Links to Exploration step
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Identification of Diseases in Newborns Using Advanced Acoustic Features of Cry Signals</title><author><name sortKey="Kheddache, Yasmina" sort="Kheddache, Yasmina" uniqKey="Kheddache Y" first="Yasmina" last="Kheddache">Yasmina Kheddache</name><affiliation><nlm:aff id="aff0001">Faculty of Science and Technology, Ziane Achour University, 3117 Djelfa, Algeria</nlm:aff></affiliation></author><author><name sortKey="Tadj, Chakib" sort="Tadj, Chakib" uniqKey="Tadj C" first="Chakib" last="Tadj">Chakib Tadj</name><affiliation><nlm:aff id="aff0002">Department of Electrical Engineering, École de technologie supérieure, H3C 1K3 Montréal (Qc), Canada</nlm:aff></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">33281921</idno><idno type="pmc">7672377</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7672377</idno><idno type="RBID">PMC:7672377</idno><idno type="doi">10.1016/j.bspc.2019.01.010</idno><date when="2019">2019</date><idno type="wicri:Area/Pmc/Corpus">000167</idno><idno type="wicri:explorRef" wicri:stream="Pmc" wicri:step="Corpus" wicri:corpus="PMC">000167</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Identification of Diseases in Newborns Using Advanced Acoustic Features of Cry Signals</title><author><name sortKey="Kheddache, Yasmina" sort="Kheddache, Yasmina" uniqKey="Kheddache Y" first="Yasmina" last="Kheddache">Yasmina Kheddache</name><affiliation><nlm:aff id="aff0001">Faculty of Science and Technology, Ziane Achour University, 3117 Djelfa, Algeria</nlm:aff></affiliation></author><author><name sortKey="Tadj, Chakib" sort="Tadj, Chakib" uniqKey="Tadj C" first="Chakib" last="Tadj">Chakib Tadj</name><affiliation><nlm:aff id="aff0002">Department of Electrical Engineering, École de technologie supérieure, H3C 1K3 Montréal (Qc), Canada</nlm:aff></affiliation></author></analytic><series><title level="j">Biomedical Signal Processing and Control</title><idno type="ISSN">1746-8094</idno><imprint><date when="2019">2019</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><p>Our challenge in the current study is to extend research on the cries of newborns for the early diagnosis of different pathologies. This paper proposes a recognition system for healthy and pathological cries using a probabilistic neural network classifier. Two different kinds of features have been used to characterize newborn cry signals: 1) acoustic features such as fundamental frequency glide (F<sub>0glide</sub>) and resonance frequencies dysregulation (RFs<sub>dys</sub>); 2) conventional features such as mel-frequency cestrum coefficients.</p><p>This paper describes the automatic estimation of the proposed characteristics and the performance evaluation of these features in identifying pathological cries. The adopted methods for F<sub>0glides</sub> and RFs<sub>dys</sub> estimation are based on the derived function of the F0 contour and the jump “J” of the RFs between two subsequent tunings, respectively. The database used contains 3250 cry samples of full-term and preterm newborns, and includes healthy and pathologic cries.</p><p>The obtained results indicate the important association between the quantified features and some studied pathologies, and also an improvement in the identification of pathologic cries. The best result obtained is 88.71% for the correct identification of health status of preterm newborns, and 82% for the correct identification of full-term infants with a specific disease. We conclude that using the proposed characteristics improves the diagnosis of pathologies in newborns. Moreover, the method applied in the estimation of these characteristics allows us to extend this study to other uninvestigated pathologies.</p></div></front><back><div1 type="bibliography"><listBibl><biblStruct><analytic><author><name sortKey="Fausto, O" uniqKey="Fausto O">O. Fausto</name></author><author><name sortKey="Galaviz, R" uniqKey="Galaviz R">R. Galaviz</name></author><author><name sortKey="Garcia, C A Reyes" uniqKey="Garcia C">C.A. Reyes Garcia</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lederman, D" uniqKey="Lederman D">D. Lederman</name></author><author><name sortKey="Zmora, E" uniqKey="Zmora E">E. Zmora</name></author><author><name sortKey="Hauschildt, S" uniqKey="Hauschildt S">S. Hauschildt</name></author><author><name sortKey="Stellzig Eisenhauer, A" uniqKey="Stellzig Eisenhauer A">A. Stellzig-Eisenhauer</name></author><author><name sortKey="Wermke, K" uniqKey="Wermke K">K. Wermke</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hariharan, M" uniqKey="Hariharan M">M. Hariharan</name></author><author><name sortKey="Yaacob, S" uniqKey="Yaacob S">S. Yaacob</name></author><author><name sortKey="Awang, S Ardeena Awatie" uniqKey="Awang S">S. Ardeena awatie Awang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kheddache, Y" uniqKey="Kheddache Y">Y Kheddache</name></author><author><name sortKey="Tadj, C" uniqKey="Tadj C">C