Links to Exploration step
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Convolutional Neural Network for Drowsiness Detection Using EEG Signals</title><author><name sortKey="Chaabene, Siwar" sort="Chaabene, Siwar" uniqKey="Chaabene S" first="Siwar" last="Chaabene">Siwar Chaabene</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Bouaziz, Bassem" sort="Bouaziz, Bassem" uniqKey="Bouaziz B" first="Bassem" last="Bouaziz">Bassem Bouaziz</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Boudaya, Amal" sort="Boudaya, Amal" uniqKey="Boudaya A" first="Amal" last="Boudaya">Amal Boudaya</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Hokelmann, Anita" sort="Hokelmann, Anita" uniqKey="Hokelmann A" first="Anita" last="Hökelmann">Anita Hökelmann</name><affiliation><nlm:aff id="af3-sensors-21-01734">Institute of Sport Science, Otto-von-Guericke University Magdeburg, 39104 Magdeburg, Germany;<email>anita.hoekelmann@ovgu.de</email></nlm:aff></affiliation></author><author><name sortKey="Ammar, Achraf" sort="Ammar, Achraf" uniqKey="Ammar A" first="Achraf" last="Ammar">Achraf Ammar</name><affiliation><nlm:aff id="af3-sensors-21-01734">Institute of Sport Science, Otto-von-Guericke University Magdeburg, 39104 Magdeburg, Germany;<email>anita.hoekelmann@ovgu.de</email></nlm:aff></affiliation><affiliation><nlm:aff id="af4-sensors-21-01734">Interdisciplinary Laboratory in Neurosciences, Physiology and Psychology: Physical Activity, Health and Learning (LINP2), UFR STAPS, UPL, Paris Nanterre University, 92000 Nanterre, France</nlm:aff></affiliation></author><author><name sortKey="Chaari, Lotfi" sort="Chaari, Lotfi" uniqKey="Chaari L" first="Lotfi" last="Chaari">Lotfi Chaari</name><affiliation><nlm:aff id="af5-sensors-21-01734">IRIT-ENSEEIHT, University of Toulouse, 31013 Toulouse, France;<email>lotfi.chaari@toulouse-inp.fr</email></nlm:aff></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">33802357</idno><idno type="pmc">7959292</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7959292</idno><idno type="RBID">PMC:7959292</idno><idno type="doi">10.3390/s21051734</idno><date when="2021">2021</date><idno type="wicri:Area/Pmc/Corpus">000284</idno><idno type="wicri:explorRef" wicri:stream="Pmc" wicri:step="Corpus" wicri:corpus="PMC">000284</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Convolutional Neural Network for Drowsiness Detection Using EEG Signals</title><author><name sortKey="Chaabene, Siwar" sort="Chaabene, Siwar" uniqKey="Chaabene S" first="Siwar" last="Chaabene">Siwar Chaabene</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Bouaziz, Bassem" sort="Bouaziz, Bassem" uniqKey="Bouaziz B" first="Bassem" last="Bouaziz">Bassem Bouaziz</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Boudaya, Amal" sort="Boudaya, Amal" uniqKey="Boudaya A" first="Amal" last="Boudaya">Amal Boudaya</name><affiliation><nlm:aff id="af1-sensors-21-01734">Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</nlm:aff></affiliation><affiliation><nlm:aff id="af2-sensors-21-01734">Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</nlm:aff></affiliation></author><author><name sortKey="Hokelmann, Anita" sort="Hokelmann, Anita" uniqKey="Hokelmann A" first="Anita" last="Hökelmann">Anita Hökelmann</name><affiliation><nlm:aff id="af3-sensors-21-01734">Institute of Sport Science, Otto-von-Guericke University Magdeburg, 39104 Magdeburg, Germany;<email>anita.hoekelmann@ovgu.de</email></nlm:aff></affiliation></author><author><name sortKey="Ammar, Achraf" sort="Ammar, Achraf" uniqKey="Ammar A" first="Achraf" last="Ammar">Achraf Ammar</name><affiliation><nlm:aff id="af3-sensors-21-01734">Institute of Sport Science, Otto-von-Guericke University Magdeburg, 39104 Magdeburg, Germany;<email>anita.hoekelmann@ovgu.de</email></nlm:aff></affiliation><affiliation><nlm:aff id="af4-sensors-21-01734">Interdisciplinary Laboratory in Neurosciences, Physiology and Psychology: Physical Activity, Health and Learning (LINP2), UFR STAPS, UPL, Paris Nanterre University, 92000 Nanterre, France</nlm:aff></affiliation></author><author><name sortKey="Chaari, Lotfi" sort="Chaari, Lotfi" uniqKey="Chaari L" first="Lotfi" last="Chaari">Lotfi Chaari</name><affiliation><nlm:aff id="af5-sensors-21-01734">IRIT-ENSEEIHT, University of Toulouse, 31013 Toulouse, France;<email>lotfi.chaari@toulouse-inp.fr</email></nlm:aff></affiliation></author></analytic><series><title level="j">Sensors (Basel, Switzerland)</title><idno type="eISSN">1424-8220</idno><imprint><date when="2021">2021</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><p>Drowsiness detection (DD) has become a relevant area of active research in biomedical signal processing. Recently, various deep learning (DL) researches based on the EEG signals have been proposed to detect fatigue conditions. The research presented in this paper proposes an EEG classification system for DD based on DL networks. However, the proposed DD system is mainly realized into two procedures; (i) data acquisition and (ii) model analysis. For the data acquisition procedure, two key steps are considered, which are the signal collection using a wearable <italic>Emotiv EPOC<inline-formula><mml:math id="mm1"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset to record 14 channels of EEG, and the signal annotation. Furthermore, a data augmentation (DA) step has been added to the proposed system to overcome the problem of over-fitting and to improve accuracy. As regards the model analysis, a comparative study is also introduced in this paper to argue the choice of DL architecture and frameworks used in our DD system. In this sense, The proposed DD protocol makes use of a convolutional neural network (CNN) architecture implemented using the Keras library. The results showed a high accuracy value (90.42%) in drowsy/awake discrimination and revealed the efficiency of the proposed DD system compared to other research works.</p></div></front><back><div1 type="bibliography"><listBibl><biblStruct><analytic><author><name sortKey="Sahayadhas, A" uniqKey="Sahayadhas A">A. Sahayadhas</name></author><author><name sortKey="Sundaraj, K" uniqKey="Sundaraj K">K. Sundaraj</name></author><author><name sortKey="Murugappan, M" uniqKey="Murugappan M">M. Murugappan</name></author><author><name sortKey="Palaniappan, R" uniqKey="Palaniappan R">R. Palaniappan</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ghandour, A" uniqKey="Ghandour A">A. Ghandour</name></author><author><name sortKey="Hammoud, H" uniqKey="Hammoud H">H. Hammoud</name></author><author><name sortKey="Al Hajj, S" uniqKey="Al Hajj S">S. Al-Hajj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Thomas, L" uniqKey="Thomas L">L. Thomas</name></author><author><name sortKey="Gast, C" uniqKey="Gast C">C. Gast</name></author><author><name sortKey="Grube, R" uniqKey="Grube R">R. Grube</name></author><author><name sortKey="Craig, K" uniqKey="Craig K">K. Craig</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Neri, D" uniqKey="Neri D">D. Neri</name></author><author><name sortKey="Shappell, S" uniqKey="Shappell S">S. Shappell</name></author><author><name sortKey="Dejohn, C" uniqKey="Dejohn C">C. DeJohn</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hu, J" uniqKey="Hu J">J. Hu</name></author><author><name sortKey="Wang, P" uniqKey="Wang P">P. Wang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Choi, Y" uniqKey="Choi Y">Y. Choi</name></author><author><name sortKey="Kwon, N" uniqKey="Kwon N">N. Kwon</name></author><author><name sortKey="Lee, S" uniqKey="Lee S">S. Lee</name></author><author><name sortKey="Shin, Y" uniqKey="Shin Y">Y. Shin</name></author><author><name sortKey="Ryo, C" uniqKey="Ryo C">C. Ryo</name></author><author><name sortKey="Park, J" uniqKey="Park J">J. Park</name></author><author><name sortKey="Shin, D" uniqKey="Shin D">D. Shin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Murugan, S" uniqKey="Murugan S">S. Murugan</name></author><author><name sortKey="Selvaraj, J" uniqKey="Selvaraj J">J. Selvaraj</name></author><author><name sortKey="Sahayadhas, A" uniqKey="Sahayadhas A">A. Sahayadhas</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author><author><name sortKey="Golubnitschaja, O" uniqKey="Golubnitschaja O">O. Golubnitschaja</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Gwak, J" uniqKey="Gwak J">J. Gwak</name></author><author><name sortKey="Hirao, A" uniqKey="Hirao A">A. Hirao</name></author><author><name sortKey="Shino, M" uniqKey="Shino M">M. Shino</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Houssaini, A" uniqKey="Houssaini A">A. Houssaini</name></author><author><name sortKey="Sabri, A" uniqKey="Sabri A">A. Sabri</name></author><author><name sortKey="Qjidaa, H" uniqKey="Qjidaa H">H. Qjidaa</name></author><author><name sortKey="Aarab, A" uniqKey="Aarab A">A. Aarab</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Boudaya, A" uniqKey="Boudaya A">A. Boudaya</name></author><author><name sortKey="Bouaziz, B" uniqKey="Bouaziz B">B. Bouaziz</name></author><author><name sortKey="Chaabene, S" uniqKey="Chaabene S">S. Chaabene</name></author><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author><author><name sortKey="Ammar, A" uniqKey="Ammar A">A. Ammar</name></author><author><name sortKey="Hokelmann, A" uniqKey="Hokelmann A">A. Hökelmann</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Murugan, S" uniqKey="Murugan S">S. Murugan</name></author><author><name sortKey="Selvaraj, J" uniqKey="Selvaraj J">J. Selvaraj</name></author><author><name sortKey="Sahayadhas, A" uniqKey="Sahayadhas A">A. Sahayadhas</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhang, L" uniqKey="Zhang L">L. Zhang</name></author><author><name sortKey="Liu, F" uniqKey="Liu F">F. Liu</name></author><author><name sortKey="Tang, J" uniqKey="Tang J">J. Tang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Dinges, D" uniqKey="Dinges D">D. Dinges</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Stanley, P" uniqKey="Stanley P">P. Stanley</name></author><author><name sortKey="Prahash, T" uniqKey="Prahash T">T. Prahash</name></author><author><name sortKey="Lal, S" uniqKey="Lal S">S. Lal</name></author><author><name sortKey="Daniel, P" uniqKey="Daniel P">P. Daniel</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Gromer, M" uniqKey="Gromer M">M. Gromer</name></author><author><name sortKey="Salb, D" uniqKey="Salb D">D. Salb</name></author><author><name sortKey="Walzer, T" uniqKey="Walzer T">T. Walzer</name></author><author><name sortKey="Madrid, N" uniqKey="Madrid N">N. Madrid</name></author><author><name sortKey="Seepold, R" uniqKey="Seepold R">R. Seepold</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Choi, H" uniqKey="Choi H">H. Choi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ahn, S" uniqKey="Ahn S">S. Ahn</name></author><author><name sortKey="Nguyen, T" uniqKey="Nguyen T">T. Nguyen</name></author><author><name sortKey="Jang, H" uniqKey="Jang H">H. Jang</name></author><author><name sortKey="Kim, J" uniqKey="Kim J">J. Kim</name></author><author><name sortKey="Jun, S" uniqKey="Jun S">S. Jun</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, F" uniqKey="Wang F">F. Wang</name></author><author><name sortKey="Wang, H" uniqKey="Wang H">H. Wang</name></author><author><name sortKey="Fu, R" uniqKey="Fu R">R. Fu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sahayadhas, A" uniqKey="Sahayadhas A">A. Sahayadhas</name></author><author><name sortKey="Sundaraj, K" uniqKey="Sundaraj K">K. Sundaraj</name></author><author><name sortKey="Murugappan, M" uniqKey="Murugappan M">M. Murugappan</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Chen, D" uniqKey="Chen D">D. Chen</name></author><author><name sortKey="Ma, Z" uniqKey="Ma Z">Z. Ma</name></author><author><name sortKey="Li, B" uniqKey="Li B">B. Li</name></author><author><name sortKey="Yan, Z" uniqKey="Yan Z">Z. Yan</name></author><author><name sortKey="Li, W" uniqKey="Li W">W. Li</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ma, Y" uniqKey="Ma Y">Y. Ma</name></author><author><name sortKey="Chen, B" uniqKey="Chen B">B. Chen</name></author><author><name sortKey="Li, R" uniqKey="Li R">R. Li</name></author><author><name sortKey="Wang, C" uniqKey="Wang C">C. Wang</name></author><author><name sortKey="Wang, J" uniqKey="Wang J">J. Wang</name></author><author><name sortKey="She, Q" uniqKey="She Q">Q. She</name></author><author><name sortKey="Luo, Z" uniqKey="Luo Z">Z. Luo</name></author><author><name sortKey="Zhang, Y" uniqKey="Zhang Y">Y. Zhang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Papadelis, C" uniqKey="Papadelis C">C. Papadelis</name></author><author><name sortKey="Chen, Z" uniqKey="Chen Z">Z. Chen</name></author><author><name sortKey="Papadeli, C K" uniqKey="Papadeli C">C.K. Papadeli</name></author><author><name sortKey="Bamidis, P" uniqKey="Bamidis P">P. Bamidis</name></author><author><name sortKey="Chouvarda, I" uniqKey="Chouvarda I">I. Chouvarda</name></author><author><name sortKey="Bekiaris, E" uniqKey="Bekiaris E">E. Bekiaris</name></author><author><name sortKey="Maglaveras, N" uniqKey="Maglaveras N">N. Maglaveras</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Larocco, J" uniqKey="Larocco J">J. LaRocco</name></author><author><name sortKey="Le, M" uniqKey="Le M">M. Le</name></author><author><name sortKey="Paeng, D" uniqKey="Paeng D">D. Paeng</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Trutschel, U" uniqKey="Trutschel U">U. Trutschel</name></author><author><name sortKey="Sirois, B" uniqKey="Sirois B">B. Sirois</name></author><author><name sortKey="Sommer, D" uniqKey="Sommer D">D. Sommer</name></author><author><name sortKey="Golz, M" uniqKey="Golz M">M. Golz</name></author><author><name sortKey="Edwards, D" uniqKey="Edwards D">D. Edwards</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Duvinage, M" uniqKey="Duvinage M">M. Duvinage</name></author><author><name sortKey="Castermans, T" uniqKey="Castermans T">T. Castermans</name></author><author><name sortKey="Petieau, M" uniqKey="Petieau M">M. Petieau</name></author><author><name sortKey="Hoellinger, T" uniqKey="Hoellinger T">T. Hoellinger</name></author><author><name sortKey="Cheron, G" uniqKey="Cheron G">G. Cheron</name></author><author><name sortKey="Dutoit, T" uniqKey="Dutoit T">T. Dutoit</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Abichou, Y" uniqKey="Abichou Y">Y. Abichou</name></author><author><name sortKey="Chaabene, S" uniqKey="Chaabene S">S. Chaabene</name></author><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Aboalayon, K" uniqKey="Aboalayon K">K. Aboalayon</name></author><author><name sortKey="Faezipour, M" uniqKey="Faezipour M">M. Faezipour</name></author><author><name sortKey="Almuhammadi, W" uniqKey="Almuhammadi W">W. Almuhammadi</name></author><author><name sortKey="Moslehpour, S" uniqKey="Moslehpour S">S. Moslehpour</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ngxande, M" uniqKey="Ngxande M">M. Ngxande</name></author><author><name sortKey="Tapamo, J" uniqKey="Tapamo J">J. Tapamo</name></author><author><name sortKey="Burke, M" uniqKey="Burke M">M. Burke</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Patil, B" uniqKey="Patil B">B. Patil</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zorgui, S" uniqKey="Zorgui S">S. Zorgui</name></author><author><name sortKey="Chaabene, S" uniqKey="Chaabene S">S. Chaabene</name></author><author><name sortKey="Bouaziz, B" uniqKey="Bouaziz B">B. Bouaziz</name></author><author><name sortKey="Batatia, H" uniqKey="Batatia H">H. Batatia</name></author><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhu, X" uniqKey="Zhu X">X. Zhu</name></author><author><name sortKey="Zheng, W" uniqKey="Zheng W">W. Zheng</name></author><author><name sortKey="Lu, B" uniqKey="Lu B">B. Lu</name></author><author><name sortKey="Chen, X" uniqKey="Chen X">X. Chen</name></author><author><name sortKey="Chen, S" uniqKey="Chen S">S. Chen</name></author><author><name sortKey="Wang, C" uniqKey="Wang C">C. Wang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, X" uniqKey="Wang X">X. Wang</name></author><author><name sortKey="Zhao, Y" uniqKey="Zhao Y">Y. Zhao</name></author><author><name sortKey="Pourpanah, F" uniqKey="Pourpanah F">F. Pourpanah</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Gragnaniello, D" uniqKey="Gragnaniello D">D. Gragnaniello</name></author><author><name sortKey="Bottino, A" uniqKey="Bottino A">A. Bottino</name></author><author><name sortKey="Cumani, S" uniqKey="Cumani S">S. Cumani</name></author><author><name sortKey="Kim, W" uniqKey="Kim W">W. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Abiodun, O I" uniqKey="Abiodun O">O.I. Abiodun</name></author><author><name sortKey="Jantan, A" uniqKey="Jantan A">A. Jantan</name></author><author><name sortKey="Omolara, A E" uniqKey="Omolara A">A.E. Omolara</name></author><author><name sortKey="Dada, K V" uniqKey="Dada K">K.V. Dada</name></author><author><name sortKey="Mohamed, N A" uniqKey="Mohamed N">N.A. Mohamed</name></author><author><name sortKey="Arshad, H" uniqKey="Arshad H">H. Arshad</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Alom, M" uniqKey="Alom M">M. Alom</name></author><author><name sortKey="Taha, T" uniqKey="Taha T">T. Taha</name></author><author><name sortKey="Yakopcic, C" uniqKey="Yakopcic C">C. Yakopcic</name></author><author><name sortKey="Westberg, S" uniqKey="Westberg S">S. Westberg</name></author><author><name sortKey="Sidike, P" uniqKey="Sidike P">P. Sidike</name></author><author><name sortKey="Nasrin, M" uniqKey="Nasrin M">M. Nasrin</name></author><author><name sortKey="Hasan, M" uniqKey="Hasan M">M. Hasan</name></author><author><name sortKey="Essen, B V" uniqKey="Essen B">B.V. Essen</name></author><author><name sortKey="Awwal, A" uniqKey="Awwal A">A. Awwal</name></author><author><name sortKey="Asari, V" uniqKey="Asari V">V. Asari</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ed Doughmi, Y" uniqKey="Ed Doughmi Y">Y. Ed-doughmi</name></author><author><name sortKey="Idrissi, N" uniqKey="Idrissi N">N. Idrissi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Jeong, J" uniqKey="Jeong J">J. Jeong</name></author><author><name sortKey="Yu, B" uniqKey="Yu B">B. Yu</name></author><author><name sortKey="Lee, D" uniqKey="Lee D">D. Lee</name></author><author><name sortKey="Lee, S" uniqKey="Lee S">S. Lee</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Vesselenyi, T" uniqKey="Vesselenyi T">T. Vesselenyi</name></author><author><name sortKey="Moca, S" uniqKey="Moca S">S. Moca</name></author><author><name sortKey="Rus, A" uniqKey="Rus A">A. Rus</name></author><author><name sortKey="Mitran, T" uniqKey="Mitran T">T. Mitran</name></author><author><name sortKey="T Taru, B" uniqKey="T Taru B">B. Tătaru</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Guarda, L" uniqKey="Guarda L">L. Guarda</name></author><author><name sortKey="Astorga, N" uniqKey="Astorga N">N. Astorga</name></author><author><name sortKey="Droguett, E" uniqKey="Droguett E">E. Droguett</name></author><author><name sortKey="Moura, M" uniqKey="Moura M">M. Moura</name></author><author><name sortKey="Ramos, M" uniqKey="Ramos M">M. Ramos</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Deng, L" uniqKey="Deng L">L. Deng</name></author><author><name sortKey="He, X" uniqKey="He X">X. He</name></author><author><name sortKey="Gao, J" uniqKey="Gao J">J. Gao</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Alaskar, H" uniqKey="Alaskar H">H. Alaskar</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, K" uniqKey="Wang K">K. Wang</name></author><author><name sortKey="Zhao, Y" uniqKey="Zhao Y">Y. Zhao</name></author><author><name sortKey="Xiong, Q" uniqKey="Xiong Q">Q. Xiong</name></author><author><name sortKey="Fan, M" uniqKey="Fan M">M. Fan</name></author><author><name sortKey="Sun, G" uniqKey="Sun G">G. Sun</name></author><author><name sortKey="Ma, L" uniqKey="Ma L">L. Ma</name></author><author><name sortKey="Liu, T" uniqKey="Liu T">T. Liu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Piekarski, M" uniqKey="Piekarski M">M. Piekarski</name></author><author><name sortKey="Korjakowska, J" uniqKey="Korjakowska J">J. Korjakowska</name></author><author><name sortKey="Wawrzyniak, A" uniqKey="Wawrzyniak A">A. Wawrzyniak</name></author><author><name sortKey="Gorgon, M" uniqKey="Gorgon M">M. Gorgon</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Chakraborty, S" uniqKey="Chakraborty S">S. Chakraborty</name></author><author><name sortKey="Aich, S" uniqKey="Aich S">S. Aich</name></author><author><name sortKey="Joo, M" uniqKey="Joo M">M. Joo</name></author><author><name sortKey="Sain, M" uniqKey="Sain M">M. Sain</name></author><author><name sortKey="Kim, H" uniqKey="Kim H">H. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Roy, Y" uniqKey="Roy Y">Y. Roy</name></author><author><name sortKey="Banville, H" uniqKey="Banville H">H. Banville</name></author><author><name sortKey="Albuquerque, I" uniqKey="Albuquerque I">I. Albuquerque</name></author><author><name sortKey="Gramfort, A" uniqKey="Gramfort A">A. Gramfort</name></author><author><name sortKey="Falk, T" uniqKey="Falk T">T. Falk</name></author><author><name sortKey="Faubert, J" uniqKey="Faubert J">J. Faubert</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Salamon, J" uniqKey="Salamon J">J. Salamon</name></author><author><name sortKey="Bello, J" uniqKey="Bello J">J. Bello</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Dwivedi, K" uniqKey="Dwivedi K">K. Dwivedi</name></author><author><name sortKey="Biswaranjan, K" uniqKey="Biswaranjan K">K. Biswaranjan</name></author><author><name sortKey="Sethi, A" uniqKey="Sethi A">A. Sethi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Doughmi, Y" uniqKey="Doughmi Y">Y. Doughmi</name></author><author><name sortKey="Idrissi, N" uniqKey="Idrissi N">N. Idrissi</name></author><author><name sortKey="Hbali, Y" uniqKey="Hbali Y">Y. Hbali</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Yang, Y" uniqKey="Yang Y">Y. Yang</name></author><author><name sortKey="Gao, Z" uniqKey="Gao Z">Z. Gao</name></author><author><name sortKey="Li, Y" uniqKey="Li Y">Y. Li</name></author><author><name sortKey="Cai, Q" uniqKey="Cai Q">Q. Cai</name></author><author><name sortKey="Marwan, N" uniqKey="Marwan N">N. Marwan</name></author><author><name sortKey="Kurths, J" uniqKey="Kurths J">J. Kurths</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Shalash, W" uniqKey="Shalash W">W. Shalash</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zeng, H" uniqKey="Zeng H">H. Zeng</name></author><author><name sortKey="Yang, C" uniqKey="Yang C">C. Yang</name></author><author><name sortKey="Dai, G" uniqKey="Dai G">G. Dai</name></author><author><name sortKey="Qin, F" uniqKey="Qin F">F. Qin</name></author><author><name sortKey="Zhang, J" uniqKey="Zhang J">J. Zhang</name></author><author><name sortKey="Kong, W" uniqKey="Kong W">W. Kong</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ko, W" uniqKey="Ko W">W. Ko</name></author><author><name sortKey="Oh, K" uniqKey="Oh K">K. Oh</name></author><author><name sortKey="Jeon, E" uniqKey="Jeon E">E. Jeon</name></author><author><name sortKey="Suk, H" uniqKey="Suk H">H. Suk</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cheng, E" uniqKey="Cheng E">E. Cheng</name></author><author><name sortKey="Young, K" uniqKey="Young K">K. Young</name></author><author><name sortKey="Lin, C" uniqKey="Lin C">C. Lin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Rahman, K" uniqKey="Rahman K">K. Rahman</name></author><author><name sortKey="Mustaffa, M" uniqKey="Mustaffa M">M. Mustaffa</name></author><author><name sortKey="Fuad, N" uniqKey="Fuad N">N. Fuad</name></author><author><name sortKey="Ahmad, M" uniqKey="Ahmad M">M. Ahmad</name></author><author><name sortKey="Ahad, R" uniqKey="Ahad R">R. Ahad</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sanchez Reolid, R" uniqKey="Sanchez Reolid R">R. Sánchez-Reolid</name></author><author><name sortKey="Garcia, A" uniqKey="Garcia A">A. García</name></author><author><name sortKey="Vicente Querol, M" uniqKey="Vicente Querol M">M. Vicente-Querol</name></author><author><name sortKey="Fernandez Aguilar, L" uniqKey="Fernandez Aguilar L">L. Fernández-Aguilar</name></author><author><name sortKey="L Pez, M" uniqKey="L Pez M">M. López</name></author><author><name sortKey="Fernandez Caballero, A" uniqKey="Fernandez Caballero A">A. Fernández-Caballero</name></author><author><name sortKey="Gonzalez, P" uniqKey="Gonzalez P">P. González</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pedrosa, P" uniqKey="Pedrosa P">P. Pedrosa</name></author><author><name sortKey="Fiedler, P" uniqKey="Fiedler P">P. Fiedler</name></author><author><name sortKey="Schinaia, L" uniqKey="Schinaia L">L. Schinaia</name></author><author><name sortKey="Vasconcelos, B" uniqKey="Vasconcelos B">B. Vasconcelos</name></author><author><name sortKey="Martins, A" uniqKey="Martins A">A. Martins</name></author><author><name sortKey="Amaral, M" uniqKey="Amaral M">M. Amaral</name></author><author><name sortKey="Comani, S" uniqKey="Comani S">S. Comani</name></author><author><name sortKey="Haueisen, J" uniqKey="Haueisen J">J. Haueisen</name></author><author><name sortKey="Fonseca, C" uniqKey="Fonseca C">C. Fonseca</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Towle, V" uniqKey="Towle V">V. Towle</name></author><author><name sortKey="Bolafios, J" uniqKey="Bolafios J">J. Bolafios</name></author><author><name sortKey="Suarez, D" uniqKey="Suarez D">D. Suarez</name></author><author><name sortKey="Tan, K" uniqKey="Tan K">K. Tan</name></author><author><name sortKey="Grzeszczuk, R" uniqKey="Grzeszczuk R">R. Grzeszczuk</name></author><author><name sortKey="Levin, D" uniqKey="Levin D">D. Levin</name></author><author><name sortKey="Cakmur, R" uniqKey="Cakmur R">R. Cakmur</name></author><author><name sortKey="Frank, S" uniqKey="Frank S">S. Frank</name></author><author><name sortKey="Spire, J" uniqKey="Spire J">J. Spire</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hu, S" uniqKey="Hu S">S. Hu</name></author><author><name sortKey="Zheng, G" uniqKey="Zheng G">G. Zheng</name></author><author><name sortKey="Peters, B" uniqKey="Peters B">B. Peters</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Mohammedi, M" uniqKey="Mohammedi M">M. Mohammedi</name></author><author><name sortKey="Omar, M" uniqKey="Omar M">M. Omar</name></author><author><name sortKey="Bouabdallah, A" uniqKey="Bouabdallah A">A. Bouabdallah</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Gebodh, N" uniqKey="Gebodh N">N. Gebodh</name></author><author><name sortKey="Esmaeilpour, Z" uniqKey="Esmaeilpour Z">Z. Esmaeilpour</name></author><author><name sortKey="Adair, D" uniqKey="Adair D">D. Adair</name></author><author><name sortKey="Chelette, K" uniqKey="Chelette K">K. Chelette</name></author><author><name sortKey="Dmochowski, J" uniqKey="Dmochowski J">J. Dmochowski</name></author><author><name sortKey="Woods, A" uniqKey="Woods A">A. Woods</name></author><author><name sortKey="Kappenman, E" uniqKey="Kappenman E">E. Kappenman</name></author><author><name sortKey="Parra, L" uniqKey="Parra L">L. Parra</name></author><author><name sortKey="Bikson, M" uniqKey="Bikson M">M. Bikson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Laruelo, A" uniqKey="Laruelo A">A. Laruelo</name></author><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author><author><name sortKey="Batatia, H" uniqKey="Batatia H">H. Batatia</name></author><author><name sortKey="Ken, S" uniqKey="Ken S">S. Ken</name></author><author><name sortKey="Rowland, B" uniqKey="Rowland B">B. Rowland</name></author><author><name sortKey="Tourneret, J Y" uniqKey="Tourneret J">J.Y. Tourneret</name></author><author><name sortKey="Laprie, A" uniqKey="Laprie A">A. Laprie</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Chaari, L" uniqKey="Chaari L">L. Chaari</name></author><author><name sortKey="Tourneret, J Y" uniqKey="Tourneret J">J.Y. Tourneret</name></author><author><name sortKey="Chaux, C" uniqKey="Chaux C">C. Chaux</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sevgi, L" uniqKey="Sevgi L">L. Sevgi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ludwig, P" uniqKey="Ludwig P">P. Ludwig</name></author><author><name sortKey="Varacallo, M" uniqKey="Varacallo M">M. Varacallo</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Teplan, M" uniqKey="Teplan M">M. Teplan</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Al Kadi, M I" uniqKey="Al Kadi M">M.I. Al-Kadi</name></author><author><name sortKey="Reaz, M B I" uniqKey="Reaz M">M.B.I. Reaz</name></author><author><name sortKey="Ali, M A" uniqKey="Ali M">M.A. Ali</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Schutze, M D" uniqKey="Schutze M">M.D. Schütze</name></author><author><name sortKey="Junghanns, K" uniqKey="Junghanns K">K. Junghanns</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Amo, C" uniqKey="Amo C">C. Amo</name></author><author><name sortKey="De Santiago, L" uniqKey="De Santiago L">L. de Santiago</name></author><author><name sortKey="Barea, R" uniqKey="Barea R">R. Barea</name></author><author><name sortKey="L Pez Dorado, A" uniqKey="L Pez Dorado A">A. López-Dorado</name></author><author><name sortKey="Boquete, L" uniqKey="Boquete L">L. Boquete</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Dkhil, M B" uniqKey="Dkhil M">M.B. Dkhil</name></author><author><name sortKey="Wali, A" uniqKey="Wali A">A. Wali</name></author><author><name sortKey="Alimi, A M" uniqKey="Alimi A">A.M. Alimi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ogino, M" uniqKey="Ogino M">M. Ogino</name></author><author><name sortKey="Mitsukura, Y" uniqKey="Mitsukura Y">Y. Mitsukura</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lin, C T" uniqKey="Lin C">C.T. Lin</name></author><author><name sortKey="Wu, R C" uniqKey="Wu R">R.C. Wu</name></author><author><name sortKey="Liang, S" uniqKey="Liang S">S. Liang</name></author><author><name sortKey="Chao, W H" uniqKey="Chao W">W.H. Chao</name></author><author><name sortKey="Chen, Y J" uniqKey="Chen Y">Y.J. Chen</name></author><author><name sortKey="Jung, T P" uniqKey="Jung T">T.P. Jung</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Makeig, S" uniqKey="Makeig S">S. Makeig</name></author><author><name sortKey="Jung, T" uniqKey="Jung T">T. Jung</name></author><author><name sortKey="Sejnowski, T" uniqKey="Sejnowski T">T. Sejnowski</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Subasi, A" uniqKey="Subasi A">A. Subasi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kar, S" uniqKey="Kar S">S. Kar</name></author><author><name sortKey="Bhagat, M" uniqKey="Bhagat M">M. Bhagat</name></author><author><name sortKey="Routray, A" uniqKey="Routray A">A. Routray</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bernardi, G" uniqKey="Bernardi G">G. Bernardi</name></author><author><name sortKey="Betta, M" uniqKey="Betta M">M. Betta</name></author><author><name sortKey="Ricciardi, E" uniqKey="Ricciardi E">E. Ricciardi</name></author><author><name sortKey="Pietrini, P" uniqKey="Pietrini P">P. Pietrini</name></author><author><name sortKey="Tononi, G" uniqKey="Tononi G">G. Tononi</name></author><author><name sortKey="Siclari, F" uniqKey="Siclari F">F. Siclari</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lashgari, E" uniqKey="Lashgari E">E. Lashgari</name></author><author><name sortKey="Liang, D" uniqKey="Liang D">D. Liang</name></author><author><name sortKey="Maoz, U" uniqKey="Maoz U">U. Maoz</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhang, Z" uniqKey="Zhang Z">Z. Zhang</name></author><author><name sortKey="Casals, J" uniqKey="Casals J">J. Casals</name></author><author><name sortKey="Cichocki, A" uniqKey="Cichocki A">A. Cichocki</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhang, Y" uniqKey="Zhang Y">Y. Zhang</name></author><author><name sortKey="Yang, S" uniqKey="Yang S">S. Yang</name></author><author><name sortKey="Liu, Y" uniqKey="Liu Y">Y. Liu</name></author><author><name sortKey="Zhang, Y" uniqKey="Zhang Y">Y. Zhang</name></author><author><name sortKey="Han, B" uniqKey="Han B">B. Han</name></author><author><name sortKey="Zhou, F" uniqKey="Zhou F">F. Zhou</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Chawla, N V" uniqKey="Chawla N">N.V. Chawla</name></author><author><name sortKey="Bowyer, K W" uniqKey="Bowyer K">K.W. Bowyer</name></author><author><name sortKey="Hall, L O" uniqKey="Hall L">L.O. Hall</name></author><author><name sortKey="Kegelmeyer, W P" uniqKey="Kegelmeyer W">W.P. Kegelmeyer</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Garcia, A" uniqKey="Garcia A">A. Garcia</name></author><author><name sortKey="Peter, K" uniqKey="Peter K">K. Peter</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Srivastava, N" uniqKey="Srivastava N">N. Srivastava</name></author><author><name sortKey="Hinton, G" uniqKey="Hinton G">G. Hinton</name></author><author><name sortKey="Krizhevsky, A" uniqKey="Krizhevsky A">A. Krizhevsky</name></author><author><name sortKey="Sutskever, I" uniqKey="Sutskever I">I. Sutskever</name></author><author><name sortKey="Salakhutdinov, R" uniqKey="Salakhutdinov R">R. Salakhutdinov</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hammad, M" uniqKey="Hammad M">M. Hammad</name></author><author><name sortKey="Plawiak, P" uniqKey="Plawiak P">P. Pławiak</name></author><author><name sortKey="Wang, K" uniqKey="Wang K">K. Wang</name></author><author><name sortKey="Acharya, U R" uniqKey="Acharya U">U.R. Acharya</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Oh, S" uniqKey="Oh S">S. Oh</name></author><author><name sortKey="Jahmunah, V" uniqKey="Jahmunah V">V. Jahmunah</name></author><author><name sortKey="Ooi, C" uniqKey="Ooi C">C. Ooi</name></author><author><name sortKey="Tan, R" uniqKey="Tan R">R. Tan</name></author><author><name sortKey="Ciaccio, E" uniqKey="Ciaccio E">E. Ciaccio</name></author><author><name sortKey="Yamakawa, T" uniqKey="Yamakawa T">T. Yamakawa</name></author><author><name sortKey="Tanabe, M" uniqKey="Tanabe M">M. Tanabe</name></author><author><name sortKey="Kobayashi, M" uniqKey="Kobayashi M">M. Kobayashi</name></author><author><name sortKey="Acharya, U" uniqKey="Acharya U">U. Acharya</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kim, J" uniqKey="Kim J">J. Kim</name></author><author><name sortKey="Seo, S" uniqKey="Seo S">S. Seo</name></author><author><name sortKey="Song, C" uniqKey="Song C">C. Song</name></author><author><name sortKey="Kim, K" uniqKey="Kim K">K. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Park, J" uniqKey="Park J">J. Park</name></author><author><name sortKey="Kim, J" uniqKey="Kim J">J. Kim</name></author><author><name sortKey="Jung, S" uniqKey="Jung S">S. Jung</name></author><author><name sortKey="Gil, Y" uniqKey="Gil Y">Y. Gil</name></author><author><name sortKey="Choi, J" uniqKey="Choi J">J. Choi</name></author><author><name sortKey="Son, H" uniqKey="Son H">H. Son</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, S" uniqKey="Wang S">S. Wang</name></author><author><name sortKey="Wang, S" uniqKey="Wang S">S. Wang</name></author><author><name sortKey="Wang, Y" uniqKey="Wang Y">Y. Wang</name></author><author><name sortKey="Wang, S" uniqKey="Wang S">S. Wang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Uyulan, C" uniqKey="Uyulan C">C. Uyulan</name></author><author><name sortKey="Erguzel, T" uniqKey="Erguzel T">T. Ergüzel</name></author><author><name sortKey="Unubol, H" uniqKey="Unubol H">H. Unubol</name></author><author><name sortKey="Cebi, M" uniqKey="Cebi M">M. Cebi</name></author><author><name sortKey="Sayar, G" uniqKey="Sayar G">G. Sayar</name></author><author><name sortKey="Asad, M" uniqKey="Asad M">M. Asad</name></author><author><name sortKey="Tarhan, N" uniqKey="Tarhan N">N. Tarhan</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hasan, M" uniqKey="Hasan M">M. Hasan</name></author><author><name sortKey="Shon, D" uniqKey="Shon D">D. Shon</name></author><author><name sortKey="Im, K" uniqKey="Im K">K. Im</name></author><author><name sortKey="Choi, H" uniqKey="Choi H">H. Choi</name></author><author><name sortKey="Yoo, D" uniqKey="Yoo D">D. Yoo</name></author><author><name sortKey="Kim, J" uniqKey="Kim J">J. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Nahid, N" uniqKey="Nahid N">N. Nahid</name></author><author><name sortKey="Rahman, A" uniqKey="Rahman A">A. Rahman</name></author><author><name sortKey="Ahad, M A R" uniqKey="Ahad M">M.A.R. Ahad</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wulan, N" uniqKey="Wulan N">N. Wulan</name></author><author><name sortKey="Wang, W" uniqKey="Wang W">W. Wang</name></author><author><name sortKey="Sun, P" uniqKey="Sun P">P. Sun</name></author><author><name sortKey="Wang, K" uniqKey="Wang K">K. Wang</name></author><author><name sortKey="Xia, Y" uniqKey="Xia Y">Y. Xia</name></author><author><name sortKey="Zhang, H" uniqKey="Zhang H">H. Zhang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhu, T" uniqKey="Zhu T">T. Zhu</name></author><author><name sortKey="Luo, W" uniqKey="Luo W">W. Luo</name></author><author><name sortKey="Yu, F" uniqKey="Yu F">F. Yu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cao, C" uniqKey="Cao C">C. Cao</name></author><author><name sortKey="Liu, F" uniqKey="Liu F">F. Liu</name></author><author><name sortKey="Tan, H" uniqKey="Tan H">H. Tan</name></author><author><name sortKey="Song, D" uniqKey="Song D">D. Song</name></author><author><name sortKey="Shu, W" uniqKey="Shu W">W. Shu</name></author><author><name sortKey="Li, W" uniqKey="Li W">W. Li</name></author><author><name sortKey="Zhou, Y" uniqKey="Zhou Y">Y. Zhou</name></author><author><name sortKey="Bo, X" uniqKey="Bo X">X. Bo</name></author><author><name sortKey="Xie, Z" uniqKey="Xie Z">Z. Xie</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Rehman, M" uniqKey="Rehman M">M. Rehman</name></author><author><name sortKey="Waris, A" uniqKey="Waris A">A. Waris</name></author><author><name sortKey="Gilani, S" uniqKey="Gilani S">S. Gilani</name></author><author><name sortKey="Jochumsen, M" uniqKey="Jochumsen M">M. Jochumsen</name></author><author><name sortKey="Niazi, I K" uniqKey="Niazi I">I.K. Niazi</name></author><author><name sortKey="Jamil, M" uniqKey="Jamil M">M. Jamil</name></author><author><name sortKey="Farina, D" uniqKey="Farina D">D. Farina</name></author><author><name sortKey="Kamavuako, E" uniqKey="Kamavuako E">E. Kamavuako</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hu, J" uniqKey="Hu J">J. Hu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Morales, J" uniqKey="Morales J">J. Morales</name></author><author><name sortKey="Salda A, R" uniqKey="Salda A R">R. Saldaña</name></author><author><name sortKey="Castolo, M" uniqKey="Castolo M">M. Castolo</name></author><author><name sortKey="Borrayo, C C E" uniqKey="Borrayo C">C.C.E. Borrayo</name></author><author><name sortKey="Ruiz, A" uniqKey="Ruiz A">A. Ruiz</name></author><author><name sortKey="Perez, H" uniqKey="Perez H">H. Perez</name></author><author><name sortKey="Ruiz, G" uniqKey="Ruiz G">G. Ruiz</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Srinivasan, V" uniqKey="Srinivasan V">V. Srinivasan</name></author><author><name sortKey="Islam, M" uniqKey="Islam M">M. Islam</name></author><author><name sortKey="Zhang, W" uniqKey="Zhang W">W. Zhang</name></author><author><name sortKey="Ren, H" uniqKey="Ren H">H. Ren</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Choi, I" uniqKey="Choi I">I. Choi</name></author><author><name sortKey="Kim, H" uniqKey="Kim H">H. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Buslaev, A" uniqKey="Buslaev A">A. Buslaev</name></author><author><name sortKey="Iglovikov, V" uniqKey="Iglovikov V">V. Iglovikov</name></author><author><name sortKey="Khvedchenya, E" uniqKey="Khvedchenya E">E. Khvedchenya</name></author><author><name sortKey="Parinov, A" uniqKey="Parinov A">A. Parinov</name></author><author><name sortKey="Druzhinin, M" uniqKey="Druzhinin M">M. Druzhinin</name></author><author><name sortKey="Kalinin, A" uniqKey="Kalinin A">A. Kalinin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Somrak, M" uniqKey="Somrak M">M. Somrak</name></author><author><name sortKey="Dzeroski, S" uniqKey="Dzeroski S">S. Džeroski</name></author><author><name sortKey="Kokalj, T" uniqKey="Kokalj T">T. Kokalj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Abadi, M" uniqKey="Abadi M">M. Abadi</name></author><author><name sortKey="Barham, P" uniqKey="Barham P">P. Barham</name></author><author><name sortKey="Chen, J" uniqKey="Chen J">J. Chen</name></author><author><name sortKey="Chen, Z" uniqKey="Chen Z">Z. Chen</name></author><author><name sortKey="Davis, A" uniqKey="Davis A">A. Davis</name></author><author><name sortKey="Dean, J" uniqKey="Dean J">J. Dean</name></author><author><name sortKey="Devin, M" uniqKey="Devin M">M. Devin</name></author><author><name sortKey="Ghemawat, S" uniqKey="Ghemawat S">S. Ghemawat</name></author><author><name sortKey="Irving, G" uniqKey="Irving G">G. Irving</name></author><author><name sortKey="Isard, M" uniqKey="Isard M">M. Isard</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Paszke, A" uniqKey="Paszke A">A. Paszke</name></author><author><name sortKey="Gross, S" uniqKey="Gross S">S. Gross</name></author><author><name sortKey="Massa, F" uniqKey="Massa F">F. Massa</name></author><author><name sortKey="Lerer, A" uniqKey="Lerer A">A. Lerer</name></author><author><name sortKey="Bradbury, J" uniqKey="Bradbury J">J. Bradbury</name></author><author><name sortKey="Chanan, G" uniqKey="Chanan G">G. Chanan</name></author><author><name sortKey="Killeen, T" uniqKey="Killeen T">T. Killeen</name></author><author><name sortKey="Lin, Z" uniqKey="Lin Z">Z. Lin</name></author><author><name sortKey="Gimelshein, N" uniqKey="Gimelshein N">N. Gimelshein</name></author><author><name sortKey="Antiga, L" uniqKey="Antiga L">L. Antiga</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Predescu, A" uniqKey="Predescu A">A. Predescu</name></author><author><name sortKey="Truica, C" uniqKey="Truica C">C. Truica</name></author><author><name sortKey="Apostol, E" uniqKey="Apostol E">E. Apostol</name></author><author><name sortKey="Mocanu, M" uniqKey="Mocanu M">M. Mocanu</name></author><author><name sortKey="Lupu, C" uniqKey="Lupu C">C. Lupu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Saleem, M H" uniqKey="Saleem M">M.H. Saleem</name></author><author><name sortKey="Potgieter, J" uniqKey="Potgieter J">J. Potgieter</name></author><author><name sortKey="Arif, K" uniqKey="Arif K">K. Arif</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Li, G" uniqKey="Li G">G. Li</name></author><author><name sortKey="Lee, C" uniqKey="Lee C">C. Lee</name></author><author><name sortKey="Jung, J" uniqKey="Jung J">J. Jung</name></author><author><name sortKey="Youn, Y" uniqKey="Youn Y">Y. Youn</name></author><author><name sortKey="Camacho, D" uniqKey="Camacho D">D. Camacho</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Shaf, A" uniqKey="Shaf A">A. Shaf</name></author><author><name sortKey="Ali, T" uniqKey="Ali T">T. Ali</name></author><author><name sortKey="Farooq, W" uniqKey="Farooq W">W. Farooq</name></author><author><name sortKey="Javaid, S" uniqKey="Javaid S">S. Javaid</name></author><author><name sortKey="Draz, U" uniqKey="Draz U">U. Draz</name></author><author><name sortKey="Yasin, S" uniqKey="Yasin S">S. Yasin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Tafsast, A" uniqKey="Tafsast A">A. Tafsast</name></author><author><name sortKey="Ferroudji, K" uniqKey="Ferroudji K">K. Ferroudji</name></author><author><name sortKey="Hadjili, M" uniqKey="Hadjili M">M. Hadjili</name></author><author><name sortKey="Bouakaz, A" uniqKey="Bouakaz A">A. Bouakaz</name></author><author><name sortKey="Benoudjit, N" uniqKey="Benoudjit N">N. Benoudjit</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Jogin, M" uniqKey="Jogin M">M. Jogin</name></author><author><name sortKey="Madhulika, M S" uniqKey="Madhulika M">M.S. Madhulika</name></author><author><name sortKey="Divya, G D" uniqKey="Divya G">G.D. Divya</name></author><author><name sortKey="Meghana, R K" uniqKey="Meghana R">R.K. Meghana</name></author><author><name sortKey="Apoorva, S" uniqKey="Apoorva S">S. Apoorva</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Garbin, C" uniqKey="Garbin C">C. Garbin</name></author><author><name sortKey="Zhu, X" uniqKey="Zhu X">X. Zhu</name></author><author><name sortKey="Marques, O" uniqKey="Marques O">O. Marques</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bisong, E" uniqKey="Bisong E">E. Bisong</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sarno, R" uniqKey="Sarno R">R. Sarno</name></author><author><name sortKey="Nugraha, B" uniqKey="Nugraha B">B. Nugraha</name></author><author><name sortKey="Munawar, M" uniqKey="Munawar M">M. Munawar</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Blaiech, H" uniqKey="Blaiech H">H. Blaiech</name></author><author><name sortKey="Neji, M" uniqKey="Neji M">M. Neji</name></author><author><name sortKey="Wali, A" uniqKey="Wali A">A. Wali</name></author><author><name sortKey="Alimi, A" uniqKey="Alimi A">A. Alimi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Nugraha, B" uniqKey="Nugraha B">B. Nugraha</name></author><author><name sortKey="Sarno, R" uniqKey="Sarno R">R. Sarno</name></author><author><name sortKey="Asfani, D" uniqKey="Asfani D">D. Asfani</name></author><author><name sortKey="Igasaki, T" uniqKey="Igasaki T">T. Igasaki</name></author><author><name sortKey="Munawar, M" uniqKey="Munawar M">M. Munawar</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Huang, R S" uniqKey="Huang R">R.S. Huang</name></author><author><name sortKey="Jung, T P" uniqKey="Jung T">T.P. Jung</name></author><author><name sortKey="Makeig, S" uniqKey="Makeig S">S. Makeig</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Majumder, S" uniqKey="Majumder S">S. Majumder</name></author><author><name sortKey="Guragain, B" uniqKey="Guragain B">B. Guragain</name></author><author><name sortKey="Wang, C" uniqKey="Wang C">C. Wang</name></author><author><name sortKey="Wilson, N" uniqKey="Wilson N">N. Wilson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zhang, H" uniqKey="Zhang H">H. Zhang</name></author><author><name sortKey="Silva, F" uniqKey="Silva F">F. Silva</name></author><author><name sortKey="Ohata, E" uniqKey="Ohata E">E. Ohata</name></author><author><name sortKey="Medeiros, A" uniqKey="Medeiros A">A. Medeiros</name></author><author><name sortKey="Filho, P" uniqKey="Filho P">P. Filho</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Trejo, L" uniqKey="Trejo L">L. Trejo</name></author><author><name sortKey="Kubitz, K" uniqKey="Kubitz K">K. Kubitz</name></author><author><name sortKey="Rosipal, R" uniqKey="Rosipal R">R. Rosipal</name></author><author><name sortKey="Kochavi, R" uniqKey="Kochavi R">R. Kochavi</name></author><author><name sortKey="Montgomery, L" uniqKey="Montgomery L">L. Montgomery</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wang, R" uniqKey="Wang R">R. Wang</name></author><author><name sortKey="Wang, Y" uniqKey="Wang Y">Y. Wang</name></author><author><name sortKey="Luo, C" uniqKey="Luo C">C. Luo</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Awais, M" uniqKey="Awais M">M. Awais</name></author><author><name sortKey="Badruddin, N" uniqKey="Badruddin N">N. Badruddin</name></author><author><name sortKey="Drieberg, M" uniqKey="Drieberg M">M. Drieberg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Nguyen, T" uniqKey="Nguyen T">T. Nguyen</name></author><author><name sortKey="Ahn, S" uniqKey="Ahn S">S. Ahn</name></author><author><name sortKey="Jang, H" uniqKey="Jang H">H. Jang</name></author><author><name sortKey="Jun, S C" uniqKey="Jun S">S.C. Jun</name></author><author><name sortKey="Kim, J G" uniqKey="Kim J">J.G. Kim</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Noori, S" uniqKey="Noori S">S. Noori</name></author><author><name sortKey="Mikaeili, M" uniqKey="Mikaeili M">M. Mikaeili</name></author></analytic></biblStruct></listBibl></div1></back></TEI><pmc article-type="research-article"><pmc-dir>properties open_access</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-ta">Sensors (Basel)</journal-id><journal-id journal-id-type="iso-abbrev">Sensors (Basel)</journal-id><journal-id journal-id-type="publisher-id">sensors</journal-id><journal-title-group><journal-title>Sensors (Basel, Switzerland)</journal-title></journal-title-group><issn pub-type="epub">1424-8220</issn><publisher><publisher-name>MDPI</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="pmid">33802357</article-id><article-id pub-id-type="pmc">7959292</article-id><article-id pub-id-type="doi">10.3390/s21051734</article-id><article-id pub-id-type="publisher-id">sensors-21-01734</article-id><article-categories><subj-group subj-group-type="heading"><subject>Article</subject></subj-group></article-categories><title-group><article-title>Convolutional Neural Network for Drowsiness Detection Using EEG Signals</article-title></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="true">https://orcid.org/0000-0002-6458-9569</contrib-id><name><surname>Chaabene</surname><given-names>Siwar</given-names></name><xref ref-type="aff" rid="af1-sensors-21-01734">1</xref><xref ref-type="aff" rid="af2-sensors-21-01734">2</xref><xref ref-type="author-notes" rid="fn1-sensors-21-01734">†</xref></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="true">https://orcid.org/0000-0002-3692-9482</contrib-id><name><surname>Bouaziz</surname><given-names>Bassem</given-names></name><xref ref-type="aff" rid="af1-sensors-21-01734">1</xref><xref ref-type="aff" rid="af2-sensors-21-01734">2</xref><xref ref-type="author-notes" rid="fn1-sensors-21-01734">†</xref></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="true">https://orcid.org/0000-0003-0486-7923</contrib-id><name><surname>Boudaya</surname><given-names>Amal</given-names></name><xref ref-type="aff" rid="af1-sensors-21-01734">1</xref><xref ref-type="aff" rid="af2-sensors-21-01734">2</xref></contrib><contrib contrib-type="author"><name><surname>Hökelmann</surname><given-names>Anita</given-names></name><xref ref-type="aff" rid="af3-sensors-21-01734">3</xref></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="true">https://orcid.org/0000-0003-0347-8053</contrib-id><name><surname>Ammar</surname><given-names>Achraf</given-names></name><xref ref-type="aff" rid="af3-sensors-21-01734">3</xref><xref ref-type="aff" rid="af4-sensors-21-01734">4</xref><xref rid="c1-sensors-21-01734" ref-type="corresp">*</xref><xref ref-type="author-notes" rid="fn2-sensors-21-01734">‡</xref></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="true">https://orcid.org/0000-0002-3590-0370</contrib-id><name><surname>Chaari</surname><given-names>Lotfi</given-names></name><xref ref-type="aff" rid="af5-sensors-21-01734">5</xref><xref ref-type="author-notes" rid="fn2-sensors-21-01734">‡</xref></contrib></contrib-group><contrib-group><contrib contrib-type="editor"><name><surname>Sparacino</surname><given-names>Giovanni</given-names></name><role>Academic Editor</role></contrib></contrib-group><aff id="af1-sensors-21-01734"><label>1</label>Multimedia InfoRmation Systems and Advanced Computing Laboratory (MIRACL), University of Sfax, Sfax 3021, Tunisia;<email>siwarchaabene@gmail.com</email> (S.C.);<email>Bassem.Bouaziz@isims.usf.tn</email> (B.B.);<email>amalboudaya71@gmail.com</email> (A.B.)</aff><aff id="af2-sensors-21-01734"><label>2</label>Digital Research Center of Sfax, B.P. 275, Sakiet Ezzit, Sfax 3021, Tunisia</aff><aff id="af3-sensors-21-01734"><label>3</label>Institute of Sport Science, Otto-von-Guericke University Magdeburg, 39104 Magdeburg, Germany;<email>anita.hoekelmann@ovgu.de</email></aff><aff id="af4-sensors-21-01734"><label>4</label>Interdisciplinary Laboratory in Neurosciences, Physiology and Psychology: Physical Activity, Health and Learning (LINP2), UFR STAPS, UPL, Paris Nanterre University, 92000 Nanterre, France</aff><aff id="af5-sensors-21-01734"><label>5</label>IRIT-ENSEEIHT, University of Toulouse, 31013 Toulouse, France;<email>lotfi.chaari@toulouse-inp.fr</email></aff><author-notes><corresp id="c1-sensors-21-01734"><label>*</label>Correspondence: <email>achraf1.ammar@ovgu.de</email></corresp><fn id="fn1-sensors-21-01734"><label>†</label><p>These authors contributed equally to this work as first author.</p></fn><fn id="fn2-sensors-21-01734"><label>‡</label><p>These authors contributed equally to this work as last author.</p></fn></author-notes><pub-date pub-type="epub"><day>03</day><month>3</month><year>2021</year></pub-date><pub-date pub-type="collection"><month>3</month><year>2021</year></pub-date><volume>21</volume><issue>5</issue><elocation-id>1734</elocation-id><history><date date-type="received"><day>05</day><month>12</month><year>2020</year></date><date date-type="accepted"><day>24</day><month>2</month><year>2021</year></date></history><permissions><copyright-statement>© 2021 by the authors.</copyright-statement><copyright-year>2021</copyright-year><license license-type="open-access"><license-p>Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (<ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/">http://creativecommons.org/licenses/by/4.0/</ext-link>).</license-p></license></permissions><abstract><p>Drowsiness detection (DD) has become a relevant area of active research in biomedical signal processing. Recently, various deep learning (DL) researches based on the EEG signals have been proposed to detect fatigue conditions. The research presented in this paper proposes an EEG classification system for DD based on DL networks. However, the proposed DD system is mainly realized into two procedures; (i) data acquisition and (ii) model analysis. For the data acquisition procedure, two key steps are considered, which are the signal collection using a wearable <italic>Emotiv EPOC<inline-formula><mml:math id="mm1"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset to record 14 channels of EEG, and the signal annotation. Furthermore, a data augmentation (DA) step has been added to the proposed system to overcome the problem of over-fitting and to improve accuracy. As regards the model analysis, a comparative study is also introduced in this paper to argue the choice of DL architecture and frameworks used in our DD system. In this sense, The proposed DD protocol makes use of a convolutional neural network (CNN) architecture implemented using the Keras library. The results showed a high accuracy value (90.42%) in drowsy/awake discrimination and revealed the efficiency of the proposed DD system compared to other research works.</p></abstract><kwd-group><kwd>drowsiness detection</kwd><kwd>EEG signals</kwd><kwd><italic>Emotiv EPOC</italic><sup>+</sup></kwd><kwd>deep learning</kwd><kwd>data augmentation</kwd><kwd>convolutional neural networks</kwd><kwd>classification</kwd><kwd>awake/drowsy states</kwd></kwd-group></article-meta></front><body><sec sec-type="intro" id="sec1-sensors-21-01734"><title>1. Introduction</title><p>Over the past three decades, we have seen changes in driving conditions and driver safety due to the vast efforts of research studies and government agencies. According to available estimates [<xref rid="B1-sensors-21-01734" ref-type="bibr">1</xref>], more than 1.3 million people die per year, and about 20 to 50 million people suffer non-fatal injuries due to road accidents. Drowsiness and fatigue, immediately after high speed and alcoholism [<xref rid="B2-sensors-21-01734" ref-type="bibr">2</xref>], are the main causes of traffic injuries in many areas such as aviation [<xref rid="B3-sensors-21-01734" ref-type="bibr">3</xref>], the military sector [<xref rid="B4-sensors-21-01734" ref-type="bibr">4</xref>] and driving [<xref rid="B5-sensors-21-01734" ref-type="bibr">5</xref>]. However, drowsiness detection (DD) researches [<xref rid="B6-sensors-21-01734" ref-type="bibr">6</xref>,<xref rid="B7-sensors-21-01734" ref-type="bibr">7</xref>] have been a subject of interest in recent years. This is now a real up to date problem in the current Covid-19 pandemic [<xref rid="B8-sensors-21-01734" ref-type="bibr">8</xref>] where medical equipment is commonly overbooked.</p><p>Drowsiness [<xref rid="B9-sensors-21-01734" ref-type="bibr">9</xref>] is an intermediate state between wakefulness and sleep. This state is mainly defined by heaviness in terms of reaction, changes in behavior, reflex reduction, and the difficulty of keeping the head in the frontal position of the vision field. In this regard, several means such as videos [<xref rid="B7-sensors-21-01734" ref-type="bibr">7</xref>,<xref rid="B10-sensors-21-01734" ref-type="bibr">10</xref>] and biomedical signals [<xref rid="B11-sensors-21-01734" ref-type="bibr">11</xref>,<xref rid="B12-sensors-21-01734" ref-type="bibr">12</xref>] have been targeted for DD. On the one side, the video-based applications for DD are efficient and robust against noise and lighting variations [<xref rid="B13-sensors-21-01734" ref-type="bibr">13</xref>]. Nevertheless, the biomedical signals are the best indicators of drowsiness relative to video features, according to [<xref rid="B14-sensors-21-01734" ref-type="bibr">14</xref>]. In this context, several biomedical signals, such as electroencephalogram (EEG) [<xref rid="B15-sensors-21-01734" ref-type="bibr">15</xref>], electrocardiogram (ECG) [<xref rid="B16-sensors-21-01734" ref-type="bibr">16</xref>], electromyogram (EMG) [<xref rid="B17-sensors-21-01734" ref-type="bibr">17</xref>] and electrooculogram (EOG), have been used for various DD studies [<xref rid="B18-sensors-21-01734" ref-type="bibr">18</xref>,<xref rid="B19-sensors-21-01734" ref-type="bibr">19</xref>,<xref rid="B20-sensors-21-01734" ref-type="bibr">20</xref>,<xref rid="B21-sensors-21-01734" ref-type="bibr">21</xref>]. Among them, EEG is probably the most efficient and promising modality of DD [<xref rid="B22-sensors-21-01734" ref-type="bibr">22</xref>,<xref rid="B23-sensors-21-01734" ref-type="bibr">23</xref>] thanks to various existing EEG-based technologies [<xref rid="B24-sensors-21-01734" ref-type="bibr">24</xref>]. Furthermore, this modality provides a good state of DD accuracy rate and also is more appropriate than percentage-of-eye-closure (PERCLOS) [<xref rid="B25-sensors-21-01734" ref-type="bibr">25</xref>] indicator in the evaluation of driver drowsiness. Thanks to its high temporal resolution, portability, and inexpensive cost, the <italic>Emotiv EPOC<inline-formula><mml:math id="mm2"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> (<uri xlink:href="https://www.emotiv.com/epoc/">https://www.emotiv.com/epoc/</uri> (accessed on 1 June 2020) headset [<xref rid="B26-sensors-21-01734" ref-type="bibr">26</xref>] is considered one of the most commonly used among the EEG-based technologies. The neurotechnology headset is a brain measuring data hardware that enables to record brain activity using fourteen electrodes placed on the participant’s scalp. In this paper, we focus on an EEG-based DD system using the <italic>Emotiv EPOC<inline-formula><mml:math id="mm3"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset to record brain activity by analyzing the drowsy or awake states.</p><p>Over this decade, many EEG-based research works related to machine learning (ML) [<xref rid="B27-sensors-21-01734" ref-type="bibr">27</xref>,<xref rid="B28-sensors-21-01734" ref-type="bibr">28</xref>,<xref rid="B29-sensors-21-01734" ref-type="bibr">29</xref>,<xref rid="B30-sensors-21-01734" ref-type="bibr">30</xref>] have been suggested in medical diagnosis, in particular for classification-based drowsiness detection tasks. Nevertheless, some limitations appear in ML applications such as the need for a massive dataset to train, limitation predictions in return, the need of an intermediary step for feature representation and drawing conclusions to detect anomalies.</p><p>In addition, deep learning (DL) researches [<xref rid="B31-sensors-21-01734" ref-type="bibr">31</xref>,<xref rid="B32-sensors-21-01734" ref-type="bibr">32</xref>] have recently shown notable progress in biomedical signal analysis especially classification-based anomaly detection. However, DL [<xref rid="B33-sensors-21-01734" ref-type="bibr">33</xref>] is now the fastest sub-field of ML technology [<xref rid="B34-sensors-21-01734" ref-type="bibr">34</xref>] based on the artificial neural networks (ANNs) [<xref rid="B35-sensors-21-01734" ref-type="bibr">35</xref>]. Interestingly, DL networks offer great potential for biomedical signals analysis through the simplification of raw input signals (i.e., through various steps including feature extraction, denoising, and feature selection) and the improvement of the classification results. Various DL models have been applied to biomedical signal analysis [<xref rid="B36-sensors-21-01734" ref-type="bibr">36</xref>] particularly for recurrent neural networks (RNNs) [<xref rid="B37-sensors-21-01734" ref-type="bibr">37</xref>], long short-term memory (LSTM) [<xref rid="B38-sensors-21-01734" ref-type="bibr">38</xref>], auto-encoder (AE) [<xref rid="B39-sensors-21-01734" ref-type="bibr">39</xref>], convolutional neural networks (CNNs) [<xref rid="B40-sensors-21-01734" ref-type="bibr">40</xref>], deep stacking networks (DSNs) [<xref rid="B41-sensors-21-01734" ref-type="bibr">41</xref>], etc. Among them, CNNs models [<xref rid="B42-sensors-21-01734" ref-type="bibr">42</xref>] are the most frequently used in biomedical signals classification for anomaly detection due to its high classification accuracy. In this sense, several biomedical signals-based CNNs studies [<xref rid="B43-sensors-21-01734" ref-type="bibr">43</xref>,<xref rid="B44-sensors-21-01734" ref-type="bibr">44</xref>,<xref rid="B45-sensors-21-01734" ref-type="bibr">45</xref>] have been suggested for anomaly detection tasks using various architectures such as CNN, visual geometry group network (VGGNet), Residual Network (ResNet), Dense Net, Inception Net, etc. In the present study, a CNN architecture is developed to classify the drowsy or awakeness states of each participant using an <italic>Emotiv EPOC<inline-formula><mml:math id="mm4"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset.</p><p>Along with the growing success of CNNs, the interest in data augmentation (DA) quickly increased. Numerous DL research works have integrated the DA technique [<xref rid="B46-sensors-21-01734" ref-type="bibr">46</xref>,<xref rid="B47-sensors-21-01734" ref-type="bibr">47</xref>] in the training step in order to avoid over-fitting and improve the performance of the networks by increasing accuracy. In our work, we integrated the DA technique to improve the performance of the proposed system.</p><p>According to [<xref rid="B48-sensors-21-01734" ref-type="bibr">48</xref>], the authors proposed an algorithm that uses features learned applying a CNN to capture various latent facial characteristics and various complex nonlinear characteristics. This system is used to warn the driver of drowsiness and to prevent traffic accidents. The trained classifier results give a classification accuracy equal to 92.33%. Likewise, in [<xref rid="B49-sensors-21-01734" ref-type="bibr">49</xref>], the authors used an RNNs architecture to detect driver fatigue in real-time. The experimental part presents good results (92.19%). In [<xref rid="B50-sensors-21-01734" ref-type="bibr">50</xref>], the authors propose a Complex Network-Based Broad Learning System (CNBLS) to differentiate between the fatigue and alert state using EEG signals. The experimental results showed an average accuracy of around 100%. In [<xref rid="B51-sensors-21-01734" ref-type="bibr">51</xref>], the authors suggest the detection of driver fatigue using a single EEG signal with the AlexNet CNN model. The achieved accuracy is respectively equal to 90% and 91%. According to [<xref rid="B52-sensors-21-01734" ref-type="bibr">52</xref>], a system composed of deep CNNs and deep residual learning with EEG signals is proposed to detect mental driver fatigue. The results showed an average accuracy reaching, respectively, to 91.788% and 92.682%. In [<xref rid="B53-sensors-21-01734" ref-type="bibr">53</xref>], the authors proposed a system to detect driver drowsiness based on differential entropy (DE) with a novel deep convolutional neural network. The experimental results showed an accuracy equal to 96%. In [<xref rid="B54-sensors-21-01734" ref-type="bibr">54</xref>], an EEG based prediction has been developed to transform the recorded EEG into an image liked feature map applying a CNN architecture. This approach offers a 40% detection score in the drowsy class.</p><p>The aim of our paper is to develop a new EEG-based DD system based on a CNN model. Our system is validated through individual performance assessment and comparison with other CNNs architectures used in biomedical signals analysis.</p><p>The rest of this paper is divided into four sections. In <xref ref-type="sec" rid="sec2-sensors-21-01734">Section 2</xref>, we introduce the suggested system using the <italic>Emotiv EPOC<inline-formula><mml:math id="mm5"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset. Moreover, we introduce the methodology used for EEG data acquisition as well as the architectures used for drowsiness analysis. In <xref ref-type="sec" rid="sec3-sensors-21-01734">Section 3</xref>, the experimental results of the proposed system are listed. A discussion is given in <xref ref-type="sec" rid="sec4-sensors-21-01734">Section 4</xref>. Finally, conclusions and future work are drawn in <xref ref-type="sec" rid="sec5-sensors-21-01734">Section 5</xref>.</p></sec><sec id="sec2-sensors-21-01734"><title>2. Materials and Methods</title><p>Our protocol introduces a new classification system between drowsiness or awakeness states using the <italic>Emotiv EPOC<inline-formula><mml:math id="mm6"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset to record 14 channels of EEG signals. The pipeline of the proposed system is represented in <xref ref-type="fig" rid="sensors-21-01734-f001">Figure 1</xref>. Data acquisition and model analysis are the two main procedures of our system. A detailed description of each procedure is given in the following subsections.</p><sec id="sec2dot1-sensors-21-01734"><title>2.1. Data Acquisition</title><p>The EEG data acquisition procedure consists of two main steps that are signal collection using the <italic>Emotiv EPOC<inline-formula><mml:math id="mm7"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset and data preprocessing. A description of each step is provided as follows.</p><sec><title>Signal Collection</title><p>The signal collection step is developed by two processes, which are the hardware and the software parts [<xref rid="B55-sensors-21-01734" ref-type="bibr">55</xref>]. The <italic>Emotiv EPOC<inline-formula><mml:math id="mm8"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> hardware is a non-invasive brain-computer interface (BCI) used for the development of the human brain and contextual research. <xref ref-type="fig" rid="sensors-21-01734-f002">Figure 2</xref> illustrates the various <italic>Emotiv EPOC<inline-formula><mml:math id="mm9"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> helmet components used in the experimental step consisting of a headset, a fourteen-sensors box, a USB key with cable for battery recharging that ensures the connection between the headset and the <italic>Emotiv Pro</italic> software, and a saline solution [<xref rid="B56-sensors-21-01734" ref-type="bibr">56</xref>] that ensures impedance and contact with the cortex. Compared to medical gel [<xref rid="B57-sensors-21-01734" ref-type="bibr">57</xref>], the saline solution is easy to use and maintains effective contact with the scalp of men and women.</p><p>The <italic>Emotiv EPOC<inline-formula><mml:math id="mm10"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset provides excellent access to professional-level brain data. As shown in <xref ref-type="fig" rid="sensors-21-01734-f003">Figure 3</xref>, this helmet contains fourteen active electrodes with two reference electrodes, which are Driven Right Leg (DRL) and Common Mode Sense (CMS). The electrodes are mounted around the participant’s scalp in the structures of the following zones: frontal and anterior parietal (AF3, AF4, F3, F4, F7, F8, FC5, FC6), temporal (T7, T8), and occipital-parietal (O1, O2, P7, P8). <xref rid="sensors-21-01734-t001" ref-type="table">Table 1</xref> presents some of the main characteristics of the <italic>Emotiv EPOC<inline-formula><mml:math id="mm11"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> helmet.</p><p>The <italic>EmotivPRO</italic> software allows visualizing the data streams in real-time including all data sources. This program configures the vertical scaling of the EEG Graphics with the multi-channel and single-channel display mode. Subsequently, the raw EEG data are exported in European Data Format (EDF) or Comma-Separated Values (CSV) formats that are considered as the input of the data preprocessing step.</p></sec></sec><sec id="sec2dot2-sensors-21-01734"><title>2.2. Data Preprocessing</title><p>The specific preprocessing steps of the EEG data revolve around the following points that are data preparation, signals annotation, and data augmentation.</p><sec id="sec2dot2dot1-sensors-21-01734"><title>2.2.1. Data Preparation</title><p>Various noise sources are targeted in the portion of the raw signal including eye blinks [<xref rid="B59-sensors-21-01734" ref-type="bibr">59</xref>,<xref rid="B60-sensors-21-01734" ref-type="bibr">60</xref>], dipolar size variance, muscle switches, inherent electrical properties and physical arrangement of various tissues [<xref rid="B61-sensors-21-01734" ref-type="bibr">61</xref>]. Data preprocessing is a preliminary step to EEG data denoising. In this context, various filters based on EEG denoising methods have been suggested as infinite impulse response (IIR) and finite impulse response (FIR) filters. Other sophisticated denoising approaches could be considered at the expense of higher computational complexity [<xref rid="B62-sensors-21-01734" ref-type="bibr">62</xref>,<xref rid="B63-sensors-21-01734" ref-type="bibr">63</xref>]. In our work, we propose to use an IIR filter that manages an impulsive signal within time and frequency domains. The IIR filter function can be expressed as the following discrete difference:<disp-formula id="FD1-sensors-21-01734"><label>(1)</label><mml:math id="mm12"><mml:mrow><mml:mrow><mml:mi>y</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>n</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:munderover><mml:mo>∑</mml:mo><mml:mrow><mml:mi>m</mml:mi><mml:mo>=</mml:mo><mml:mn>0</mml:mn></mml:mrow><mml:mi>M</mml:mi></mml:munderover><mml:msub><mml:mi>b</mml:mi><mml:mi>m</mml:mi></mml:msub><mml:mi>x</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>n</mml:mi><mml:mo>−</mml:mo><mml:mi>m</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>−</mml:mo><mml:munderover><mml:mo>∑</mml:mo><mml:mrow><mml:mi>m</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn></mml:mrow><mml:mi>N</mml:mi></mml:munderover><mml:msub><mml:mi>a</mml:mi><mml:mi>m</mml:mi></mml:msub><mml:mi>y</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>n</mml:mi><mml:mo>−</mml:mo><mml:mi>m</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>,</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula>where <italic>y</italic>(<italic>n</italic>) refers to the filtered signal, <italic>x</italic>(<italic>n</italic>) represents the input signal, <inline-formula><mml:math id="mm13"><mml:mrow><mml:msub><mml:mi>b</mml:mi><mml:mi>m</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> and <inline-formula><mml:math id="mm14"><mml:mrow><mml:msub><mml:mi>a</mml:mi><mml:mi>m</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> refer to the coefficients of the filter, and N represents the order of the filter. Subsequently, we convert the EEG signal from the time domain to the frequency domain using the fast Fourier transform (FFT) [<xref rid="B64-sensors-21-01734" ref-type="bibr">64</xref>]. The key task of the FFT is to take to 1024 samples from the input signal in the time domain and generate the output frequency of 128 Hz in the spectrum domain. In this work, alpha and theta waves analysis are accomplished using the FFT by adopting standardized EEG data.</p></sec><sec id="sec2dot2dot2-sensors-21-01734"><title>2.2.2. Signals Annotation</title><p>The central nervous system (CNS) [<xref rid="B65-sensors-21-01734" ref-type="bibr">65</xref>] consists of the spinal cord, the cerebellum, and the brain. The latter is divided into two parts: the right and left hemispheres. There are four lobes in each hemisphere, which are frontal, parietal, occipital, and temporal. Predominantly, the EEG signal is split into large spectral frequency bands related to EEG processors and rhythms of various frequency waves [<xref rid="B66-sensors-21-01734" ref-type="bibr">66</xref>,<xref rid="B67-sensors-21-01734" ref-type="bibr">67</xref>]. Brainwaves are usually classified into five frequency and amplitude bands [<xref rid="B66-sensors-21-01734" ref-type="bibr">66</xref>] including Gamma, Beta, Alpha, Theta, and Delta where each band wave refers to identifying states of participants. Other mixed bands, such as Alpha-Theta (5–9 Hz) [<xref rid="B68-sensors-21-01734" ref-type="bibr">68</xref>], have also appeared, which refers to waking and relaxation states. <xref rid="sensors-21-01734-t002" ref-type="table">Table 2</xref> presents a brief description of each brainwave from EEG signals.</p><p>The main functions associated with the six brainwave frequencies are described in the following in order to identify the electrodes that allow the detection of drowsy/awake states.</p><list list-type="bullet"><list-item><p>Gamma bands have a frequency ranging from 30 to 70 Hz and an amplitude value between 3 µV to 5 µV. These waves are used to detect Alzheimer’s disease [<xref rid="B69-sensors-21-01734" ref-type="bibr">69</xref>].</p></list-item><list-item><p>Beta wave is generated from the cortex region with frequency values from 13 to 30 Hz and a low amplitude ranging from 2 to 20 µV. These waves are related to awake states and various pathologies and symptoms of drugs.</p></list-item><list-item><p>Alpha band is produced from the thalamus area with a frequency ranging between 8 to 13 Hz and amplitude values between 20 to 60 µV. This band is detected with eyes closed to generating relaxation and awake states with attenuating drowsiness.</p></list-item><list-item><p>Theta wave is produced from the neocortex and hippocampus areas of the brain with frequency values from 4 to 7 Hz and an amplitude ranging from 20 to 100 µV. This band is correlated with a drowsiness state.</p></list-item><list-item><p>Delta wave is produced from the thalamus with a spectrum range of 4 Hz and an amplitude ranging from 20 to 200 µV. The wave is shown in the deep stage of sleep.</p></list-item><list-item><p>Alpha-Theta waves have a frequency ranging from 5 to 9 Hz and amplitude values between 20 to 100 µV. These bands refer to awake and drowsy states.</p></list-item></list><p>Furthermore, drowsiness is an intermediate state between awakeness (i.e., wakefulness) to sleep. During awakeness, <italic>beta</italic> waves are analyzed in the human brain [<xref rid="B70-sensors-21-01734" ref-type="bibr">70</xref>]. The drowsy stage is called stage 1 of sleep, the correlation is assured by <italic>alpha</italic> and <italic>theta</italic> bands [<xref rid="B71-sensors-21-01734" ref-type="bibr">71</xref>,<xref rid="B72-sensors-21-01734" ref-type="bibr">72</xref>,<xref rid="B73-sensors-21-01734" ref-type="bibr">73</xref>,<xref rid="B74-sensors-21-01734" ref-type="bibr">74</xref>]. The decrease in the <italic>alpha</italic> band and the rise in the <italic>theta</italic> frequency band expresses drowsiness [<xref rid="B75-sensors-21-01734" ref-type="bibr">75</xref>]. The drowsy state is a transitional phase between wakefulness and sleep, which is experienced in theta brain waves. This step is characterized by a decrease in the EEG waves frequency with an increase in their amplitude. The third and fourth steps are related to deep sleep, which is characterized by a low frequency and high amplitude fluctuation of the delta waves [<xref rid="B76-sensors-21-01734" ref-type="bibr">76</xref>]. According to this analysis, we support that the alpha-theta waves are the best bands for detecting the drowsy state. Our annotation is based on the study of <italic>Alpha-Theta</italic> waves for drowsiness/awakeness detection from, respectively, the occipital and temporal regions. The illustration of our annotation for the awake and drowsy states mentioned by O1, O2, T7, and T8 is shown in <xref ref-type="fig" rid="sensors-21-01734-f004">Figure 4</xref>. During the awakeness state, the amplitude is characterized by the lowest value while the drowsiness state is characterized by the highest value.</p></sec><sec id="sec2dot2dot3-sensors-21-01734"><title>2.2.3. Data Augmentation</title><p>In the recent year, DA [<xref rid="B77-sensors-21-01734" ref-type="bibr">77</xref>] has been shown to achieve significant performance for DL with increasing accuracy and stability and reducing over-fitting. As developed in [<xref rid="B46-sensors-21-01734" ref-type="bibr">46</xref>], DA is a process in which new data are artificially created from the current data on the training phase. In [<xref rid="B78-sensors-21-01734" ref-type="bibr">78</xref>], the need for developing a DA technique contributes to avoiding over-fitting, improves classification accuracy and stability [<xref rid="B47-sensors-21-01734" ref-type="bibr">47</xref>,<xref rid="B79-sensors-21-01734" ref-type="bibr">79</xref>] then better generalizes on new data and enhances performance in imbalanced class issues [<xref rid="B80-sensors-21-01734" ref-type="bibr">80</xref>]. Furthermore, DA allows improving the efficiency of CNN in the BCI field by overcoming the problems of small datasets. DA effectiveness varied considerably across techniques. However, sampling methods, noise addition, windows sliding, and Fourier transform are considered as the classic examples in signal classification tasks. Growingly, the DA techniques are used with DL networks on EEG signals works to generate new samples based on existing training data [<xref rid="B46-sensors-21-01734" ref-type="bibr">46</xref>]. This technique presents various advantages as it increases the model robustness against the variability of the input without decreasing the efficient capacity [<xref rid="B81-sensors-21-01734" ref-type="bibr">81</xref>]. In our work, DA steps are applied only to the training set in order to prevent over-fitting. The main idea of this procedure is to generate new samples by labeling retraining data transformations. The proposed DA method is considered as the opposite operation to dropout [<xref rid="B82-sensors-21-01734" ref-type="bibr">82</xref>] where a small volume of training data are duplicated randomly and appended to the training set. For instance, each EEG segment of the training set added a form of opposite operation to the dropout where the segments were extended by duplicating the vectors at random time points to a fixed length in the time dimension.</p></sec></sec><sec id="sec2dot3-sensors-21-01734"><title>2.3. Model Analysis</title><p>Choices of the different architectures and frameworks of DL used in the proposed DD system are argued by a comparative study. This section introduces our DD system based on comparative analysis.</p><sec id="sec2dot3dot1-sensors-21-01734"><title>2.3.1. Comparative Study</title><p>Simple CNN, ResNet, WaveNet, and Inception are among the best CNNs networks widely used in biomedical signals analysis studies. Based on recent works [<xref rid="B42-sensors-21-01734" ref-type="bibr">42</xref>,<xref rid="B83-sensors-21-01734" ref-type="bibr">83</xref>,<xref rid="B84-sensors-21-01734" ref-type="bibr">84</xref>,<xref rid="B85-sensors-21-01734" ref-type="bibr">85</xref>,<xref rid="B86-sensors-21-01734" ref-type="bibr">86</xref>,<xref rid="B87-sensors-21-01734" ref-type="bibr">87</xref>,<xref rid="B88-sensors-21-01734" ref-type="bibr">88</xref>,<xref rid="B89-sensors-21-01734" ref-type="bibr">89</xref>,<xref rid="B90-sensors-21-01734" ref-type="bibr">90</xref>,<xref rid="B91-sensors-21-01734" ref-type="bibr">91</xref>,<xref rid="B92-sensors-21-01734" ref-type="bibr">92</xref>,<xref rid="B93-sensors-21-01734" ref-type="bibr">93</xref>,<xref rid="B94-sensors-21-01734" ref-type="bibr">94</xref>,<xref rid="B95-sensors-21-01734" ref-type="bibr">95</xref>,<xref rid="B96-sensors-21-01734" ref-type="bibr">96</xref>,<xref rid="B97-sensors-21-01734" ref-type="bibr">97</xref>], a comparative analysis is provided in the following using various performance criteria as <italic>complexity</italic>, <italic>1D-dimension</italic>, <italic>performance</italic> and <italic>time-consumption</italic>. In this regard, specific three tests (2, 3 and 4 states) with various values are given for each criterion as following.
<list list-type="bullet"><list-item><p>2 states (0, 1),</p></list-item><list-item><p>3 states (0, 0.5, 1),</p></list-item><list-item><p>4 states (0, 0.33, 0.66, 1),</p></list-item></list>where 0 value is the low level, 1 value represents the high level, 0.33, 0.5, and 0.66 are intermediate levels. <xref rid="sensors-21-01734-t003" ref-type="table">Table 3</xref> indicates the score of the architectures with 2, 3, and 4 states.</p><p>For instance, 0 value indicates more <italic>complexity</italic> and <italic>time-consumption</italic>, low <italic>performance</italic> and unused for <italic>1D-dimension</italic>, while a value of 1 indicates less <italic>complexity</italic> and <italic>time-consumption</italic>, high <italic>performance</italic> and widely used for <italic>1D-dimension</italic>. The highest score is identified by the best architecture used in biomedical signals classification. According to the reported results, the high total value is presented by the Simple CNN architecture.</p><p>As regards the choice of the DL framework, there are numerous open-source frameworks [<xref rid="B98-sensors-21-01734" ref-type="bibr">98</xref>,<xref rid="B99-sensors-21-01734" ref-type="bibr">99</xref>], such as keras [<xref rid="B100-sensors-21-01734" ref-type="bibr">100</xref>], tensorflow [<xref rid="B101-sensors-21-01734" ref-type="bibr">101</xref>], and pytorch [<xref rid="B102-sensors-21-01734" ref-type="bibr">102</xref>]. In the developing of DL models, the Keras framework offers a high level in build blocks by using particular libraries, such as TensorFlow, dedicated for operations characterized by a low level [<xref rid="B103-sensors-21-01734" ref-type="bibr">103</xref>]. In this context, we have used the Keras DL library with a sequential model applied to the binary classification. Keras is used to build the architectures with TensorFlow backend [<xref rid="B104-sensors-21-01734" ref-type="bibr">104</xref>]. This framework presents high-level application programming interfaces (APIs) developed on top of TensorFlow. This model is characterized by its easy use and its simplicity.</p><p>Regarding the choice of the optimization algorithm, many optimizers exist in the literature such as Adam [<xref rid="B105-sensors-21-01734" ref-type="bibr">105</xref>], Stochastic Gradient Descent Optimizer (SGD) [<xref rid="B106-sensors-21-01734" ref-type="bibr">106</xref>] and Root Mean Square Propagation (RMS prop) [<xref rid="B107-sensors-21-01734" ref-type="bibr">107</xref>]. In this context, SGD is the most popular optimizer, which is simple and effective for finding optimal values in a neural network. In this work, we have used an SGD optimizer.</p></sec><sec id="sec2dot3dot2-sensors-21-01734"><title>2.3.2. Proposed Simple CNN Model</title><p>The diagram of the proposed CNN used in our DD system is presented in <xref ref-type="fig" rid="sensors-21-01734-f005">Figure 5</xref>. All the EEG windows with 3.75 s are the input of our proposed model. Via four convolutional and one max-pooling layers, EEG signals move followed by seven batch-normalization and one fully connected layer. All layers are equipped with the activation function of the rectified linear unit (ReLU). The pooling process chooses the maximum pooling procedure that can accomplish both reduction of dimensionality and invariance. In addition, dropout processing [<xref rid="B82-sensors-21-01734" ref-type="bibr">82</xref>] is used to reduce the risk of over-fitting. Throughout the structure of our network, the fully connected layer serves as a classifier when mapping between high and low dimensions. The different layers of the proposed CNN model used in our DD system are detailed in the following.</p><list list-type="bullet"><list-item><p><bold>Convolutional layers</bold></p><p>The layers allow filter application and features extraction [<xref rid="B108-sensors-21-01734" ref-type="bibr">108</xref>] based on the input EEG signals. The equation below presents the convolution operation.
<disp-formula id="FD2-sensors-21-01734"><label>(2)</label><mml:math id="mm15"><mml:mrow><mml:mrow><mml:msub><mml:mi>Y</mml:mi><mml:mi>i</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:msub><mml:mi>b</mml:mi><mml:mi>i</mml:mi></mml:msub><mml:mo>+</mml:mo><mml:munder><mml:mo>∑</mml:mo><mml:mi>n</mml:mi></mml:munder><mml:msub><mml:mi>W</mml:mi><mml:mrow><mml:mi>i</mml:mi><mml:mi>n</mml:mi></mml:mrow></mml:msub><mml:mo>∗</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mi>n</mml:mi></mml:msub><mml:mo>,</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula>where ∗ is the convolution operation, <inline-formula><mml:math id="mm16"><mml:mrow><mml:msub><mml:mi>Y</mml:mi><mml:mi>i</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> presents the feature map, <inline-formula><mml:math id="mm17"><mml:mrow><mml:msub><mml:mi>b</mml:mi><mml:mi>i</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> is the bias term, <inline-formula><mml:math id="mm18"><mml:mrow><mml:msub><mml:mi>W</mml:mi><mml:mrow><mml:mi>i</mml:mi><mml:mi>n</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> is the sub-kernel of channel and <inline-formula><mml:math id="mm19"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mi>n</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> is the input signal. <xref rid="sensors-21-01734-t004" ref-type="table">Table 4</xref> presents a description of the four convolutional layers purpose.</p></list-item><list-item><p><bold>BatchNormalization layers</bold></p><p>As known in DL, there are two fundamental problems [<xref rid="B109-sensors-21-01734" ref-type="bibr">109</xref>], which are the over-fitting and the long training duration. The Batch Normalization (BN) layers are used to scale and speed up the learning process. Accordingly, each BN stratum normalizes the previous activation layer by subtracting the average batches, as well as divides it by the standard deviation.</p></list-item><list-item><p><bold>Dropout layer</bold></p><p>Each dropout layer is considered as a regularization technique and allows to improve over-adjustment on neural networks in which it decreases the error rate in the classification process. In the proposed model, the value of dropout is equal to 0.2. To avoid over-fitting, we have inactivated 20% of the neurons. We have used three dropout layers in our model.</p></list-item><list-item><p><bold>Max-Pooling1D layer</bold></p><p>The sample-based discretization max-pooling-1D blocks is used to sub-sample each input layer by reducing its dimensionality and decreasing the number of the parameters to learn, thereby reducing calculation costs.</p></list-item><list-item><p><bold>Flatten layer</bold></p><p>A multidimensional data output is given in the previous step, which cannot be read directly from this neural network, and the model is therefore flattened.</p></list-item><list-item><p><bold>Dense layers</bold></p><p>The dense layer has the role of describing the connectivity with the next and intermediate layers of neurons. We have used two fully connected layers in our architecture. In the first dense of our model, we used a hidden layer of 128 neurons to have better classification results. For the second dense, the value of the final neuron is equal to 1. Binary classification is applied in this work, so a single neuron is sufficient to denote class “1” or “0”.</p></list-item></list></sec></sec></sec><sec id="sec3-sensors-21-01734"><title>3. Experimental Validation</title><p>A description of our dataset and experiments without and with DA were provided in the following subsections for the efficiency assessment of the proposed DD scheme. Our experiments have been performed using the power of GPU (Graphical Processing Unit) provided by the Google AI (Artificial Intelligence) platform and Colab [<xref rid="B110-sensors-21-01734" ref-type="bibr">110</xref>].</p><sec id="sec3dot1-sensors-21-01734"><title>3.1. Dataset</title><p>Our EEG signal collection contains forty-two records of six men and eight women aged between fourteen and sixty-four with normal mental health. For each person, we made three recordings lasting sixteen minutes over the day: in the morning, afternoon, and evening. For each recording, the total number of rows of data is equal to 123,648. In order to identify the state of each participant, we divided the EEG signal into windows of 3.75 s. In this sense, we split each EEG recording into 256 different sets (segments) and the length of each segment is equal to 483. Based on the proposed data annotation step in our method, a deeper analysis of the brain is the preliminary phase in the detection of each participant’s state. In this regard, we categorized the different participants according to drowsy and awake states. <xref rid="sensors-21-01734-t005" ref-type="table">Table 5</xref> presents the detailed results for each participating state.</p><p>Our data are divided into two parts, with 80% and 20%, respectively, as training for the train model and testing for the predict model. There were (20,286, 256) recordings in total in which (16,422, 256) were used for training and (3864, 256) for testing. Therefore, the training set data is divided into two parts, with 80% and 20% as training and validation. There were (16,422, 256) recording in total in which (13,137, 256) were used for training and (3285, 256) for validation.</p></sec><sec id="sec3dot2-sensors-21-01734"><title>3.2. Experimental Details</title><p>The different parameters as filters, kernel-size, padding, kernel-initializer, and activation of the four convolutional layers have the same values, respectively, 512, 32, same, normal, and relu. The parameter values of the remaining layers are detailed in <xref rid="sensors-21-01734-t006" ref-type="table">Table 6</xref>.</p><p>We aim to reach the best accuracy rate by using a minimum number of electrodes that provide information about the drowsiness state. In [<xref rid="B111-sensors-21-01734" ref-type="bibr">111</xref>,<xref rid="B112-sensors-21-01734" ref-type="bibr">112</xref>,<xref rid="B113-sensors-21-01734" ref-type="bibr">113</xref>], the authors discover that the pre-frontal and occipital cortex are the most important channel to better diagnose the drowsiness state. Furthermore, previous work [<xref rid="B114-sensors-21-01734" ref-type="bibr">114</xref>] indicates that occipital, parietal, central and frontal regions are useful for drowsiness detection. According to the recent related work [<xref rid="B115-sensors-21-01734" ref-type="bibr">115</xref>], the authors provide that the frontal, occipital and parietal are the best selected areas for DD. To select the relevant channels that enable the best accuracy in the proposed DD system, we suggest comparing the different results recorded by various numbers of electrodes. To reach the converge of our model, we used 15 epochs for all experiments. To this regard, we choose the following recorded data:<list list-type="bullet"><list-item><p>Recording by 14 electrodes including the frontal and the anterior parietal (AF3, AF4, F3, F4, F7, F8, FC5, FC6), the temporal (T7, T8), and the occipital-parietal (O1, O2, P7, P8).</p></list-item><list-item><p>Recording by 7 (AF3, F7, F3, T7, O2, P8, F8) electrodes from parietal, occipital, pre-frontal and temporal areas.</p></list-item><list-item><p>Recording by 4 (T7,T8, O1 and O2) electrodes from the temporal and occipital areas.</p></list-item><list-item><p>Recording by 2 (O1 and O2) electrodes from the occipital area.</p></list-item></list></p><sec id="sec3dot2dot1-sensors-21-01734"><title>3.2.1. Experiments without DA</title><p><xref rid="sensors-21-01734-t007" ref-type="table">Table 7</xref> presents the reported testing and training accuracies, respectively, with two, four, seven, and fourteen electrodes. From the reported results, the different accuracy values related to the training and validation sets as well as testing sets are low. One can notice that the training accuracy is quite stable over different electrode configurations, while test accuracy presents more disparity and lower values. These high classification error rates on the testing set indicate low generalization capacity of the proposed model when used without DA.</p><p>In the next experiments, a DA step is added to the training set to improve the classification performance (accuracy) of the proposed DD system, thereafter to select the most efficient number of electrodes associated with the best results.</p></sec><sec id="sec3dot2dot2-sensors-21-01734"><title>3.2.2. Experiments with DA</title><p>In the present work, we solve the data limitation problem by adding the DA step to increase the performance of the proposed CNN model. The DA step is only processed for the training set by using 20 duplicates. In this regard, the vector value of the training set is doubling from (13,524, 256) to (132,058, 256). The reported training, validation and testing accuracies, respectively, with two, four, seven, and fourteen electrodes are presented in <xref rid="sensors-21-01734-t008" ref-type="table">Table 8</xref>. We can notice that DA allows to drastically improve the performance of the proposed model while used with seven electrodes, especially for the testing set. As regards training, the four configurations perform similarly with very good accuracies.</p><p>After evaluating our model with the use of the DA technique, we can select the best acquisition configuration, i.e., seven electrodes. To this regard, we use AF3, F7, F3 and F8 electrodes from the frontal, T7 the temporal, O2 the occipital and P8 the parietal areas. The values mentioned in <xref rid="sensors-21-01734-t008" ref-type="table">Table 8</xref> present the average accuracies of three runs for each experiment. <xref rid="sensors-21-01734-t009" ref-type="table">Table 9</xref> gives an example of the average accuracy of seven electrodes with DA.</p><p>Using the selected electrodes, <xref ref-type="fig" rid="sensors-21-01734-f006">Figure 6</xref> displays the training and validation accuracy and loss. Using 15 epochs, we find that the train and validation accuracy improves, and the training and validation loss decreases. This shows that the proposed CNN-based DD system has been trained to achieve up to 98.81% highest training accuracy with 90.42% highest testing accuracy for the prediction in order to automatically classify the EEG signals in drowsy/awake states.</p><p>To further quantitatively evaluate the performance of the proposed model, True Positive (TP), True Negative (TN), False Positive (FP), and False Negative (FN) rates are used to evaluate metrics [<xref rid="B116-sensors-21-01734" ref-type="bibr">116</xref>] such as accuracy, precision, recall, and F1 score calculated as follows:<disp-formula id="FD3-sensors-21-01734"><label>(3)</label><mml:math id="mm20"><mml:mrow><mml:mrow><mml:mi>A</mml:mi><mml:mi>c</mml:mi><mml:mi>c</mml:mi><mml:mi>u</mml:mi><mml:mi>r</mml:mi><mml:mi>a</mml:mi><mml:mi>c</mml:mi><mml:mi>y</mml:mi><mml:mo>=</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>T</mml:mi><mml:mi>P</mml:mi><mml:mo>+</mml:mo><mml:mi>T</mml:mi><mml:mi>N</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>/</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>T</mml:mi><mml:mi>P</mml:mi><mml:mo>+</mml:mo><mml:mi>T</mml:mi><mml:mi>N</mml:mi><mml:mo>+</mml:mo><mml:mi>F</mml:mi><mml:mi>P</mml:mi><mml:mo>+</mml:mo><mml:mi>F</mml:mi><mml:mi>N</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>.</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula><disp-formula id="FD4-sensors-21-01734"><label>(4)</label><mml:math id="mm21"><mml:mrow><mml:mrow><mml:mi>P</mml:mi><mml:mi>r</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>i</mml:mi><mml:mi>s</mml:mi><mml:mi>i</mml:mi><mml:mi>o</mml:mi><mml:mi>n</mml:mi><mml:mo>=</mml:mo><mml:mrow><mml:mi>T</mml:mi><mml:mi>P</mml:mi></mml:mrow><mml:mo>/</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>T</mml:mi><mml:mi>P</mml:mi><mml:mo>+</mml:mo><mml:mi>F</mml:mi><mml:mi>P</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>.</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula><disp-formula id="FD5-sensors-21-01734"><label>(5)</label><mml:math id="mm22"><mml:mrow><mml:mrow><mml:mi>R</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>a</mml:mi><mml:mi>l</mml:mi><mml:mi>l</mml:mi><mml:mo>=</mml:mo><mml:mrow><mml:mi>T</mml:mi><mml:mi>P</mml:mi></mml:mrow><mml:mo>/</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>T</mml:mi><mml:mi>P</mml:mi><mml:mo>+</mml:mo><mml:mi>F</mml:mi><mml:mi>N</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>.</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula><disp-formula id="FD6-sensors-21-01734"><label>(6)</label><mml:math id="mm23"><mml:mrow><mml:mrow><mml:mi>F</mml:mi><mml:mn>1</mml:mn><mml:mi>s</mml:mi><mml:mi>c</mml:mi><mml:mi>o</mml:mi><mml:mi>r</mml:mi><mml:mi>e</mml:mi><mml:mo>=</mml:mo><mml:mo>(</mml:mo><mml:mrow><mml:mi>P</mml:mi><mml:mi>r</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>i</mml:mi><mml:mi>s</mml:mi><mml:mi>i</mml:mi><mml:mi>o</mml:mi><mml:mi>n</mml:mi><mml:mo>∗</mml:mo><mml:mi>R</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>a</mml:mi><mml:mi>l</mml:mi><mml:mi>l</mml:mi></mml:mrow><mml:mo>)</mml:mo><mml:mo>/</mml:mo><mml:mo>(</mml:mo><mml:mrow><mml:mi>P</mml:mi><mml:mi>r</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>i</mml:mi><mml:mi>s</mml:mi><mml:mi>i</mml:mi><mml:mi>o</mml:mi><mml:mi>n</mml:mi><mml:mo>+</mml:mo><mml:mi>R</mml:mi><mml:mi>e</mml:mi><mml:mi>c</mml:mi><mml:mi>a</mml:mi><mml:mi>l</mml:mi><mml:mi>l</mml:mi></mml:mrow><mml:mo>)</mml:mo><mml:mo>.</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula></p><p>In the experimental configuration with DA, the highest accuracy value is equal to 90.42%, the precision is equal to 86.51%, the recall value is equal to 89%, while the F1-score value is equal to 88%. This high precision rate indicates the capacity of the model to not miss drowsy alarms.</p><p>To visualize the performance of the proposed model, we used the confusion matrix that is represented in <xref ref-type="fig" rid="sensors-21-01734-f007">Figure 7</xref>, where 2667 presents the TP, 231 presents the FP, 139 presents the FN and 827 presents the TN.</p><p>Additionally, the cross validation method is used in order to evaluate our model with seven electrodes. <xref rid="sensors-21-01734-t010" ref-type="table">Table 10</xref> presents all the experimental results with different folds.</p></sec></sec><sec id="sec3dot3-sensors-21-01734"><title>3.3. Comparison</title><p>In order to evaluate the effectiveness of the proposed CNN model, we compared the performance measures of our model with that of several different CNNs architectures, as mentioned in <xref rid="sensors-21-01734-t011" ref-type="table">Table 11</xref>, such as Inception (Conv1d_3, Conv1d_4, Conv1d_5, Max_Pooling1d_1, Concatenate_1, Batch_Normalization, Dropout, Flatten, Dense, Batch_Normalization and Dense_), WaveNet (import WaveNet) and ResNet (Conv1d_46, Conv1d_47, Conv1d_45, Add_14, Activation_14, Batchnormalization_14, Dropout_7, Flatten_5, Dense_17, Batchnormalization_15 and Dense_18).</p><p>Additionally, we compare our work with recent DD systems in the literature. In [<xref rid="B54-sensors-21-01734" ref-type="bibr">54</xref>], the authors propose a system based on the EEG signal processing image, which converts the EEG signal into an image-like signal 2-D function map and then transfers them to the CNN model for DD. This architecture is composed of two convolutional and pooling layers with one fully connected layer. The total accuracy in the prediction imbalanced dataset result is equal to 71.15%. In [<xref rid="B40-sensors-21-01734" ref-type="bibr">40</xref>], the authors suggest a DD system based on a DL model. Using spectrograms from the channels of EEG signals, the proposed system is developed to the ULg Multimodality Drowsiness Database. The used ConVNets model is composed of three convolutional and max-pooling layers with one fully connected layer. An accuracy of 86% is achieved in this work. We implement these two DL architectures using our EEG data. <xref rid="sensors-21-01734-t012" ref-type="table">Table 12</xref> indicates the accuracy values of the testing set using the competing DD systems. It is noteworthy that the proposed DD system gives the best accuracy classification of drowsy/awake states.</p></sec></sec><sec sec-type="discussion" id="sec4-sensors-21-01734"><title>4. Discussion</title><p>EEG data are being increasingly used to analyze drowsiness through the control of mental states, fatigue progression, and tiredness over time [<xref rid="B117-sensors-21-01734" ref-type="bibr">117</xref>]. Interestingly, reported studies in the literature indicate a specific trend to reduce the number of used electrodes [<xref rid="B118-sensors-21-01734" ref-type="bibr">118</xref>,<xref rid="B119-sensors-21-01734" ref-type="bibr">119</xref>]. From a practical point of view, reducing the number of electrodes ensures better comfort for the driver. In this paper, we started by using fourteen electrodes and we reduced the number to seven, four, and two electrodes. However, brain regions, such as the parietal, frontal, and occipital lobes, tend to be more vulnerable than other areas for DD. To this regard, alpha and theta waves from the occipital and the temporal area reveal a high indicator for DD. During drowsiness, exhaustion, and insufficient attention, the alpha band demonstrates an increase in-band power, while the theta band indicates the state of deep relaxation during the first phase of slow sleep. In fact, these waves reflect the state between sleep and wholeness. Therefore, comparative behavioral testing of alpha and theta waves can be beneficial for effective DD. The proposed DD system is divided into two steps as data acquisition and model analysis. The first step contains three steps, signal collection, data annotation, and data augmentation (DA). An <italic>Emotiv EPOC<inline-formula><mml:math id="mm24"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset is used for signal collection. Subsequently, we have annotated our dataset according to the amplitudes of alpha and theta waves. By incorporating the DA step to improve performance, we have done two experimental tests: with and without DA. For model analysis, we have built a CNN model in which implementation is done using the Keras framework. The average values of the accuracy, F1-score, precision, and recall showed a high classification rate using seven electrodes, in comparison to other competing methods.</p></sec><sec sec-type="conclusions" id="sec5-sensors-21-01734"><title>5. Conclusions and Future Work</title><p>This paper proposes a new DD system based on EEG signals using a CNN architecture. An <italic>Emotiv EPOC<inline-formula><mml:math id="mm25"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset is used for signal collection. Furthermore, our EEG data has been annotated to detect drowsiness based on the analysis of alpha and theta waves from the occipital and temporal area. A study has been conducted to select the most suitable number of electrodes. Obtained results are coherent with the state-of-the-art. In this context, we proposed a system for DD using only seven electrodes. The proposed system achieves an average classification accuracy of 90.14%. In future work, EEG can be considered with other physiological assessment tools, such as EOG, ECG and Near-Infrared Spectroscopy (NIRS) [<xref rid="B120-sensors-21-01734" ref-type="bibr">120</xref>,<xref rid="B121-sensors-21-01734" ref-type="bibr">121</xref>], which help to improve accuracy rate. We will also consider validating our system on larger datasets, especially collected under real driving conditions.</p></sec></body><back><fn-group><fn><p><bold>Publisher’s Note:</bold> MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.</p></fn></fn-group><notes><title>Author Contributions</title><p>Conceptualization, S.C., B.B., A.B., A.H., A.A. and L.C.; data curation, B.B.; formal analysis, B.B.; funding acquisition, A.H. and A.A.; investigation, S.C., B.B., A.B., A.A. and L.C.; methodology, S.C., B.B., A.B. and L.C.; project administration, B.B. and L.C.; software, S.C., B.B. and A.B.; supervision, B.B., A.A. and L.C.; validation, B.B. and L.C.; writing—original draft, S.C., B.B. and A.B.; writing—review and editing, A.H., A.A. and L.C. All authors have read and agreed to the published version of the manuscript.</p></notes><notes><title>Funding</title><p>This research received no external funding.</p></notes><notes><title>Institutional Review Board Statement</title><p>The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the local Institutional Review Board.</p></notes><notes><title>Informed Consent Statement</title><p>Informed consent was obtained from all subjects involved in the study.</p></notes><notes notes-type="data-availability"><title>Data Availability Statement</title><p>Data are available from the authors (S.C., B.B., or A.B.) upon reasonable request. Dataset: <uri xlink:href="https://github.com/bassem-bouaziz/Drowsiness_Detection">https://github.com/bassem-bouaziz/Drowsiness_Detection</uri> (accessed on 1 June 2020).</p></notes><notes notes-type="COI-statement"><title>Conflicts of Interest</title><p>The authors declare no conflict of interest.</p></notes><glossary><title>Abbreviations</title><p>The following abbreviations are used in this manuscript:
<array orientation="portrait"><tbody><tr><td align="left" valign="middle" rowspan="1" colspan="1">ANNs</td><td align="left" valign="middle" rowspan="1" colspan="1">Artificial Neural Networks</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">AE</td><td align="left" valign="middle" rowspan="1" colspan="1">Auto-encoder</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">APIs</td><td align="left" valign="middle" rowspan="1" colspan="1">Application Programming Interfaces</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">AI</td><td align="left" valign="middle" rowspan="1" colspan="1">Artificial Intelligence</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">BCI</td><td align="left" valign="middle" rowspan="1" colspan="1">Brain Computer Interface</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">BN</td><td align="left" valign="middle" rowspan="1" colspan="1">BatchNormalization</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">CNN</td><td align="left" valign="middle" rowspan="1" colspan="1">Convolutional Neural Network</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">CNBLS </td><td align="left" valign="middle" rowspan="1" colspan="1">Complex Network-based Broad Learning System</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">CMS</td><td align="left" valign="middle" rowspan="1" colspan="1">Common Mode Sense</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">CSV</td><td align="left" valign="middle" rowspan="1" colspan="1">Comma Separated Values</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">CNS</td><td align="left" valign="middle" rowspan="1" colspan="1">Central Nervous System</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DD</td><td align="left" valign="middle" rowspan="1" colspan="1">Drowsiness Detection</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DL</td><td align="left" valign="middle" rowspan="1" colspan="1">Deep Learning</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DA</td><td align="left" valign="middle" rowspan="1" colspan="1">Data Augmentation</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DSNs</td><td align="left" valign="middle" rowspan="1" colspan="1">Deep Stacking Networks</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DE</td><td align="left" valign="middle" rowspan="1" colspan="1">Differential Entropy</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">DRL</td><td align="left" valign="middle" rowspan="1" colspan="1">Driven Right Leg</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">EEG</td><td align="left" valign="middle" rowspan="1" colspan="1">Electroencephalogram</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">ECG</td><td align="left" valign="middle" rowspan="1" colspan="1">Electrocardiogram</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">EMG</td><td align="left" valign="middle" rowspan="1" colspan="1">Electromyogram</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">EOG</td><td align="left" valign="middle" rowspan="1" colspan="1">Electrooculogram</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">EDF</td><td align="left" valign="middle" rowspan="1" colspan="1">European Data Interface</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">FIR</td><td align="left" valign="middle" rowspan="1" colspan="1">Finite Impulse Response</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">FFT</td><td align="left" valign="middle" rowspan="1" colspan="1">Fast Fourier Transformation</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">FP</td><td align="left" valign="middle" rowspan="1" colspan="1">False Positive</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">FN</td><td align="left" valign="middle" rowspan="1" colspan="1">False Negative</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">GPU</td><td align="left" valign="middle" rowspan="1" colspan="1">Graphics Processing Unit</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">IIR</td><td align="left" valign="middle" rowspan="1" colspan="1">Infinite Impulse Response</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">LSTM</td><td align="left" valign="middle" rowspan="1" colspan="1">Long Short Term Memory</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">ML</td><td align="left" valign="middle" rowspan="1" colspan="1">Machine Learning</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">NIRS</td><td align="left" valign="middle" rowspan="1" colspan="1">Near Infrared Spectroscopy</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">PERCLOS</td><td align="left" valign="middle" rowspan="1" colspan="1">Percentage of eye closure</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">RNNs</td><td align="left" valign="middle" rowspan="1" colspan="1">Recurrent Neural Networks</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">ResNet</td><td align="left" valign="middle" rowspan="1" colspan="1">Residual Network</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">RMSprop</td><td align="left" valign="middle" rowspan="1" colspan="1">Root Mean Squence Propagation</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">ReLU</td><td align="left" valign="middle" rowspan="1" colspan="1">Rectified Linear Unit</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">SGD</td><td align="left" valign="middle" rowspan="1" colspan="1">Stochastic Gradient Descent Optimizer</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">TP</td><td align="left" valign="middle" rowspan="1" colspan="1">True Positive</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">TN</td><td align="left" valign="middle" rowspan="1" colspan="1">True Negative</td></tr><tr><td align="left" valign="middle" rowspan="1" colspan="1">VGGNet</td><td align="left" valign="middle" rowspan="1" colspan="1">Visual Geometry Group Network</td></tr></tbody></array></p></glossary><ref-list><title>References</title><ref id="B1-sensors-21-01734"><label>1.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sahayadhas</surname><given-names>A.</given-names></name><name><surname>Sundaraj</surname><given-names>K.</given-names></name><name><surname>Murugappan</surname><given-names>M.</given-names></name><name><surname>Palaniappan</surname><given-names>R.</given-names></name></person-group><article-title>Physiological Signal based Detection of Driver Hypovigilance using Higher Order Spectra</article-title><source>Expert Syst. Appl.</source><year>2015</year><volume>42</volume><fpage>8669</fpage><lpage>8677</lpage><pub-id pub-id-type="doi">10.1016/j.eswa.2015.07.021</pub-id></element-citation></ref><ref id="B2-sensors-21-01734"><label>2.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ghandour</surname><given-names>A.</given-names></name><name><surname>Hammoud</surname><given-names>H.</given-names></name><name><surname>Al-Hajj</surname><given-names>S.</given-names></name></person-group><article-title>Analyzing Factors Associated with Fatal Road Crashes: A Machine Learning Approach</article-title><source>Int. J. Environ. Res. Public Health</source><year>2020</year><volume>17</volume><elocation-id>4111</elocation-id><pub-id pub-id-type="doi">10.3390/ijerph17114111</pub-id><pub-id pub-id-type="pmid">32526945</pub-id></element-citation></ref><ref id="B3-sensors-21-01734"><label>3.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Thomas</surname><given-names>L.</given-names></name><name><surname>Gast</surname><given-names>C.</given-names></name><name><surname>Grube</surname><given-names>R.</given-names></name><name><surname>Craig</surname><given-names>K.</given-names></name></person-group><article-title>Fatigue Detection in Commercial Flight Operations: Results Using Physiological Measures</article-title><source>Procedia Manuf.</source><year>2015</year><volume>3</volume><fpage>2357</fpage><lpage>2364</lpage><pub-id pub-id-type="doi">10.1016/j.promfg.2015.07.383</pub-id></element-citation></ref><ref id="B4-sensors-21-01734"><label>4.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Neri</surname><given-names>D.</given-names></name><name><surname>Shappell</surname><given-names>S.</given-names></name><name><surname>DeJohn</surname><given-names>C.</given-names></name></person-group><article-title>Simulated Sustained Flight Operations and Performance, Part 1: Effects of Fatigue</article-title><source>Mil. Psychol.</source><year>1992</year><volume>4</volume><fpage>137</fpage><lpage>155</lpage><pub-id pub-id-type="doi">10.1207/s15327876mp0403_2</pub-id></element-citation></ref><ref id="B5-sensors-21-01734"><label>5.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hu</surname><given-names>J.</given-names></name><name><surname>Wang</surname><given-names>P.</given-names></name></person-group><article-title>Noise Robustness Analysis of Performance for EEG-Based Driver Fatigue Detection Using Different Entropy Feature Sets</article-title><source>Entropy</source><year>2017</year><volume>19</volume><fpage>385</fpage></element-citation></ref><ref id="B6-sensors-21-01734"><label>6.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Choi</surname><given-names>Y.</given-names></name><name><surname>Kwon</surname><given-names>N.</given-names></name><name><surname>Lee</surname><given-names>S.</given-names></name><name><surname>Shin</surname><given-names>Y.</given-names></name><name><surname>Ryo</surname><given-names>C.</given-names></name><name><surname>Park</surname><given-names>J.</given-names></name><name><surname>Shin</surname><given-names>D.</given-names></name></person-group><article-title>Hypovigilance Detection for UCAV Operators Based on a Hidden Markov Model</article-title><source>Comput. Math. Methods Med.</source><year>2014</year><volume>2014</volume><fpage>567</fpage><lpage>645</lpage><pub-id pub-id-type="doi">10.1155/2014/567645</pub-id></element-citation></ref><ref id="B7-sensors-21-01734"><label>7.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Murugan</surname><given-names>S.</given-names></name><name><surname>Selvaraj</surname><given-names>J.</given-names></name><name><surname>Sahayadhas</surname><given-names>A.</given-names></name></person-group><article-title>Driver Hypovigilance Detection for Safe Driving using Infrared Camera</article-title><source>Proceedings of the Fifth International Conference on Inventive Computation Technologies (ICICT)</source><conf-loc>Tamilnadu, India</conf-loc><conf-date>26–28 February 2020</conf-date><fpage>413</fpage><lpage>418</lpage></element-citation></ref><ref id="B8-sensors-21-01734"><label>8.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chaari</surname><given-names>L.</given-names></name><name><surname>Golubnitschaja</surname><given-names>O.</given-names></name></person-group><article-title>Covid-19 pandemic by the “real-time” monitoring: The Tunisian case and lessons for global epidemics in the context of 3PM strategies</article-title><source>EPMA J.</source><year>2020</year><volume>11</volume><fpage>133</fpage><lpage>138</lpage><pub-id pub-id-type="doi">10.1007/s13167-020-00207-0</pub-id></element-citation></ref><ref id="B9-sensors-21-01734"><label>9.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gwak</surname><given-names>J.</given-names></name><name><surname>Hirao</surname><given-names>A.</given-names></name><name><surname>Shino</surname><given-names>M.</given-names></name></person-group><article-title>An Investigation of Early Detection of Driver Drowsiness Using Ensemble Machine Learning Based on Hybrid Sensing</article-title><source>Appl. Sci.</source><year>2020</year><volume>10</volume><elocation-id>2890</elocation-id><pub-id pub-id-type="doi">10.3390/app10082890</pub-id></element-citation></ref><ref id="B10-sensors-21-01734"><label>10.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Houssaini</surname><given-names>A.</given-names></name><name><surname>Sabri</surname><given-names>A.</given-names></name><name><surname>Qjidaa</surname><given-names>H.</given-names></name><name><surname>Aarab</surname><given-names>A.</given-names></name></person-group><article-title>Real-Time Driver’s Hypovigilance Detection using Facial Landmarks</article-title><source>Proceedings of the International Conference on Wireless Technologies, Embedded and Intelligent Systems (WITS)</source><conf-loc>Fez, Morocco</conf-loc><conf-date>30 May 2019</conf-date><fpage>1</fpage><lpage>4</lpage></element-citation></ref><ref id="B11-sensors-21-01734"><label>11.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Boudaya</surname><given-names>A.</given-names></name><name><surname>Bouaziz</surname><given-names>B.</given-names></name><name><surname>Chaabene</surname><given-names>S.</given-names></name><name><surname>Chaari</surname><given-names>L.</given-names></name><name><surname>Ammar</surname><given-names>A.</given-names></name><name><surname>Hökelmann</surname><given-names>A.</given-names></name></person-group><article-title>EEG-Based Hypo-vigilance Detection Using Convolutional Neural Network</article-title><source>Proceedings of the International Conference on Smart Living and Public Health (ICOST)</source><conf-loc>Hammamet, Tunisia</conf-loc><conf-date>24–26 June 2020</conf-date><fpage>69</fpage><lpage>78</lpage></element-citation></ref><ref id="B12-sensors-21-01734"><label>12.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Murugan</surname><given-names>S.</given-names></name><name><surname>Selvaraj</surname><given-names>J.</given-names></name><name><surname>Sahayadhas</surname><given-names>A.</given-names></name></person-group><article-title>Detection and analysis: Driver state with electrocardiogram (ECG)</article-title><source>Phys. Eng. Sci. Med.</source><year>2020</year><volume>43</volume><fpage>525</fpage><lpage>537</lpage><pub-id pub-id-type="doi">10.1007/s13246-020-00853-8</pub-id><pub-id pub-id-type="pmid">32524437</pub-id></element-citation></ref><ref id="B13-sensors-21-01734"><label>13.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>L.</given-names></name><name><surname>Liu</surname><given-names>F.</given-names></name><name><surname>Tang</surname><given-names>J.</given-names></name></person-group><article-title>Real-Time System for Driver Fatigue Detection by RGB-D Camera</article-title><source>Assoc. Comput. Mach.</source><year>2015</year><volume>6</volume><pub-id pub-id-type="doi">10.1145/2629482</pub-id></element-citation></ref><ref id="B14-sensors-21-01734"><label>14.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dinges</surname><given-names>D.</given-names></name></person-group><article-title>An overview of sleepiness and accidents</article-title><source>J. Sleep Res.</source><year>1995</year><volume>4</volume><fpage>4</fpage><lpage>14</lpage><pub-id pub-id-type="doi">10.1111/j.1365-2869.1995.tb00220.x</pub-id><pub-id pub-id-type="pmid">10607205</pub-id></element-citation></ref><ref id="B15-sensors-21-01734"><label>15.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Stanley</surname><given-names>P.</given-names></name><name><surname>Prahash</surname><given-names>T.</given-names></name><name><surname>Lal</surname><given-names>S.</given-names></name><name><surname>Daniel</surname><given-names>P.</given-names></name></person-group><article-title>Embedded based drowsiness detection using EEG signals</article-title><source>Proceedings of the IEEE International Conference on Power, Control, Signals and Instrumentation Engineering (ICPCSI)</source><conf-loc>Chennai, India</conf-loc><conf-date>21–22 September 2017</conf-date><fpage>2596</fpage><lpage>2600</lpage></element-citation></ref><ref id="B16-sensors-21-01734"><label>16.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gromer</surname><given-names>M.</given-names></name><name><surname>Salb</surname><given-names>D.</given-names></name><name><surname>Walzer</surname><given-names>T.</given-names></name><name><surname>Madrid</surname><given-names>N.</given-names></name><name><surname>Seepold</surname><given-names>R.</given-names></name></person-group><article-title>ECG sensor for detection of driver’s drowsiness</article-title><source>Procedia Comput. Sci.</source><year>2019</year><volume>159</volume><fpage>1938</fpage><lpage>1946</lpage><pub-id pub-id-type="doi">10.1016/j.procs.2019.09.366</pub-id></element-citation></ref><ref id="B17-sensors-21-01734"><label>17.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Choi</surname><given-names>H.</given-names></name></person-group><article-title>EMG Feature Extraction for Driver’s Drowsiness Using RF Wireless Power Transmission Method</article-title><source>Int. J. Eng. Adv. Technol. IJEAT</source><year>2019</year><volume>8</volume><fpage>494</fpage><lpage>497</lpage></element-citation></ref><ref id="B18-sensors-21-01734"><label>18.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ahn</surname><given-names>S.</given-names></name><name><surname>Nguyen</surname><given-names>T.</given-names></name><name><surname>Jang</surname><given-names>H.</given-names></name><name><surname>Kim</surname><given-names>J.</given-names></name><name><surname>Jun</surname><given-names>S.</given-names></name></person-group><article-title>Exploring neuro-physiological correlates of drivers’ mental fatigue caused by sleep deprivation using simultaneous EEG, ECG, and fNIRS data</article-title><source>Front. Hum. Neurosci.</source><year>2016</year><volume>10</volume><fpage>219</fpage><pub-id pub-id-type="doi">10.3389/fnhum.2016.00219</pub-id><pub-id pub-id-type="pmid">27242483</pub-id></element-citation></ref><ref id="B19-sensors-21-01734"><label>19.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>F.</given-names></name><name><surname>Wang</surname><given-names>H.</given-names></name><name><surname>Fu</surname><given-names>R.</given-names></name></person-group><article-title>Real-Time ECG-Based Detection of Fatigue Driving Using Sample Entropy</article-title><source>Entropy</source><year>2018</year><volume>20</volume><elocation-id>196</elocation-id><pub-id pub-id-type="doi">10.3390/e20030196</pub-id><pub-id pub-id-type="pmid">33265287</pub-id></element-citation></ref><ref id="B20-sensors-21-01734"><label>20.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sahayadhas</surname><given-names>A.</given-names></name><name><surname>Sundaraj</surname><given-names>K.</given-names></name><name><surname>Murugappan</surname><given-names>M.</given-names></name></person-group><article-title>Electromyogram signal based hypovigilance detection</article-title><source>Biomed. Res.</source><year>2014</year><volume>25</volume><fpage>281</fpage><lpage>288</lpage></element-citation></ref><ref id="B21-sensors-21-01734"><label>21.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chen</surname><given-names>D.</given-names></name><name><surname>Ma</surname><given-names>Z.</given-names></name><name><surname>Li</surname><given-names>B.</given-names></name><name><surname>Yan</surname><given-names>Z.</given-names></name><name><surname>Li</surname><given-names>W.</given-names></name></person-group><article-title>Drowsiness Detection with Electrooculography Signal Using a System Dynamics Approach</article-title><source>J. Dyn. Syst. Meas. Control</source><year>2017</year><volume>139</volume><fpage>081003</fpage><pub-id pub-id-type="doi">10.1115/1.4035611</pub-id></element-citation></ref><ref id="B22-sensors-21-01734"><label>22.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ma</surname><given-names>Y.</given-names></name><name><surname>Chen</surname><given-names>B.</given-names></name><name><surname>Li</surname><given-names>R.</given-names></name><name><surname>Wang</surname><given-names>C.</given-names></name><name><surname>Wang</surname><given-names>J.</given-names></name><name><surname>She</surname><given-names>Q.</given-names></name><name><surname>Luo</surname><given-names>Z.</given-names></name><name><surname>Zhang</surname><given-names>Y.</given-names></name></person-group><article-title>Driving Fatigue Detection from EEG Using a Modified PCANet Method</article-title><source>Comput. Intell. Neurosci.</source><year>2019</year><volume>2019</volume><fpage>4721863</fpage><pub-id pub-id-type="doi">10.1155/2019/4721863</pub-id><pub-id pub-id-type="pmid">31396270</pub-id></element-citation></ref><ref id="B23-sensors-21-01734"><label>23.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Papadelis</surname><given-names>C.</given-names></name><name><surname>Chen</surname><given-names>Z.</given-names></name><name><surname>Papadeli</surname><given-names>C.K.</given-names></name><name><surname>Bamidis</surname><given-names>P.</given-names></name><name><surname>Chouvarda</surname><given-names>I.</given-names></name><name><surname>Bekiaris</surname><given-names>E.</given-names></name><name><surname>Maglaveras</surname><given-names>N.</given-names></name></person-group><article-title>Monitoring sleepiness with on-board electrophysiological recordings for preventing sleep-deprived traffic accidents</article-title><source>Clin. Neurophysiol. Off. J. Int. Fed. Clin. Neurophysiol.</source><year>2007</year><volume>118</volume><fpage>1906</fpage><lpage>1922</lpage><pub-id pub-id-type="doi">10.1016/j.clinph.2007.04.031</pub-id><pub-id pub-id-type="pmid">17652020</pub-id></element-citation></ref><ref id="B24-sensors-21-01734"><label>24.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>LaRocco</surname><given-names>J.</given-names></name><name><surname>Le</surname><given-names>M.</given-names></name><name><surname>Paeng</surname><given-names>D.</given-names></name></person-group><article-title>A Systemic Review of Available Low-Cost EEG Headsets Used for Drowsiness Detection</article-title><source>Front. Neuroinform.</source><year>2020</year><volume>14</volume><fpage>42</fpage><pub-id pub-id-type="doi">10.3389/fninf.2020.553352</pub-id><pub-id pub-id-type="pmid">33178004</pub-id></element-citation></ref><ref id="B25-sensors-21-01734"><label>25.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Trutschel</surname><given-names>U.</given-names></name><name><surname>Sirois</surname><given-names>B.</given-names></name><name><surname>Sommer</surname><given-names>D.</given-names></name><name><surname>Golz</surname><given-names>M.</given-names></name><name><surname>Edwards</surname><given-names>D.</given-names></name></person-group><article-title>PERCLOS: An Alertness Measure of the Past</article-title><source>Proceedings of the 6th International Driving Symposium on Human Factors in Driver Assessment, Training, and Vehicle Design: Driving Assessment 2011</source><conf-loc>Lake Tahoe, CA, USA</conf-loc><conf-date>27–30 June 2011</conf-date><fpage>172</fpage><lpage>179</lpage><pub-id pub-id-type="doi">10.17077/drivingassessment.1394</pub-id></element-citation></ref><ref id="B26-sensors-21-01734"><label>26.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Duvinage</surname><given-names>M.</given-names></name><name><surname>Castermans</surname><given-names>T.</given-names></name><name><surname>Petieau</surname><given-names>M.</given-names></name><name><surname>Hoellinger</surname><given-names>T.</given-names></name><name><surname>Cheron</surname><given-names>G.</given-names></name><name><surname>Dutoit</surname><given-names>T.</given-names></name></person-group><article-title>Performance of the Emotiv Epoc headset for P300-based applications</article-title><source>BioMed. Eng. Online</source><year>2013</year><volume>12</volume><fpage>56</fpage><pub-id pub-id-type="doi">10.1186/1475-925X-12-56</pub-id><pub-id pub-id-type="pmid">23800158</pub-id></element-citation></ref><ref id="B27-sensors-21-01734"><label>27.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Abichou</surname><given-names>Y.</given-names></name><name><surname>Chaabene</surname><given-names>S.</given-names></name><name><surname>Chaari</surname><given-names>L.</given-names></name></person-group><article-title>A sleep monitoring method with EEG signals</article-title><source>Proceedings of the International Conference on Digital Health Technologies (ICDHT)</source><conf-loc>Hammamet, Tunisia</conf-loc><conf-date>9–11 December 2019</conf-date><fpage>1</fpage><lpage>8</lpage></element-citation></ref><ref id="B28-sensors-21-01734"><label>28.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Aboalayon</surname><given-names>K.</given-names></name><name><surname>Faezipour</surname><given-names>M.</given-names></name><name><surname>Almuhammadi</surname><given-names>W.</given-names></name><name><surname>Moslehpour</surname><given-names>S.</given-names></name></person-group><article-title>Sleep Stage Classification Using EEG Signal Analysis: A Comprehensive Survey and New Investigation</article-title><source>Entropy</source><year>2016</year><volume>18</volume><elocation-id>272</elocation-id><pub-id pub-id-type="doi">10.3390/e18090272</pub-id></element-citation></ref><ref id="B29-sensors-21-01734"><label>29.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Ngxande</surname><given-names>M.</given-names></name><name><surname>Tapamo</surname><given-names>J.</given-names></name><name><surname>Burke</surname><given-names>M.</given-names></name></person-group><article-title>Driver drowsiness detection using behavioral measures and machine learning techniques: A review of state-of-art techniques</article-title><source>Proceedings of the Pattern Recognition Association of South Africa and Robotics and Mechatronics (PRASA-RobMech)</source><conf-loc>Bloemfontein, South Africa</conf-loc><conf-date>30 November–1 December 2017</conf-date><fpage>156</fpage><lpage>161</lpage></element-citation></ref><ref id="B30-sensors-21-01734"><label>30.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Patil</surname><given-names>B.</given-names></name></person-group><article-title>Experimental Study on Assessment on Impact of Biometric Parameters on Drowsiness based on Yawning & head movement using Support Vector Machine</article-title><source>Int. J. Comput. Sci. Manag. Res.</source><year>2013</year><volume>2</volume><pub-id pub-id-type="doi">10.13140/RG.2.1.2416.5522</pub-id></element-citation></ref><ref id="B31-sensors-21-01734"><label>31.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Zorgui</surname><given-names>S.</given-names></name><name><surname>Chaabene</surname><given-names>S.</given-names></name><name><surname>Bouaziz</surname><given-names>B.</given-names></name><name><surname>Batatia</surname><given-names>H.</given-names></name><name><surname>Chaari</surname><given-names>L.</given-names></name></person-group><article-title>A Convolutional Neural Network for Lentigo Diagnosis</article-title><source>Proceedings of the International Conference on Smart Living and Public Health (ICOST)</source><conf-loc>Hammamet, Tunisia</conf-loc><conf-date>24–26 June 2020</conf-date><fpage>89</fpage><lpage>99</lpage></element-citation></ref><ref id="B32-sensors-21-01734"><label>32.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Zhu</surname><given-names>X.</given-names></name><name><surname>Zheng</surname><given-names>W.</given-names></name><name><surname>Lu</surname><given-names>B.</given-names></name><name><surname>Chen</surname><given-names>X.</given-names></name><name><surname>Chen</surname><given-names>S.</given-names></name><name><surname>Wang</surname><given-names>C.</given-names></name></person-group><article-title>EOG-based drowsiness detection using convolutional neural networks</article-title><source>Proceedings of the International Joint Conference on Neural Networks</source><conf-loc>Beijing, China</conf-loc><conf-date>6–11 July 2014</conf-date><fpage>128</fpage><lpage>134</lpage></element-citation></ref><ref id="B33-sensors-21-01734"><label>33.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>X.</given-names></name><name><surname>Zhao</surname><given-names>Y.</given-names></name><name><surname>Pourpanah</surname><given-names>F.</given-names></name></person-group><article-title>Recent advances in deep learning</article-title><source>Int. J. Mach. Learn. Cybern.</source><year>2020</year><volume>11</volume><fpage>747</fpage><lpage>750</lpage><pub-id pub-id-type="doi">10.1007/s13042-020-01096-5</pub-id></element-citation></ref><ref id="B34-sensors-21-01734"><label>34.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gragnaniello</surname><given-names>D.</given-names></name><name><surname>Bottino</surname><given-names>A.</given-names></name><name><surname>Cumani</surname><given-names>S.</given-names></name><name><surname>Kim</surname><given-names>W.</given-names></name></person-group><article-title>Special Issue on Advances in Deep Learning</article-title><source>Appl. Sci.</source><year>2020</year><volume>10</volume><elocation-id>3172</elocation-id><pub-id pub-id-type="doi">10.3390/app10093172</pub-id></element-citation></ref><ref id="B35-sensors-21-01734"><label>35.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Abiodun</surname><given-names>O.I.</given-names></name><name><surname>Jantan</surname><given-names>A.</given-names></name><name><surname>Omolara</surname><given-names>A.E.</given-names></name><name><surname>Dada</surname><given-names>K.V.</given-names></name><name><surname>Mohamed</surname><given-names>N.A.</given-names></name><name><surname>Arshad</surname><given-names>H.</given-names></name></person-group><article-title>State-of-the-art in artificial neural network applications: A survey</article-title><source>Heliyon</source><year>2018</year><volume>4</volume><fpage>e00938</fpage><pub-id pub-id-type="doi">10.1016/j.heliyon.2018.e00938</pub-id><pub-id pub-id-type="pmid">30519653</pub-id></element-citation></ref><ref id="B36-sensors-21-01734"><label>36.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Alom</surname><given-names>M.</given-names></name><name><surname>Taha</surname><given-names>T.</given-names></name><name><surname>Yakopcic</surname><given-names>C.</given-names></name><name><surname>Westberg</surname><given-names>S.</given-names></name><name><surname>Sidike</surname><given-names>P.</given-names></name><name><surname>Nasrin</surname><given-names>M.</given-names></name><name><surname>Hasan</surname><given-names>M.</given-names></name><name><surname>Essen</surname><given-names>B.V.</given-names></name><name><surname>Awwal</surname><given-names>A.</given-names></name><name><surname>Asari</surname><given-names>V.</given-names></name></person-group><article-title>A State-of-the-Art Survey on Deep Learning Theory and Architectures</article-title><source>Electronics</source><year>2019</year><volume>8</volume><elocation-id>292</elocation-id><pub-id pub-id-type="doi">10.3390/electronics8030292</pub-id></element-citation></ref><ref id="B37-sensors-21-01734"><label>37.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Ed-doughmi</surname><given-names>Y.</given-names></name><name><surname>Idrissi</surname><given-names>N.</given-names></name></person-group><article-title>Driver Fatigue Detection using Recurrent Neural Networks</article-title><source>Proceedings of the 2nd International Conference on Networking, Information Systems & Security</source><conf-loc>Rabat, Morocco</conf-loc><conf-date>27–28 March 2019</conf-date><year>2019</year><fpage>1</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1145/3320326.3320376</pub-id></element-citation></ref><ref id="B38-sensors-21-01734"><label>38.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Jeong</surname><given-names>J.</given-names></name><name><surname>Yu</surname><given-names>B.</given-names></name><name><surname>Lee</surname><given-names>D.</given-names></name><name><surname>Lee</surname><given-names>S.</given-names></name></person-group><article-title>Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals</article-title><source>Brain Sci.</source><year>2019</year><volume>9</volume><elocation-id>348</elocation-id><pub-id pub-id-type="doi">10.3390/brainsci9120348</pub-id></element-citation></ref><ref id="B39-sensors-21-01734"><label>39.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Vesselenyi</surname><given-names>T.</given-names></name><name><surname>Moca</surname><given-names>S.</given-names></name><name><surname>Rus</surname><given-names>A.</given-names></name><name><surname>Mitran</surname><given-names>T.</given-names></name><name><surname>Tătaru</surname><given-names>B.</given-names></name></person-group><article-title>Driver drowsiness detection using ANN image processing</article-title><source>IOP Conf. Ser. Mater. Sci. Eng.</source><year>2017</year><volume>252</volume><fpage>012097</fpage><pub-id pub-id-type="doi">10.1088/1757-899X/252/1/012097</pub-id></element-citation></ref><ref id="B40-sensors-21-01734"><label>40.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Guarda</surname><given-names>L.</given-names></name><name><surname>Astorga</surname><given-names>N.</given-names></name><name><surname>Droguett</surname><given-names>E.</given-names></name><name><surname>Moura</surname><given-names>M.</given-names></name><name><surname>Ramos</surname><given-names>M.</given-names></name></person-group><article-title>Drowsiness Detection Using Electroencephalography Signals: A Deep Learning Based Model</article-title><source>Proceedings of the Probabilistic Safety Assessment and Management PSAM</source><conf-loc>Los Angeles, CA, USA</conf-loc><conf-date>16 September 2018</conf-date></element-citation></ref><ref id="B41-sensors-21-01734"><label>41.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Deng</surname><given-names>L.</given-names></name><name><surname>He</surname><given-names>X.</given-names></name><name><surname>Gao</surname><given-names>J.</given-names></name></person-group><article-title>Deep stacking networks for information retrieval</article-title><source>Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing</source><conf-loc>Vancouver, BC, Canada</conf-loc><conf-date>26–31 May 2013</conf-date><fpage>3153</fpage><lpage>3157</lpage></element-citation></ref><ref id="B42-sensors-21-01734"><label>42.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Alaskar</surname><given-names>H.</given-names></name></person-group><article-title>Convolutional Neural Network Application in Biomedical Signals</article-title><source>J. Comput. Sci. Inf. Technol.</source><year>2018</year><volume>6</volume><fpage>45</fpage><lpage>59</lpage><pub-id pub-id-type="doi">10.15640/jcsit.v6n2a5</pub-id></element-citation></ref><ref id="B43-sensors-21-01734"><label>43.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>K.</given-names></name><name><surname>Zhao</surname><given-names>Y.</given-names></name><name><surname>Xiong</surname><given-names>Q.</given-names></name><name><surname>Fan</surname><given-names>M.</given-names></name><name><surname>Sun</surname><given-names>G.</given-names></name><name><surname>Ma</surname><given-names>L.</given-names></name><name><surname>Liu</surname><given-names>T.</given-names></name></person-group><article-title>Research on Healthy Anomaly Detection Model Based on Deep Learning from Multiple Time-Series Physiological Signals</article-title><source>Sci. Program.</source><year>2016</year><volume>2016</volume><fpage>5642856</fpage><pub-id pub-id-type="doi">10.1155/2016/5642856</pub-id></element-citation></ref><ref id="B44-sensors-21-01734"><label>44.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Piekarski</surname><given-names>M.</given-names></name><name><surname>Korjakowska</surname><given-names>J.</given-names></name><name><surname>Wawrzyniak</surname><given-names>A.</given-names></name><name><surname>Gorgon</surname><given-names>M.</given-names></name></person-group><article-title>Convolutional neural network architecture for beam instabilities identification in Synchrotron Radiation Systems as an anomaly detection problem</article-title><source>Measurement</source><year>2020</year><volume>165</volume><fpage>108116</fpage><pub-id pub-id-type="doi">10.1016/j.measurement.2020.108116</pub-id></element-citation></ref><ref id="B45-sensors-21-01734"><label>45.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chakraborty</surname><given-names>S.</given-names></name><name><surname>Aich</surname><given-names>S.</given-names></name><name><surname>Joo</surname><given-names>M.</given-names></name><name><surname>Sain</surname><given-names>M.</given-names></name><name><surname>Kim</surname><given-names>H.</given-names></name></person-group><article-title>A Multichannel Convolutional Neural Network Architecture for the Detection of the State of Mind Using Physiological Signals from Wearable Devices</article-title><source>J. Healthc. Eng.</source><year>2019</year><volume>2019</volume><fpage>5397814</fpage><pub-id pub-id-type="doi">10.1155/2019/5397814</pub-id><pub-id pub-id-type="pmid">31687119</pub-id></element-citation></ref><ref id="B46-sensors-21-01734"><label>46.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Roy</surname><given-names>Y.</given-names></name><name><surname>Banville</surname><given-names>H.</given-names></name><name><surname>Albuquerque</surname><given-names>I.</given-names></name><name><surname>Gramfort</surname><given-names>A.</given-names></name><name><surname>Falk</surname><given-names>T.</given-names></name><name><surname>Faubert</surname><given-names>J.</given-names></name></person-group><article-title>Deep learning-based electroencephalography analysis: A systematic review</article-title><source>arXiv</source><year>2019</year><pub-id pub-id-type="arxiv">1901.05498</pub-id><pub-id pub-id-type="doi">10.1088/1741-2552/ab260c</pub-id></element-citation></ref><ref id="B47-sensors-21-01734"><label>47.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Salamon</surname><given-names>J.</given-names></name><name><surname>Bello</surname><given-names>J.</given-names></name></person-group><article-title>Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification</article-title><source>arXiv</source><year>2016</year><pub-id pub-id-type="arxiv">1608.04363</pub-id><pub-id pub-id-type="doi">10.1109/LSP.2017.2657381</pub-id></element-citation></ref><ref id="B48-sensors-21-01734"><label>48.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Dwivedi</surname><given-names>K.</given-names></name><name><surname>Biswaranjan</surname><given-names>K.</given-names></name><name><surname>Sethi</surname><given-names>A.</given-names></name></person-group><article-title>Drowsy driver detection using representation learning</article-title><source>Proceedings of the IEEE International Advance Computing Conference (IACC)</source><conf-loc>Gurgaon, India</conf-loc><conf-date>21–22 February 2014</conf-date><fpage>995</fpage><lpage>999</lpage></element-citation></ref><ref id="B49-sensors-21-01734"><label>49.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Doughmi</surname><given-names>Y.</given-names></name><name><surname>Idrissi</surname><given-names>N.</given-names></name><name><surname>Hbali</surname><given-names>Y.</given-names></name></person-group><article-title>Real-Time System for Driver Fatigue Detection Based on a Recurrent Neuronal Network</article-title><source>J. Imaging</source><year>2020</year><volume>6</volume><elocation-id>8</elocation-id><pub-id pub-id-type="doi">10.3390/jimaging6030008</pub-id></element-citation></ref><ref id="B50-sensors-21-01734"><label>50.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yang</surname><given-names>Y.</given-names></name><name><surname>Gao</surname><given-names>Z.</given-names></name><name><surname>Li</surname><given-names>Y.</given-names></name><name><surname>Cai</surname><given-names>Q.</given-names></name><name><surname>Marwan</surname><given-names>N.</given-names></name><name><surname>Kurths</surname><given-names>J.</given-names></name></person-group><article-title>A Complex Network-Based Broad Learning System for Detecting Driver Fatigue From EEG Signals</article-title><source>IEEE Trans. Syst. Man Cybernet. Syst.</source><year>2019</year><fpage>1</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1109/TSMC.2019.2956022</pub-id></element-citation></ref><ref id="B51-sensors-21-01734"><label>51.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Shalash</surname><given-names>W.</given-names></name></person-group><article-title>Driver Fatigue Detection with Single EEG Channel Using Transfer Learning</article-title><source>Proceedings of the IEEE International Conference on Imaging System and Techniques</source><conf-loc>Abu Dabi, United Arab Emirates</conf-loc><conf-date>9–10 December 2019</conf-date></element-citation></ref><ref id="B52-sensors-21-01734"><label>52.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zeng</surname><given-names>H.</given-names></name><name><surname>Yang</surname><given-names>C.</given-names></name><name><surname>Dai</surname><given-names>G.</given-names></name><name><surname>Qin</surname><given-names>F.</given-names></name><name><surname>Zhang</surname><given-names>J.</given-names></name><name><surname>Kong</surname><given-names>W.</given-names></name></person-group><article-title>EEG classification of driver mental states by deep learning</article-title><source>Cogn. Neurodyn.</source><year>2018</year><volume>12</volume><fpage>597</fpage><lpage>606</lpage><pub-id pub-id-type="doi">10.1007/s11571-018-9496-y</pub-id><pub-id pub-id-type="pmid">30483367</pub-id></element-citation></ref><ref id="B53-sensors-21-01734"><label>53.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Ko</surname><given-names>W.</given-names></name><name><surname>Oh</surname><given-names>K.</given-names></name><name><surname>Jeon</surname><given-names>E.</given-names></name><name><surname>Suk</surname><given-names>H.</given-names></name></person-group><article-title>VIGNet: A Deep Convolutional Neural Network for EEG-based Driver Vigilance Estimation</article-title><source>Proceedings of the 8th International Winter Conference on Brain-Computer Interface (BCI)</source><conf-loc>Gangwon, Korea</conf-loc><conf-date>26–28 February 2020</conf-date><fpage>1</fpage><lpage>3</lpage></element-citation></ref><ref id="B54-sensors-21-01734"><label>54.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Cheng</surname><given-names>E.</given-names></name><name><surname>Young</surname><given-names>K.</given-names></name><name><surname>Lin</surname><given-names>C.</given-names></name></person-group><article-title>Image-based EEG signal processing for driving fatigue prediction</article-title><source>Proceedings of the 2018 International Automatic Control Conference (CACS)</source><conf-loc>Taoyuan, Taiwan</conf-loc><conf-date>4–7 November 2018</conf-date><fpage>1</fpage><lpage>5</lpage></element-citation></ref><ref id="B55-sensors-21-01734"><label>55.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Rahman</surname><given-names>K.</given-names></name><name><surname>Mustaffa</surname><given-names>M.</given-names></name><name><surname>Fuad</surname><given-names>N.</given-names></name><name><surname>Ahmad</surname><given-names>M.</given-names></name><name><surname>Ahad</surname><given-names>R.</given-names></name></person-group><article-title>Body Motion Control via Brain Signal Response</article-title><source>Proceedings of the IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)</source><conf-loc>Sarawak, Malaysia</conf-loc><conf-date>3–6 December 2018</conf-date><fpage>696</fpage><lpage>700</lpage></element-citation></ref><ref id="B56-sensors-21-01734"><label>56.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sánchez-Reolid</surname><given-names>R.</given-names></name><name><surname>García</surname><given-names>A.</given-names></name><name><surname>Vicente-Querol</surname><given-names>M.</given-names></name><name><surname>Fernández-Aguilar</surname><given-names>L.</given-names></name><name><surname>López</surname><given-names>M.</given-names></name><name><surname>Fernández-Caballero</surname><given-names>A.</given-names></name><name><surname>González</surname><given-names>P.</given-names></name></person-group><article-title>Artificial Neural Networks to Assess Emotional States from Brain-Computer Interface</article-title><source>Electronics</source><year>2018</year><volume>7</volume><elocation-id>384</elocation-id><pub-id pub-id-type="doi">10.3390/electronics7120384</pub-id></element-citation></ref><ref id="B57-sensors-21-01734"><label>57.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pedrosa</surname><given-names>P.</given-names></name><name><surname>Fiedler</surname><given-names>P.</given-names></name><name><surname>Schinaia</surname><given-names>L.</given-names></name><name><surname>Vasconcelos</surname><given-names>B.</given-names></name><name><surname>Martins</surname><given-names>A.</given-names></name><name><surname>Amaral</surname><given-names>M.</given-names></name><name><surname>Comani</surname><given-names>S.</given-names></name><name><surname>Haueisen</surname><given-names>J.</given-names></name><name><surname>Fonseca</surname><given-names>C.</given-names></name></person-group><article-title>Alginate-based hydrogels as an alternative to electrolytic gels for rapid EEG monitoring and easy cleaning procedures</article-title><source>Sens. Actuators B Chem.</source><year>2017</year><volume>247</volume><fpage>231</fpage><lpage>237</lpage><pub-id pub-id-type="doi">10.1016/j.snb.2017.02.164</pub-id></element-citation></ref><ref id="B58-sensors-21-01734"><label>58.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Towle</surname><given-names>V.</given-names></name><name><surname>Bolafios</surname><given-names>J.</given-names></name><name><surname>Suarez</surname><given-names>D.</given-names></name><name><surname>Tan</surname><given-names>K.</given-names></name><name><surname>Grzeszczuk</surname><given-names>R.</given-names></name><name><surname>Levin</surname><given-names>D.</given-names></name><name><surname>Cakmur</surname><given-names>R.</given-names></name><name><surname>Frank</surname><given-names>S.</given-names></name><name><surname>Spire</surname><given-names>J.</given-names></name></person-group><article-title>The spatial location of EEG electrodes: Locating the best-fitting sphere relative to cortical anatomy</article-title><source>Electroencephalogr. Clin. Neurophysiol.</source><year>1993</year><volume>86</volume><fpage>1</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1016/0013-4694(93)90061-Y</pub-id><pub-id pub-id-type="pmid">7678386</pub-id></element-citation></ref><ref id="B59-sensors-21-01734"><label>59.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hu</surname><given-names>S.</given-names></name><name><surname>Zheng</surname><given-names>G.</given-names></name><name><surname>Peters</surname><given-names>B.</given-names></name></person-group><article-title>Driver fatigue detection from electroencephalogram spectrum after electrooculography artefact removal</article-title><source>Intell. Transp. Syst. IET</source><year>2013</year><volume>7</volume><fpage>105</fpage><lpage>113</lpage><pub-id pub-id-type="doi">10.1049/iet-its.2012.0045</pub-id></element-citation></ref><ref id="B60-sensors-21-01734"><label>60.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Mohammedi</surname><given-names>M.</given-names></name><name><surname>Omar</surname><given-names>M.</given-names></name><name><surname>Bouabdallah</surname><given-names>A.</given-names></name></person-group><article-title>Automatic removal of ocular artifacts in EEG signals for driver’s drowsiness detection: A survey</article-title><source>Proceedings of the 2018 International Conference on Smart Communications in Network Technologies (SaCoNeT)</source><conf-loc>El Oued, Algeria</conf-loc><conf-date>27–31 October 2018</conf-date><fpage>188</fpage><lpage>193</lpage></element-citation></ref><ref id="B61-sensors-21-01734"><label>61.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gebodh</surname><given-names>N.</given-names></name><name><surname>Esmaeilpour</surname><given-names>Z.</given-names></name><name><surname>Adair</surname><given-names>D.</given-names></name><name><surname>Chelette</surname><given-names>K.</given-names></name><name><surname>Dmochowski</surname><given-names>J.</given-names></name><name><surname>Woods</surname><given-names>A.</given-names></name><name><surname>Kappenman</surname><given-names>E.</given-names></name><name><surname>Parra</surname><given-names>L.</given-names></name><name><surname>Bikson</surname><given-names>M.</given-names></name></person-group><article-title>Inherent physiological artifacts in EEG during tDCS</article-title><source>Neuroimage</source><year>2018</year><volume>185</volume><fpage>408</fpage><lpage>424</lpage><pub-id pub-id-type="doi">10.1016/j.neuroimage.2018.10.025</pub-id><pub-id pub-id-type="pmid">30321643</pub-id></element-citation></ref><ref id="B62-sensors-21-01734"><label>62.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Laruelo</surname><given-names>A.</given-names></name><name><surname>Chaari</surname><given-names>L.</given-names></name><name><surname>Batatia</surname><given-names>H.</given-names></name><name><surname>Ken</surname><given-names>S.</given-names></name><name><surname>Rowland</surname><given-names>B.</given-names></name><name><surname>Tourneret</surname><given-names>J.Y.</given-names></name><name><surname>Laprie</surname><given-names>A.</given-names></name></person-group><article-title>Hybrid Sparse Regularization for Magnetic Resonance Spectroscopy</article-title><source>Proceedings of the IEEE International Conference of Engineering in Medicine and Biology Society (EMBC)</source><conf-loc>Osaka, Japan</conf-loc><conf-date>3–7 July 2013</conf-date><fpage>3</fpage><lpage>7</lpage></element-citation></ref><ref id="B63-sensors-21-01734"><label>63.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Chaari</surname><given-names>L.</given-names></name><name><surname>Tourneret</surname><given-names>J.Y.</given-names></name><name><surname>Chaux</surname><given-names>C.</given-names></name></person-group><article-title>Sparse signal recovery using a Bernouilli generalized Gaussian prior</article-title><source>Proceedings of the European Signal Processing Conference (EUSIPCO)</source><conf-loc>Nice, France</conf-loc><conf-date>31 August–4 September 2015</conf-date></element-citation></ref><ref id="B64-sensors-21-01734"><label>64.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sevgi</surname><given-names>L.</given-names></name></person-group><article-title>Numerical Fourier Transforms: DFT and FFT</article-title><source>IEEE Antennas Propag. Mag.</source><year>2007</year><volume>49</volume><fpage>238</fpage><lpage>243</lpage><pub-id pub-id-type="doi">10.1109/MAP.2007.4293982</pub-id></element-citation></ref><ref id="B65-sensors-21-01734"><label>65.</label><element-citation publication-type="web"><person-group person-group-type="author"><name><surname>Ludwig</surname><given-names>P.</given-names></name><name><surname>Varacallo</surname><given-names>M.</given-names></name></person-group><article-title>Neuroanatomy, Central Nervous System (CNS)</article-title><comment>Available online: <ext-link ext-link-type="uri" xlink:href="https://www.researchgate.net/publication/329717386_Neuroanatomy_Central_Nervous_System_CNS">https://www.researchgate.net/publication/329717386_Neuroanatomy_Central_Nervous_System_CNS</ext-link></comment><date-in-citation content-type="access-date" iso-8601-date="2021-01-15">(accessed on 15 January 2021)</date-in-citation></element-citation></ref><ref id="B66-sensors-21-01734"><label>66.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Teplan</surname><given-names>M.</given-names></name></person-group><article-title>Fundamental of EEG Measurement</article-title><source>Meas. Sci. Rev.</source><year>2002</year><volume>2</volume><fpage>1</fpage><lpage>11</lpage></element-citation></ref><ref id="B67-sensors-21-01734"><label>67.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Al-Kadi</surname><given-names>M.I.</given-names></name><name><surname>Reaz</surname><given-names>M.B.I.</given-names></name><name><surname>Ali</surname><given-names>M.A.</given-names></name></person-group><article-title>Evolution of Electroencephalogram Signal Analysis Techniques during Anesthesia</article-title><source>Sensors</source><year>2013</year><volume>13</volume><fpage>6605</fpage><lpage>6635</lpage><pub-id pub-id-type="doi">10.3390/s130506605</pub-id><pub-id pub-id-type="pmid">23686141</pub-id></element-citation></ref><ref id="B68-sensors-21-01734"><label>68.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Schütze</surname><given-names>M.D.</given-names></name><name><surname>Junghanns</surname><given-names>K.</given-names></name></person-group><article-title>The Difficulty of Staying Awake During Alpha/Theta Neurofeedback Training</article-title><source>Appl. Psychophysiol. Biofeedback</source><year>2015</year><volume>40</volume><fpage>85</fpage><lpage>94</lpage><pub-id pub-id-type="doi">10.1007/s10484-015-9278-9</pub-id><pub-id pub-id-type="pmid">25835580</pub-id></element-citation></ref><ref id="B69-sensors-21-01734"><label>69.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Amo</surname><given-names>C.</given-names></name><name><surname>de Santiago</surname><given-names>L.</given-names></name><name><surname>Barea</surname><given-names>R.</given-names></name><name><surname>López-Dorado</surname><given-names>A.</given-names></name><name><surname>Boquete</surname><given-names>L.</given-names></name></person-group><article-title>Analysis of Gamma-Band Activity from Human EEG Using Empirical Mode Decomposition</article-title><source>Sensors</source><year>2017</year><volume>17</volume><fpage>989</fpage></element-citation></ref><ref id="B70-sensors-21-01734"><label>70.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dkhil</surname><given-names>M.B.</given-names></name><name><surname>Wali</surname><given-names>A.</given-names></name><name><surname>Alimi</surname><given-names>A.M.</given-names></name></person-group><article-title>Drowsy Driver Detection by EEG Analysis Using Fast Fourier Transform</article-title><source>arXiv</source><year>2018</year><pub-id pub-id-type="arxiv">1806.07286v1</pub-id></element-citation></ref><ref id="B71-sensors-21-01734"><label>71.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ogino</surname><given-names>M.</given-names></name><name><surname>Mitsukura</surname><given-names>Y.</given-names></name></person-group><article-title>Portable Drowsiness Detection through Use of a Prefrontal Single-Channel Electroencephalogram</article-title><source>Sensors</source><year>2018</year><volume>18</volume><elocation-id>4477</elocation-id><pub-id pub-id-type="doi">10.3390/s18124477</pub-id><pub-id pub-id-type="pmid">30567347</pub-id></element-citation></ref><ref id="B72-sensors-21-01734"><label>72.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lin</surname><given-names>C.T.</given-names></name><name><surname>Wu</surname><given-names>R.C.</given-names></name><name><surname>Liang</surname><given-names>S.</given-names></name><name><surname>Chao</surname><given-names>W.H.</given-names></name><name><surname>Chen</surname><given-names>Y.J.</given-names></name><name><surname>Jung</surname><given-names>T.P.</given-names></name></person-group><article-title>EEG-based drowsiness estimation for safety driving using independent component analysis</article-title><source>IEEE Trans. Circuits Syst. I Regul. Pap.</source><year>2005</year><volume>52</volume><fpage>2726</fpage><lpage>2738</lpage></element-citation></ref><ref id="B73-sensors-21-01734"><label>73.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Makeig</surname><given-names>S.</given-names></name><name><surname>Jung</surname><given-names>T.</given-names></name><name><surname>Sejnowski</surname><given-names>T.</given-names></name></person-group><article-title>Awareness during drowsiness: Dynamics and electrophysiological correlates</article-title><source>Can. J. Exp. Psychol.</source><year>2000</year><volume>54</volume><fpage>266</fpage><lpage>273</lpage><pub-id pub-id-type="doi">10.1037/h0087346</pub-id><pub-id pub-id-type="pmid">11195717</pub-id></element-citation></ref><ref id="B74-sensors-21-01734"><label>74.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Subasi</surname><given-names>A.</given-names></name></person-group><article-title>Automatic recognition of alertness level from EEG by using neural network and wavelet coefficients</article-title><source>Expert Syst. Appl.</source><year>2005</year><volume>28</volume><fpage>701</fpage><lpage>711</lpage><pub-id pub-id-type="doi">10.1016/j.eswa.2004.12.027</pub-id></element-citation></ref><ref id="B75-sensors-21-01734"><label>75.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kar</surname><given-names>S.</given-names></name><name><surname>Bhagat</surname><given-names>M.</given-names></name><name><surname>Routray</surname><given-names>A.</given-names></name></person-group><article-title>EEG signal analysis for the assessment and quantification of driver’s fatigue</article-title><source>Transp. Res. Part F Traffic Psychol. Behav.</source><year>2010</year><volume>13</volume><fpage>297</fpage><lpage>306</lpage><pub-id pub-id-type="doi">10.1016/j.trf.2010.06.006</pub-id></element-citation></ref><ref id="B76-sensors-21-01734"><label>76.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bernardi</surname><given-names>G.</given-names></name><name><surname>Betta</surname><given-names>M.</given-names></name><name><surname>Ricciardi</surname><given-names>E.</given-names></name><name><surname>Pietrini</surname><given-names>P.</given-names></name><name><surname>Tononi</surname><given-names>G.</given-names></name><name><surname>Siclari</surname><given-names>F.</given-names></name></person-group><article-title>Regional Delta Waves In Human Rapid Eye Movement Sleep</article-title><source>J. Neurosci.</source><year>2019</year><volume>39</volume><fpage>2686</fpage><lpage>2697</lpage><pub-id pub-id-type="doi">10.1523/JNEUROSCI.2298-18.2019</pub-id><pub-id pub-id-type="pmid">30737310</pub-id></element-citation></ref><ref id="B77-sensors-21-01734"><label>77.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lashgari</surname><given-names>E.</given-names></name><name><surname>Liang</surname><given-names>D.</given-names></name><name><surname>Maoz</surname><given-names>U.</given-names></name></person-group><article-title>Data Augmentation for Deep-Learning-Based Electroencephalography</article-title><source>J. Neurosci. Methods</source><year>2020</year><volume>346</volume><fpage>108885</fpage><pub-id pub-id-type="doi">10.1016/j.jneumeth.2020.108885</pub-id><pub-id pub-id-type="pmid">32745492</pub-id></element-citation></ref><ref id="B78-sensors-21-01734"><label>78.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>Z.</given-names></name><name><surname>Casals</surname><given-names>J.</given-names></name><name><surname>Cichocki</surname><given-names>A.</given-names></name></person-group><article-title>A Novel Deep Learning Approach with Data Augmentation to Classify Motor Imagery Signals</article-title><source>IEEE Access</source><year>2019</year><volume>7</volume><fpage>5945</fpage><lpage>15954</lpage><pub-id pub-id-type="doi">10.1109/ACCESS.2019.2895133</pub-id></element-citation></ref><ref id="B79-sensors-21-01734"><label>79.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>Y.</given-names></name><name><surname>Yang</surname><given-names>S.</given-names></name><name><surname>Liu</surname><given-names>Y.</given-names></name><name><surname>Zhang</surname><given-names>Y.</given-names></name><name><surname>Han</surname><given-names>B.</given-names></name><name><surname>Zhou</surname><given-names>F.</given-names></name></person-group><article-title>Integration of 24 Feature Types to Accurately Detect and Predict Seizures Using Scalp EEG Signals</article-title><source>Sensors</source><year>2018</year><volume>18</volume><elocation-id>1372</elocation-id><pub-id pub-id-type="doi">10.3390/s18051372</pub-id></element-citation></ref><ref id="B80-sensors-21-01734"><label>80.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chawla</surname><given-names>N.V.</given-names></name><name><surname>Bowyer</surname><given-names>K.W.</given-names></name><name><surname>Hall</surname><given-names>L.O.</given-names></name><name><surname>Kegelmeyer</surname><given-names>W.P.</given-names></name></person-group><article-title>SMOTE: Synthetic Minority Over-sampling Technique</article-title><source>J. Artif. Intell. Res.</source><year>2002</year><volume>16</volume><fpage>321</fpage><lpage>357</lpage><pub-id pub-id-type="doi">10.1613/jair.953</pub-id></element-citation></ref><ref id="B81-sensors-21-01734"><label>81.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Garcia</surname><given-names>A.</given-names></name><name><surname>Peter</surname><given-names>K.</given-names></name></person-group><article-title>Further advantages of data augmentation on convolutional neural networks</article-title><source>Proceedings of the 27th International Conference on Artificial Neural Networks</source><conf-loc>Rhodes, Greece</conf-loc><conf-date>27 September 2018</conf-date><fpage>95</fpage><lpage>103</lpage></element-citation></ref><ref id="B82-sensors-21-01734"><label>82.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Srivastava</surname><given-names>N.</given-names></name><name><surname>Hinton</surname><given-names>G.</given-names></name><name><surname>Krizhevsky</surname><given-names>A.</given-names></name><name><surname>Sutskever</surname><given-names>I.</given-names></name><name><surname>Salakhutdinov</surname><given-names>R.</given-names></name></person-group><article-title>Dropout: A Simple Way to Prevent Neural Networks from Overfitting</article-title><source>J. Mach. Learn. Res.</source><year>2014</year><volume>15</volume><fpage>1929</fpage><lpage>1958</lpage></element-citation></ref><ref id="B83-sensors-21-01734"><label>83.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hammad</surname><given-names>M.</given-names></name><name><surname>Pławiak</surname><given-names>P.</given-names></name><name><surname>Wang</surname><given-names>K.</given-names></name><name><surname>Acharya</surname><given-names>U.R.</given-names></name></person-group><article-title>ResNet-Attention model for human authentication using ECG signals</article-title><source>Expert Syst.</source><year>2020</year><volume>1</volume><fpage>e12547</fpage><pub-id pub-id-type="doi">10.1111/exsy.12547</pub-id></element-citation></ref><ref id="B84-sensors-21-01734"><label>84.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Oh</surname><given-names>S.</given-names></name><name><surname>Jahmunah</surname><given-names>V.</given-names></name><name><surname>Ooi</surname><given-names>C.</given-names></name><name><surname>Tan</surname><given-names>R.</given-names></name><name><surname>Ciaccio</surname><given-names>E.</given-names></name><name><surname>Yamakawa</surname><given-names>T.</given-names></name><name><surname>Tanabe</surname><given-names>M.</given-names></name><name><surname>Kobayashi</surname><given-names>M.</given-names></name><name><surname>Acharya</surname><given-names>U.</given-names></name></person-group><article-title>Classification of heart sound signals using a novel deep WaveNet model</article-title><source>Comput. Methods Programs Biomed.</source><year>2020</year><volume>196</volume><fpage>105604</fpage><pub-id pub-id-type="doi">10.1016/j.cmpb.2020.105604</pub-id><pub-id pub-id-type="pmid">32593061</pub-id></element-citation></ref><ref id="B85-sensors-21-01734"><label>85.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kim</surname><given-names>J.</given-names></name><name><surname>Seo</surname><given-names>S.</given-names></name><name><surname>Song</surname><given-names>C.</given-names></name><name><surname>Kim</surname><given-names>K.</given-names></name></person-group><article-title>Assessment of Electrocardiogram Rhythms by GoogLeNet Deep Neural Network Architecture</article-title><source>J. Healthc. Eng.</source><year>2019</year><volume>2019</volume><fpage>2826901</fpage><pub-id pub-id-type="doi">10.1155/2019/2826901</pub-id><pub-id pub-id-type="pmid">31183029</pub-id></element-citation></ref><ref id="B86-sensors-21-01734"><label>86.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Park</surname><given-names>J.</given-names></name><name><surname>Kim</surname><given-names>J.</given-names></name><name><surname>Jung</surname><given-names>S.</given-names></name><name><surname>Gil</surname><given-names>Y.</given-names></name><name><surname>Choi</surname><given-names>J.</given-names></name><name><surname>Son</surname><given-names>H.</given-names></name></person-group><article-title>ECG-Signal Multi-Classification Model Based on Squeeze-and-Excitation Residual Neural Networks</article-title><source>Appl. Sci.</source><year>2020</year><volume>10</volume><elocation-id>6495</elocation-id><pub-id pub-id-type="doi">10.3390/app10186495</pub-id></element-citation></ref><ref id="B87-sensors-21-01734"><label>87.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>S.</given-names></name><name><surname>Wang</surname><given-names>S.</given-names></name><name><surname>Wang</surname><given-names>Y.</given-names></name><name><surname>Wang</surname><given-names>S.</given-names></name></person-group><article-title>Time-ResNeXt for epilepsy recognition based on EEG signals in wireless networks</article-title><source>EURASIP J. Wireless Commun. Netw.</source><year>2020</year><volume>2020</volume><fpage>195</fpage><pub-id pub-id-type="doi">10.1186/s13638-020-01810-5</pub-id></element-citation></ref><ref id="B88-sensors-21-01734"><label>88.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Uyulan</surname><given-names>C.</given-names></name><name><surname>Ergüzel</surname><given-names>T.</given-names></name><name><surname>Unubol</surname><given-names>H.</given-names></name><name><surname>Cebi</surname><given-names>M.</given-names></name><name><surname>Sayar</surname><given-names>G.</given-names></name><name><surname>Asad</surname><given-names>M.</given-names></name><name><surname>Tarhan</surname><given-names>N.</given-names></name></person-group><article-title>Major Depressive Disorder Classification Based on Different Convolutional Neural Network Models: Deep Learning Approach</article-title><source>Clin. EEG Neurosci.</source><year>2020</year><pub-id pub-id-type="doi">10.1177/1550059420916634</pub-id><pub-id pub-id-type="pmid">32491928</pub-id></element-citation></ref><ref id="B89-sensors-21-01734"><label>89.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hasan</surname><given-names>M.</given-names></name><name><surname>Shon</surname><given-names>D.</given-names></name><name><surname>Im</surname><given-names>K.</given-names></name><name><surname>Choi</surname><given-names>H.</given-names></name><name><surname>Yoo</surname><given-names>D.</given-names></name><name><surname>Kim</surname><given-names>J.</given-names></name></person-group><article-title>Sleep State Classification Using Power Spectral Density and Residual Neural Network with Multichannel EEG Signals</article-title><source>Appl. Sci.</source><year>2020</year><volume>10</volume><elocation-id>7639</elocation-id><pub-id pub-id-type="doi">10.3390/app10217639</pub-id></element-citation></ref><ref id="B90-sensors-21-01734"><label>90.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Nahid</surname><given-names>N.</given-names></name><name><surname>Rahman</surname><given-names>A.</given-names></name><name><surname>Ahad</surname><given-names>M.A.R.</given-names></name></person-group><article-title>Deep Learning Based Surface EMG Hand Gesture Classification for Low-Cost Myoelectric Prosthetic Hand</article-title><source>Proceedings of the 9th International Conference on Informatics, Electronics & Vision (ICIEV) and 2020 4th International Conference on Imaging, Vision & Pattern Recognition (icIVPR)</source><conf-loc>Kitakyushu, Japan</conf-loc><conf-date>26–29 August 2020</conf-date></element-citation></ref><ref id="B91-sensors-21-01734"><label>91.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wulan</surname><given-names>N.</given-names></name><name><surname>Wang</surname><given-names>W.</given-names></name><name><surname>Sun</surname><given-names>P.</given-names></name><name><surname>Wang</surname><given-names>K.</given-names></name><name><surname>Xia</surname><given-names>Y.</given-names></name><name><surname>Zhang</surname><given-names>H.</given-names></name></person-group><article-title>Generating electrocardiogram signals by deep learning</article-title><source>Neurocomputing</source><year>2020</year><volume>404</volume><fpage>122</fpage><lpage>136</lpage><pub-id pub-id-type="doi">10.1016/j.neucom.2020.04.076</pub-id></element-citation></ref><ref id="B92-sensors-21-01734"><label>92.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhu</surname><given-names>T.</given-names></name><name><surname>Luo</surname><given-names>W.</given-names></name><name><surname>Yu</surname><given-names>F.</given-names></name></person-group><article-title>Convolution- and Attention-Based Neural Network for Automated Sleep Stage Classification</article-title><source>Int. J. Environ. Res. Public Health</source><year>2020</year><volume>17</volume><elocation-id>4152</elocation-id><pub-id pub-id-type="doi">10.3390/ijerph17114152</pub-id><pub-id pub-id-type="pmid">32532084</pub-id></element-citation></ref><ref id="B93-sensors-21-01734"><label>93.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cao</surname><given-names>C.</given-names></name><name><surname>Liu</surname><given-names>F.</given-names></name><name><surname>Tan</surname><given-names>H.</given-names></name><name><surname>Song</surname><given-names>D.</given-names></name><name><surname>Shu</surname><given-names>W.</given-names></name><name><surname>Li</surname><given-names>W.</given-names></name><name><surname>Zhou</surname><given-names>Y.</given-names></name><name><surname>Bo</surname><given-names>X.</given-names></name><name><surname>Xie</surname><given-names>Z.</given-names></name></person-group><article-title>Deep Learning and Its Applications in Biomedicine</article-title><source>Genom. Proteom. Bioinform.</source><year>2018</year><volume>16</volume><fpage>17</fpage><lpage>32</lpage><pub-id pub-id-type="doi">10.1016/j.gpb.2017.07.003</pub-id><pub-id pub-id-type="pmid">29522900</pub-id></element-citation></ref><ref id="B94-sensors-21-01734"><label>94.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rehman</surname><given-names>M.</given-names></name><name><surname>Waris</surname><given-names>A.</given-names></name><name><surname>Gilani</surname><given-names>S.</given-names></name><name><surname>Jochumsen</surname><given-names>M.</given-names></name><name><surname>Niazi</surname><given-names>I.K.</given-names></name><name><surname>Jamil</surname><given-names>M.</given-names></name><name><surname>Farina</surname><given-names>D.</given-names></name><name><surname>Kamavuako</surname><given-names>E.</given-names></name></person-group><article-title>Multiday EMG-Based Classification of Hand Motions with Deep Learning Techniques</article-title><source>Sensors</source><year>2018</year><volume>18</volume><elocation-id>2497</elocation-id><pub-id pub-id-type="doi">10.3390/s18082497</pub-id><pub-id pub-id-type="pmid">30071617</pub-id></element-citation></ref><ref id="B95-sensors-21-01734"><label>95.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hu</surname><given-names>J.</given-names></name></person-group><article-title>Comparison of Different Features and Classifiers for Driver Fatigue Detection Based on a Single EEG Channel</article-title><source>Comput. Math. Methods Med.</source><year>2017</year><volume>2017</volume><fpage>5109530</fpage><pub-id pub-id-type="doi">10.1155/2017/5109530</pub-id><pub-id pub-id-type="pmid">28255330</pub-id></element-citation></ref><ref id="B96-sensors-21-01734"><label>96.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Morales</surname><given-names>J.</given-names></name><name><surname>Saldaña</surname><given-names>R.</given-names></name><name><surname>Castolo</surname><given-names>M.</given-names></name><name><surname>Borrayo</surname><given-names>C.C.E.</given-names></name><name><surname>Ruiz</surname><given-names>A.</given-names></name><name><surname>Perez</surname><given-names>H.</given-names></name><name><surname>Ruiz</surname><given-names>G.</given-names></name></person-group><article-title>Deep Learning for the Classification of Genomic Signals</article-title><source>Comput. Intell. Image Process.</source><year>2020</year><volume>2020</volume><fpage>7698590</fpage></element-citation></ref><ref id="B97-sensors-21-01734"><label>97.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Srinivasan</surname><given-names>V.</given-names></name><name><surname>Islam</surname><given-names>M.</given-names></name><name><surname>Zhang</surname><given-names>W.</given-names></name><name><surname>Ren</surname><given-names>H.</given-names></name></person-group><article-title>Finger Movement Classification from Myoelectric Signals Using Convolutional Neural Networks</article-title><source>Proceedings of the IEEE International Conference on Robotics and Biomimetics (ROBIO)</source><conf-loc>Kuala Lumpur, Malaysia</conf-loc><conf-date>12–15 December 2018</conf-date><fpage>1070</fpage><lpage>1075</lpage></element-citation></ref><ref id="B98-sensors-21-01734"><label>98.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Choi</surname><given-names>I.</given-names></name><name><surname>Kim</surname><given-names>H.</given-names></name></person-group><article-title>An On-Device Deep Learning Approach to Battery Saving on Industrial Mobile Terminals</article-title><source>Sensors</source><year>2020</year><volume>20</volume><elocation-id>4044</elocation-id><pub-id pub-id-type="doi">10.3390/s20144044</pub-id></element-citation></ref><ref id="B99-sensors-21-01734"><label>99.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Buslaev</surname><given-names>A.</given-names></name><name><surname>Iglovikov</surname><given-names>V.</given-names></name><name><surname>Khvedchenya</surname><given-names>E.</given-names></name><name><surname>Parinov</surname><given-names>A.</given-names></name><name><surname>Druzhinin</surname><given-names>M.</given-names></name><name><surname>Kalinin</surname><given-names>A.</given-names></name></person-group><article-title>Albumentations: Fast and flexible image augmentations</article-title><source>Information</source><year>2020</year><volume>11</volume><elocation-id>125</elocation-id><pub-id pub-id-type="doi">10.3390/info11020125</pub-id></element-citation></ref><ref id="B100-sensors-21-01734"><label>100.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Somrak</surname><given-names>M.</given-names></name><name><surname>Džeroski</surname><given-names>S.</given-names></name><name><surname>Kokalj</surname><given-names>T.</given-names></name></person-group><article-title>Learning to Classify Structures in ALS-Derived Visualizations of Ancient Maya Settlements with CNN</article-title><source>Remote Sens.</source><year>2020</year><volume>12</volume><elocation-id>2215</elocation-id><pub-id pub-id-type="doi">10.3390/rs12142215</pub-id></element-citation></ref><ref id="B101-sensors-21-01734"><label>101.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Abadi</surname><given-names>M.</given-names></name><name><surname>Barham</surname><given-names>P.</given-names></name><name><surname>Chen</surname><given-names>J.</given-names></name><name><surname>Chen</surname><given-names>Z.</given-names></name><name><surname>Davis</surname><given-names>A.</given-names></name><name><surname>Dean</surname><given-names>J.</given-names></name><name><surname>Devin</surname><given-names>M.</given-names></name><name><surname>Ghemawat</surname><given-names>S.</given-names></name><name><surname>Irving</surname><given-names>G.</given-names></name><name><surname>Isard</surname><given-names>M.</given-names></name><etal></etal></person-group><article-title>TensorFlow: A System for Large-Scale Machine Learning</article-title><source>Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation</source><conf-loc>Savannah, GA, USA</conf-loc><conf-date>2–4 November 2016</conf-date><fpage>265</fpage><lpage>283</lpage></element-citation></ref><ref id="B102-sensors-21-01734"><label>102.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Paszke</surname><given-names>A.</given-names></name><name><surname>Gross</surname><given-names>S.</given-names></name><name><surname>Massa</surname><given-names>F.</given-names></name><name><surname>Lerer</surname><given-names>A.</given-names></name><name><surname>Bradbury</surname><given-names>J.</given-names></name><name><surname>Chanan</surname><given-names>G.</given-names></name><name><surname>Killeen</surname><given-names>T.</given-names></name><name><surname>Lin</surname><given-names>Z.</given-names></name><name><surname>Gimelshein</surname><given-names>N.</given-names></name><name><surname>Antiga</surname><given-names>L.</given-names></name><etal></etal></person-group><article-title>PyTorch: An Imperative Style, High-Performance Deep Learning Library</article-title><source>Adv. Neural Inf. Process. Syst.</source><year>2019</year><volume>32</volume><fpage>8026</fpage><lpage>8037</lpage></element-citation></ref><ref id="B103-sensors-21-01734"><label>103.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Predescu</surname><given-names>A.</given-names></name><name><surname>Truica</surname><given-names>C.</given-names></name><name><surname>Apostol</surname><given-names>E.</given-names></name><name><surname>Mocanu</surname><given-names>M.</given-names></name><name><surname>Lupu</surname><given-names>C.</given-names></name></person-group><article-title>An Advanced Learning-Based Multiple Model Control Supervisor for Pumping Stations in a Smart Water Distribution System</article-title><source>Mathematics</source><year>2020</year><volume>8</volume><elocation-id>887</elocation-id><pub-id pub-id-type="doi">10.3390/math8060887</pub-id></element-citation></ref><ref id="B104-sensors-21-01734"><label>104.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Saleem</surname><given-names>M.H.</given-names></name><name><surname>Potgieter</surname><given-names>J.</given-names></name><name><surname>Arif</surname><given-names>K.</given-names></name></person-group><article-title>Plant Disease Classification: A Comparative Evaluation of Convolutional Neural Networks and Deep Learning Optimizers</article-title><source>Plants</source><year>2020</year><volume>9</volume><elocation-id>1319</elocation-id><pub-id pub-id-type="doi">10.3390/plants9101319</pub-id></element-citation></ref><ref id="B105-sensors-21-01734"><label>105.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>G.</given-names></name><name><surname>Lee</surname><given-names>C.</given-names></name><name><surname>Jung</surname><given-names>J.</given-names></name><name><surname>Youn</surname><given-names>Y.</given-names></name><name><surname>Camacho</surname><given-names>D.</given-names></name></person-group><article-title>Deep learning for EEG data analytics: A survey</article-title><source>Concurr. Comput. Pract. Exp.</source><year>2019</year><volume>32</volume><fpage>e5199</fpage><pub-id pub-id-type="doi">10.1002/cpe.5199</pub-id></element-citation></ref><ref id="B106-sensors-21-01734"><label>106.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Shaf</surname><given-names>A.</given-names></name><name><surname>Ali</surname><given-names>T.</given-names></name><name><surname>Farooq</surname><given-names>W.</given-names></name><name><surname>Javaid</surname><given-names>S.</given-names></name><name><surname>Draz</surname><given-names>U.</given-names></name><name><surname>Yasin</surname><given-names>S.</given-names></name></person-group><article-title>Two Classes Classification Using Different Optimizers in Convolutional Neural Network</article-title><source>Proceedings of the 2018 IEEE 21st International Multi-Topic Conference (INMIC)</source><conf-loc>Karachi, Pakistan</conf-loc><conf-date>1–2 November 2018</conf-date><fpage>1</fpage><lpage>6</lpage></element-citation></ref><ref id="B107-sensors-21-01734"><label>107.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Tafsast</surname><given-names>A.</given-names></name><name><surname>Ferroudji</surname><given-names>K.</given-names></name><name><surname>Hadjili</surname><given-names>M.</given-names></name><name><surname>Bouakaz</surname><given-names>A.</given-names></name><name><surname>Benoudjit</surname><given-names>N.</given-names></name></person-group><article-title>Automatic Microemboli Characterization Using Convolutional Neural Networks and Radio Frequency Signals</article-title><source>Proceedings of the International Conference on Communications and Electrical Engineering (ICCEE)</source><conf-loc>El Oued, Algeria</conf-loc><conf-date>17–18 December 2018</conf-date><fpage>1</fpage><lpage>4</lpage></element-citation></ref><ref id="B108-sensors-21-01734"><label>108.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Jogin</surname><given-names>M.</given-names></name><name><surname>Madhulika</surname><given-names>M.S.</given-names></name><name><surname>Divya</surname><given-names>G.D.</given-names></name><name><surname>Meghana</surname><given-names>R.K.</given-names></name><name><surname>Apoorva</surname><given-names>S.</given-names></name></person-group><article-title>Feature Extraction using Convolution Neural Networks (CNN) and Deep Learning</article-title><source>Proceedings of the 2018 3rd IEEE International Conference on Recent Trends in Electronics, Information Communication Technology (RTEICT)</source><conf-loc>Bangalore, India</conf-loc><conf-date>18–19 May 2018</conf-date><fpage>2319</fpage><lpage>2323</lpage></element-citation></ref><ref id="B109-sensors-21-01734"><label>109.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Garbin</surname><given-names>C.</given-names></name><name><surname>Zhu</surname><given-names>X.</given-names></name><name><surname>Marques</surname><given-names>O.</given-names></name></person-group><article-title>Dropout vs. batch normalization: An empirical study of their impact to deep learning</article-title><source>Multimed. Tools Appl.</source><year>2020</year><volume>79</volume><fpage>1</fpage><lpage>39</lpage><pub-id pub-id-type="doi">10.1007/s11042-019-08453-9</pub-id></element-citation></ref><ref id="B110-sensors-21-01734"><label>110.</label><element-citation publication-type="book"><person-group person-group-type="author"><name><surname>Bisong</surname><given-names>E.</given-names></name></person-group><source>Building Machine Learning and Deep Learning Models on Google Cloud Platform</source><publisher-name>Springer</publisher-name><publisher-loc>Berlin/Heidelberg, Germany</publisher-loc><year>2019</year><fpage>59</fpage><lpage>64</lpage></element-citation></ref><ref id="B111-sensors-21-01734"><label>111.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sarno</surname><given-names>R.</given-names></name><name><surname>Nugraha</surname><given-names>B.</given-names></name><name><surname>Munawar</surname><given-names>M.</given-names></name></person-group><article-title>Real Time Fatigue-Driver Detection from Electroencephalography Using Emotiv EPOC+</article-title><source>Int. Rev. Comput. Softw. IRECOS</source><year>2016</year><volume>11</volume><fpage>214</fpage><pub-id pub-id-type="doi">10.15866/irecos.v11i3.8562</pub-id></element-citation></ref><ref id="B112-sensors-21-01734"><label>112.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Blaiech</surname><given-names>H.</given-names></name><name><surname>Neji</surname><given-names>M.</given-names></name><name><surname>Wali</surname><given-names>A.</given-names></name><name><surname>Alimi</surname><given-names>A.</given-names></name></person-group><article-title>Emotion recognition by analysis of EEG signals</article-title><source>Proceedings of the 13th International Conference on Hybrid Intelligent Systems</source><conf-loc>Arlington, VA, USA</conf-loc><conf-date>6–10 November 2013</conf-date><fpage>312</fpage><lpage>318</lpage></element-citation></ref><ref id="B113-sensors-21-01734"><label>113.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nugraha</surname><given-names>B.</given-names></name><name><surname>Sarno</surname><given-names>R.</given-names></name><name><surname>Asfani</surname><given-names>D.</given-names></name><name><surname>Igasaki</surname><given-names>T.</given-names></name><name><surname>Munawar</surname><given-names>M.</given-names></name></person-group><article-title>Classification of driver fatigue state based on EEG using Emotiv EPOC+</article-title><source>J. Theor. Appl. Inf. Technol.</source><year>2016</year><volume>86</volume><fpage>347</fpage><lpage>359</lpage></element-citation></ref><ref id="B114-sensors-21-01734"><label>114.</label><element-citation publication-type="book"><person-group person-group-type="author"><name><surname>Huang</surname><given-names>R.S.</given-names></name><name><surname>Jung</surname><given-names>T.P.</given-names></name><name><surname>Makeig</surname><given-names>S.</given-names></name></person-group><article-title>Tonic Changes in EEG Power Spectra during Simulated Driving</article-title><source>International Conference on Foundations of Augmented Cognition</source><publisher-name>Springer</publisher-name><publisher-loc>Berlin, Germany</publisher-loc><year>2009</year><fpage>394</fpage><lpage>403</lpage></element-citation></ref><ref id="B115-sensors-21-01734"><label>115.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Majumder</surname><given-names>S.</given-names></name><name><surname>Guragain</surname><given-names>B.</given-names></name><name><surname>Wang</surname><given-names>C.</given-names></name><name><surname>Wilson</surname><given-names>N.</given-names></name></person-group><article-title>On-board Drowsiness Detection using EEG: Current Status and Future Prospects</article-title><source>Proceedings of the IEEE International Conference on Electro Information Technology (EIT)</source><conf-loc>Brookings, SD, USA</conf-loc><conf-date>20–22 May 2019</conf-date><fpage>483</fpage><lpage>490</lpage></element-citation></ref><ref id="B116-sensors-21-01734"><label>116.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>H.</given-names></name><name><surname>Silva</surname><given-names>F.</given-names></name><name><surname>Ohata</surname><given-names>E.</given-names></name><name><surname>Medeiros</surname><given-names>A.</given-names></name><name><surname>Filho</surname><given-names>P.</given-names></name></person-group><article-title>Bi-Dimensional Approach Based on Transfer Learning for Alcoholism Pre-disposition Classification via EEG Signals</article-title><source>Front. Hum. Neurosci.</source><year>2020</year><volume>14</volume><fpage>365</fpage><pub-id pub-id-type="doi">10.3389/fnhum.2020.00365</pub-id><pub-id pub-id-type="pmid">33061900</pub-id></element-citation></ref><ref id="B117-sensors-21-01734"><label>117.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Trejo</surname><given-names>L.</given-names></name><name><surname>Kubitz</surname><given-names>K.</given-names></name><name><surname>Rosipal</surname><given-names>R.</given-names></name><name><surname>Kochavi</surname><given-names>R.</given-names></name><name><surname>Montgomery</surname><given-names>L.</given-names></name></person-group><article-title>EEG-Based Estimation and Classification of Mental Fatigue</article-title><source>Psychology</source><year>2015</year><volume>6</volume><fpage>572</fpage><lpage>589</lpage><pub-id pub-id-type="doi">10.4236/psych.2015.65055</pub-id></element-citation></ref><ref id="B118-sensors-21-01734"><label>118.</label><element-citation publication-type="confproc"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>R.</given-names></name><name><surname>Wang</surname><given-names>Y.</given-names></name><name><surname>Luo</surname><given-names>C.</given-names></name></person-group><article-title>EEG-Based Real-Time Drowsiness Detection Using Hilbert-Huang Transform</article-title><source>Proceedings of the 7th International Conference on Intelligent Human-Machine Systems and Cybernetics</source><conf-loc>Hangzhou, China</conf-loc><conf-date>26–27 August 2015</conf-date><fpage>195</fpage><lpage>198</lpage></element-citation></ref><ref id="B119-sensors-21-01734"><label>119.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Awais</surname><given-names>M.</given-names></name><name><surname>Badruddin</surname><given-names>N.</given-names></name><name><surname>Drieberg</surname><given-names>M.</given-names></name></person-group><article-title>A Hybrid Approach to Detect Driver Drowsiness Utilizing Physiological Signals to Improve System Performance and Wearability</article-title><source>Sensors</source><year>2017</year><volume>17</volume><elocation-id>1991</elocation-id><pub-id pub-id-type="doi">10.3390/s17091991</pub-id></element-citation></ref><ref id="B120-sensors-21-01734"><label>120.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nguyen</surname><given-names>T.</given-names></name><name><surname>Ahn</surname><given-names>S.</given-names></name><name><surname>Jang</surname><given-names>H.</given-names></name><name><surname>Jun</surname><given-names>S.C.</given-names></name><name><surname>Kim</surname><given-names>J.G.</given-names></name></person-group><article-title>Utilization of a combined EEG/NIRS system to predict driver drowsiness</article-title><source>Sci. Rep.</source><year>2017</year><volume>7</volume><fpage>43933</fpage><pub-id pub-id-type="pmid">28266633</pub-id></element-citation></ref><ref id="B121-sensors-21-01734"><label>121.</label><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Noori</surname><given-names>S.</given-names></name><name><surname>Mikaeili</surname><given-names>M.</given-names></name></person-group><article-title>Driving Drowsiness Detection Using Fusion of Electroencephalography, Electrooculography, and Driving Quality Signals</article-title><source>J. Med. Signal. Sens.</source><year>2016</year><volume>6</volume><fpage>39</fpage><lpage>46</lpage></element-citation></ref></ref-list></back><floats-group><fig id="sensors-21-01734-f001" orientation="portrait" position="float"><label>Figure 1</label><caption><p>Pipeline of the proposed drowsiness detection (DD) system.</p></caption><graphic xlink:href="sensors-21-01734-g001"></graphic></fig><fig id="sensors-21-01734-f002" orientation="portrait" position="float"><label>Figure 2</label><caption><p>The different components of the <italic>Emotiv EPOC<inline-formula><mml:math id="mm26"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> headset: (<bold>a</bold>) helmet, (<bold>b</bold>) fourteen-sensors box, (<bold>c</bold>) saline solution and (<bold>d</bold>) USB Key with cable.</p></caption><graphic xlink:href="sensors-21-01734-g002"></graphic></fig><fig id="sensors-21-01734-f003" orientation="portrait" position="float"><label>Figure 3</label><caption><p>Location of the <italic>Emotiv EPOC<inline-formula><mml:math id="mm27"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> helmet of the International System (10–20) [<xref rid="B58-sensors-21-01734" ref-type="bibr">58</xref>].</p></caption><graphic xlink:href="sensors-21-01734-g003"></graphic></fig><fig id="sensors-21-01734-f004" orientation="portrait" position="float"><label>Figure 4</label><caption><p>Example annotation of drowsy (<bold>a</bold>) and awake (<bold>b</bold>) of our electroencephalogram (EEG) signal collection.</p></caption><graphic xlink:href="sensors-21-01734-g004"></graphic></fig><fig id="sensors-21-01734-f005" orientation="portrait" position="float"><label>Figure 5</label><caption><p>Diagram of the proposed convolutional neural network (CNN) model.</p></caption><graphic xlink:href="sensors-21-01734-g005"></graphic></fig><fig id="sensors-21-01734-f006" orientation="portrait" position="float"><label>Figure 6</label><caption><p>(<bold>a</bold>) Accuracy graph, (<bold>b</bold>) loss graph.</p></caption><graphic xlink:href="sensors-21-01734-g006"></graphic></fig><fig id="sensors-21-01734-f007" orientation="portrait" position="float"><label>Figure 7</label><caption><p>The highest results of the confusion matrix of 7 electrodes with DA.</p></caption><graphic xlink:href="sensors-21-01734-g007"></graphic></fig><table-wrap id="sensors-21-01734-t001" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t001_Table 1</object-id><label>Table 1</label><caption><p>The characteristics of the <italic>Emotiv EPOC<inline-formula><mml:math id="mm28"><mml:mrow><mml:msup><mml:mrow></mml:mrow><mml:mo>+</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula></italic> helmet.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Characteristics</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">EEG Headset</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Number of channels</td><td align="center" valign="middle" rowspan="1" colspan="1">14 (plus 2 references CMS and DRL)</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Channel names</td><td align="center" valign="middle" rowspan="1" colspan="1">AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, AF4</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Sampling rate</td><td align="center" valign="middle" rowspan="1" colspan="1">128 SPS (2048 Hz internal)</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Sampling method</td><td align="center" valign="middle" rowspan="1" colspan="1">Sequential sampling</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Bandwidth</td><td align="center" valign="middle" rowspan="1" colspan="1">0.2–45 Hz, Digital notch filters at 50 Hz and 60 Hz</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Resolution</td><td align="center" valign="middle" rowspan="1" colspan="1">14 bits</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Filtration</td><td align="center" valign="middle" rowspan="1" colspan="1">Sinc filter</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Dynamic range</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">8400 µV (microvolts)</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t002" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t002_Table 2</object-id><label>Table 2</label><caption><p>Characteristics of brain waves.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Brainwaves</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Description</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Frequency Interval</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Location</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Gamma</td><td align="center" valign="middle" rowspan="1" colspan="1">Refers to hyper-vigilance state</td><td align="center" valign="middle" rowspan="1" colspan="1">>30 Hz</td><td align="center" valign="middle" rowspan="1" colspan="1">—</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Beta</td><td align="center" valign="middle" rowspan="1" colspan="1">Refers to alert state</td><td align="center" valign="middle" rowspan="1" colspan="1">13 to 30 Hz</td><td align="center" valign="middle" rowspan="1" colspan="1">Frontal and Central</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Alpha</td><td align="center" valign="middle" rowspan="1" colspan="1">Refers to waking state</td><td align="center" valign="middle" rowspan="1" colspan="1">8 to 13 Hz</td><td align="center" valign="middle" rowspan="1" colspan="1">Frontal and Occipital</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Theta</td><td align="center" valign="middle" rowspan="1" colspan="1">Refers to the half-sleep</td><td align="center" valign="middle" rowspan="1" colspan="1">4 to 7 Hz</td><td align="center" valign="middle" rowspan="1" colspan="1">Temporal and Median</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Alpha-Theta</td><td align="center" valign="middle" rowspan="1" colspan="1">Refers to waking and relaxation states</td><td align="center" valign="middle" rowspan="1" colspan="1">5 to 9 Hz</td><td align="center" valign="middle" rowspan="1" colspan="1">Temporal and Occipital</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Delta</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Refers to consciousness and sleep states</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">0.5 to 4 Hz</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Frontal lobe</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t003" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t003_Table 3</object-id><label>Table 3</label><caption><p>The architectures scores with 2, 3, and 4 states.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">CNNs Architectures</th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">ResNet</th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Inception</th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">WaveNet</th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Simple CNN</th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th><th align="left" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1"></th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">References</td><td colspan="3" align="left" valign="middle" rowspan="1">[<xref rid="B86-sensors-21-01734" ref-type="bibr">86</xref>,<xref rid="B87-sensors-21-01734" ref-type="bibr">87</xref>,<xref rid="B88-sensors-21-01734" ref-type="bibr">88</xref>,<xref rid="B89-sensors-21-01734" ref-type="bibr">89</xref>,<xref rid="B90-sensors-21-01734" ref-type="bibr">90</xref>]</td><td align="left" valign="middle" rowspan="1" colspan="1">[<xref rid="B96-sensors-21-01734" ref-type="bibr">96</xref>,<xref rid="B97-sensors-21-01734" ref-type="bibr">97</xref>]</td><th align="left" valign="middle" rowspan="1" colspan="1"></th><th align="left" valign="middle" rowspan="1" colspan="1"></th><td align="left" valign="middle" rowspan="1" colspan="1">[<xref rid="B84-sensors-21-01734" ref-type="bibr">84</xref>,<xref rid="B91-sensors-21-01734" ref-type="bibr">91</xref>]</td><th align="left" valign="middle" rowspan="1" colspan="1"></th><th align="left" valign="middle" rowspan="1" colspan="1"></th><td align="left" valign="middle" rowspan="1" colspan="1">[<xref rid="B42-sensors-21-01734" ref-type="bibr">42</xref>,<xref rid="B92-sensors-21-01734" ref-type="bibr">92</xref>,<xref rid="B93-sensors-21-01734" ref-type="bibr">93</xref>,<xref rid="B94-sensors-21-01734" ref-type="bibr">94</xref>,<xref rid="B95-sensors-21-01734" ref-type="bibr">95</xref>]</td><th align="left" valign="middle" rowspan="1" colspan="1"></th><th align="left" valign="middle" rowspan="1" colspan="1"></th></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">States number</td><td align="center" valign="middle" rowspan="1" colspan="1">2</td><td align="center" valign="middle" rowspan="1" colspan="1">3</td><td align="center" valign="middle" rowspan="1" colspan="1">4</td><td align="center" valign="middle" rowspan="1" colspan="1">2</td><td align="center" valign="middle" rowspan="1" colspan="1">3</td><td align="center" valign="middle" rowspan="1" colspan="1">4</td><td align="center" valign="middle" rowspan="1" colspan="1">2</td><td align="center" valign="middle" rowspan="1" colspan="1">3</td><td align="center" valign="middle" rowspan="1" colspan="1">4</td><td align="center" valign="middle" rowspan="1" colspan="1">2</td><td align="center" valign="middle" rowspan="1" colspan="1">3</td><td align="center" valign="middle" rowspan="1" colspan="1">4</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1"><italic>Complexity</italic></td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.33</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.33</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.33</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1"><italic>Performance</italic></td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.33</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1"><italic>Time-consumption</italic></td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1"><italic>1D-dimension</italic></td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td><td align="center" valign="middle" rowspan="1" colspan="1">0.5</td><td align="center" valign="middle" rowspan="1" colspan="1">0.66</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td><td align="center" valign="middle" rowspan="1" colspan="1">1</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Total</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2.5</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2.31</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">0</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">1.33</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2.15</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">3</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">2.65</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">4</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">3.5</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">3.66</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t004" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t004_Table 4</object-id><label>Table 4</label><caption><p>Convolutional layers parameters.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Parameters</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Role</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Filters</td><td align="center" valign="middle" rowspan="1" colspan="1">Feature extraction</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Kernel size</td><td align="center" valign="middle" rowspan="1" colspan="1">Convolutional window specification</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Kernel initializer</td><td align="center" valign="middle" rowspan="1" colspan="1">Initialization of all values</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Activation</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Applied after performing the convolution</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t005" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t005_Table 5</object-id><label>Table 5</label><caption><p>Detailed table of each participant’s status.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Participants</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Morning</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Afternoon</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Evening</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">P1 (26 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P2 (46 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P3 (37 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P4 (35 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P5 (64 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P6 (62 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P7 (53 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P8 (63 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P9 (59 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P10 (24 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P11 (17 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P12 (22 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">P13 (14 years)</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td><td align="center" valign="middle" rowspan="1" colspan="1">Drowsy</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">P14 (43 years)</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Awake</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Drowsy</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t006" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t006_Table 6</object-id><label>Table 6</label><caption><p>Summary of our model.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Participants</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Morning</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Afternoon</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Evening</th></tr><tr><th align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Layer Num</th><th align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Type</th><th align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Output Shape</th><th align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Parameters</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 1</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 2)</td><td align="center" valign="middle" rowspan="1" colspan="1">1024</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 2</td><td align="center" valign="middle" rowspan="1" colspan="1">Conv 1D</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 512)</td><td align="center" valign="middle" rowspan="1" colspan="1">33,280</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 3</td><td align="center" valign="middle" rowspan="1" colspan="1">Conv 1D</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 512)</td><td align="center" valign="middle" rowspan="1" colspan="1">8,389,120</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 4</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 512)</td><td align="center" valign="middle" rowspan="1" colspan="1">2048</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 5</td><td align="center" valign="middle" rowspan="1" colspan="1">Dropout</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 512)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 6</td><td align="center" valign="middle" rowspan="1" colspan="1">Conv 1DN</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">4,194,560</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 7</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">1024</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 8</td><td align="center" valign="middle" rowspan="1" colspan="1">Dropout</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 9</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">1024</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 10</td><td align="center" valign="middle" rowspan="1" colspan="1">Conv 1D</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">2,097,408</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 11</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 256, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">1024</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 12</td><td align="center" valign="middle" rowspan="1" colspan="1">Maxpool 1D</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 2, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 13</td><td align="center" valign="middle" rowspan="1" colspan="1">Dropout</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 2, 256)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 14</td><td align="center" valign="middle" rowspan="1" colspan="1">Flatten</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 512)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 15</td><td align="center" valign="middle" rowspan="1" colspan="1">Dense</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 128)</td><td align="center" valign="middle" rowspan="1" colspan="1">65,664</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 16</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None,128)</td><td align="center" valign="middle" rowspan="1" colspan="1">512</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 17</td><td align="center" valign="middle" rowspan="1" colspan="1">Dropout</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 128)</td><td align="center" valign="middle" rowspan="1" colspan="1">0</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Layer 18</td><td align="center" valign="middle" rowspan="1" colspan="1">Batch Normalization</td><td align="center" valign="middle" rowspan="1" colspan="1">(None, 128)</td><td align="center" valign="middle" rowspan="1" colspan="1">512</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Layer 19</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Dense</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">(None, 1)</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">129</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t007" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t007_Table 7</object-id><label>Table 7</label><caption><p>Training, validation and testing accuracy of the various numbers of electrodes without data augmentation (DA).</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Number of Electrods</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">2</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">4</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">7</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">14</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy train</td><td align="center" valign="middle" rowspan="1" colspan="1">78.20%</td><td align="center" valign="middle" rowspan="1" colspan="1">85.82%</td><td align="center" valign="middle" rowspan="1" colspan="1">88.22%</td><td align="center" valign="middle" rowspan="1" colspan="1">90.46%</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy Validation</td><td align="center" valign="middle" rowspan="1" colspan="1">74.33%</td><td align="center" valign="middle" rowspan="1" colspan="1">80.09%</td><td align="center" valign="middle" rowspan="1" colspan="1">86.30%</td><td align="center" valign="middle" rowspan="1" colspan="1">87.95%</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Accuracy test</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">68.79%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">54.14%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">72.41%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">79.43%</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t008" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t008_Table 8</object-id><label>Table 8</label><caption><p>Training, validation and testing accuracy of the various numbers of electrodes with DA.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Number of Electrods</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">2</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">4</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">7</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">14</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy train</td><td align="center" valign="middle" rowspan="1" colspan="1">94.30%</td><td align="center" valign="middle" rowspan="1" colspan="1">97.25%</td><td align="center" valign="middle" rowspan="1" colspan="1">98.88%</td><td align="center" valign="middle" rowspan="1" colspan="1">93.69%</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy Validation</td><td align="center" valign="middle" rowspan="1" colspan="1">78.14%</td><td align="center" valign="middle" rowspan="1" colspan="1">86.06%</td><td align="center" valign="middle" rowspan="1" colspan="1">93.27%</td><td align="center" valign="middle" rowspan="1" colspan="1">89.22%</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Accuracy test</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">77.41%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">78.49%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.14%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">82.07%</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t009" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t009_Table 9</object-id><label>Table 9</label><caption><p>Average accuracies of training, validation and testing of 7 electrodes with DA.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Run</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">1</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">2</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">3</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Average Accuracy</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy train</td><td align="center" valign="middle" rowspan="1" colspan="1">98.94%</td><td align="center" valign="middle" rowspan="1" colspan="1">98.90%</td><td align="center" valign="middle" rowspan="1" colspan="1">98.81 %</td><td align="center" valign="middle" rowspan="1" colspan="1">98.88%</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy Validation</td><td align="center" valign="middle" rowspan="1" colspan="1">92.15%</td><td align="center" valign="middle" rowspan="1" colspan="1">93.88%</td><td align="center" valign="middle" rowspan="1" colspan="1">93.79%</td><td align="center" valign="middle" rowspan="1" colspan="1">93.27%</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Accuracy test</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.01%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.42%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.14%</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t010" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t010_Table 10</object-id><label>Table 10</label><caption><p>The experimental results of cross-validation for 7 electrodes with DA.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Train and Validation Sets</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">80%, 20%</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">60%, 40%</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">40%, 60%</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">20%, 80%</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy train</td><td align="center" valign="middle" rowspan="1" colspan="1">98.94%</td><td align="center" valign="middle" rowspan="1" colspan="1">98.81 %</td><td align="center" valign="middle" rowspan="1" colspan="1">98.66%</td><td align="center" valign="middle" rowspan="1" colspan="1">98.83%</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy Validation</td><td align="center" valign="middle" rowspan="1" colspan="1">92.15%</td><td align="center" valign="middle" rowspan="1" colspan="1">89.82%</td><td align="center" valign="middle" rowspan="1" colspan="1">88.32%</td><td align="center" valign="middle" rowspan="1" colspan="1">89.48%</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Accuracy test</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.01%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">88.20%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">84.94%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">84.96%</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t011" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t011_Table 11</object-id><label>Table 11</label><caption><p>Accuracy comparison of the proposed CNN model with ResNet, Inception and WaveNet models.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Models</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Proposed CNN</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Inception</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Resnet</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Wavenet</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy train</td><td align="center" valign="middle" rowspan="1" colspan="1">98.88%</td><td align="center" valign="middle" rowspan="1" colspan="1">88.91%</td><td align="center" valign="middle" rowspan="1" colspan="1">79.03%</td><td align="center" valign="middle" rowspan="1" colspan="1">71.54%</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">Accuracy Validation</td><td align="center" valign="middle" rowspan="1" colspan="1">93.27%</td><td align="center" valign="middle" rowspan="1" colspan="1">67.70%</td><td align="center" valign="middle" rowspan="1" colspan="1">69.86%</td><td align="center" valign="middle" rowspan="1" colspan="1">67.40%</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Accuracy test</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.14%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">74.87%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">72.80%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">75%</td></tr></tbody></table></table-wrap><table-wrap id="sensors-21-01734-t012" orientation="portrait" position="float"><object-id pub-id-type="pii">sensors-21-01734-t012_Table 12</object-id><label>Table 12</label><caption><p>Accuracy test comparison with related works.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">DD Methodology</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Accuracy Test</th><th align="center" valign="middle" style="border-bottom:solid thin;border-top:solid thin" rowspan="1" colspan="1">Classification Method</th></tr></thead><tbody><tr><td align="center" valign="middle" rowspan="1" colspan="1">E.J. Cheng et al. [<xref rid="B54-sensors-21-01734" ref-type="bibr">54</xref>]</td><td align="center" valign="middle" rowspan="1" colspan="1">74.95%</td><td align="center" valign="middle" rowspan="1" colspan="1">CNN</td></tr><tr><td align="center" valign="middle" rowspan="1" colspan="1">L. Guarda et al. [<xref rid="B40-sensors-21-01734" ref-type="bibr">40</xref>]</td><td align="center" valign="middle" rowspan="1" colspan="1">83.93%</td><td align="center" valign="middle" rowspan="1" colspan="1">ConvNets</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">Proposed Method</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">90.14%</td><td align="center" valign="middle" style="border-bottom:solid thin" rowspan="1" colspan="1">CNN</td></tr></tbody></table></table-wrap></floats-group></pmc></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sante/explor/MaghrebDataLibMedV2/Data/Pmc/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000284 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd -nk 000284 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Sante
|area= MaghrebDataLibMedV2
|flux= Pmc
|étape= Corpus
|type= RBID
|clé=
|texte=
}}
| This area was generated with Dilib version V0.6.38. |  |