Tadj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Manfredi, C" uniqKey="Manfredi C">C Manfredi</name></author><author><name sortKey="Andrea, B" uniqKey="Andrea B">B Andrea</name></author><author><name sortKey="Melino, D" uniqKey="Melino D">D Melino</name></author><author><name sortKey="Viellevoyec, R" uniqKey="Viellevoyec R">R Viellevoyec</name></author><author><name sortKey="Masendu, K" uniqKey="Masendu K">K Masendu</name></author><author><name sortKey="Silvia, O" uniqKey="Silvia O">O Silvia</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lagasse, Ll" uniqKey="Lagasse L">LL LaGasse</name></author><author><name sortKey="Neal, Ar" uniqKey="Neal A">AR Neal</name></author><author><name sortKey="Lester, Bm" uniqKey="Lester B">BM Lester</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Orlandia, S" uniqKey="Orlandia S">S. Orlandia</name></author><author><name sortKey="Bandiniab, A" uniqKey="Bandiniab A">A. Bandiniab</name></author><author><name sortKey="Fiaschia, F F" uniqKey="Fiaschia F">F.F. Fiaschia</name></author><author><name sortKey="Manfredia, C" uniqKey="Manfredia C">C. Manfredia</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wasz Hockert, O" uniqKey="Wasz Hockert O">O Wasz-Hockert</name></author><author><name sortKey="Michelsson, K" uniqKey="Michelsson K">K Michelsson</name></author><author><name sortKey="Lind, J" uniqKey="Lind J">J Lind</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Michelson, K" uniqKey="Michelson K">K Michelson</name></author><author><name sortKey="Todd De Barra, H" uniqKey="Todd De Barra H">H Todd de Barra</name></author><author><name sortKey="Michelson, O" uniqKey="Michelson O">O Michelson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lester, Bm" uniqKey="Lester B">BM Lester</name></author><author><name sortKey="Lagasse, Ll" uniqKey="Lagasse L">LL LaGasse</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Corwin, Michael J" uniqKey="Corwin M">Michael J Corwin</name></author><author><name sortKey="Kayne, Herbert" uniqKey="Kayne H">Herbert Kayne</name></author><author><name sortKey="Lester, Barry M" uniqKey="Lester B">Barry M Lester</name></author><author><name sortKey="Sepkoski, Carol" uniqKey="Sepkoski C">Carol Sepkoski</name></author><author><name sortKey="Mclaughlin, Sarah" uniqKey="Mclaughlin S">Sarah McLaughlin</name></author><author><name sortKey="Golub, Howard L" uniqKey="Golub H">Howard L Golub</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Orozco, J" uniqKey="Orozco J">J Orozco</name></author><author><name sortKey="Garcia, C A R" uniqKey="Garcia C">C.A.R Garcia</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hesam Farsaie, Alaie" uniqKey="Hesam Farsaie A">Alaie Hesam Farsaie</name></author><author><name sortKey="Chakib, Tadj" uniqKey="Chakib T">Tadj Chakib</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Amaro Camargo, E" uniqKey="Amaro Camargo E">E Amaro-Camargo</name></author><author><name sortKey="Reyes Garcia, C" uniqKey="Reyes Garcia C">C Reyes García</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cano Ortiz, Sergio D" uniqKey="Cano Ortiz S">Sergio D Cano Ortiz</name></author><author><name sortKey="Beceiro, Daniel I Escobedo" uniqKey="Beceiro D">Daniel I Escobedo Beceiro</name></author><author><name sortKey="Ekkel, Taco" uniqKey="Ekkel T">Taco Ekkel</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Fisichelli, Vincent R" uniqKey="Fisichelli V">Vincent R Fisichelli</name></author><author><name sortKey="Karelitz, S" uniqKey="Karelitz S">S. Karelitz</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kheddache, Y" uniqKey="Kheddache Y">Y Kheddache</name></author><author><name sortKey="Tadj, C" uniqKey="Tadj C">C Tadj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kheddache, Y" uniqKey="Kheddache Y">Y Kheddache</name></author><author><name sortKey="Tadj, C" uniqKey="Tadj C">C Tadj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kheddache, Y" uniqKey="Kheddache Y">Y Kheddache</name></author><author><name sortKey="Tadj, C" uniqKey="Tadj C">C Tadj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Aalto, A" uniqKey="Aalto A">A. Aalto</name></author><author><name sortKey="Aalto, D" uniqKey="Aalto D">D. Aalto</name></author><author><name sortKey="Malinen, J" uniqKey="Malinen J">J. Malinen</name></author><author><name sortKey="Vainio, M" uniqKey="Vainio M">M Vainio</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Aalto, A" uniqKey="Aalto A">A. Aalto</name></author><author><name sortKey="Aalto, D" uniqKey="Aalto D">D. Aalto</name></author><author><name sortKey="Malinen, J" uniqKey="Malinen J">J. Malinen</name></author><author><name sortKey="Vainio, M" uniqKey="Vainio M">M Vainio</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Titze, Ingo R" uniqKey="Titze I">Ingo R. Titze</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Honda, Kiyoshi" uniqKey="Honda K">Kiyoshi honda</name></author><author><name sortKey="Hirai, Hiroyuki" uniqKey="Hirai H">Hiroyuki hirai</name></author><author><name sortKey="Masaki, Shinobu" uniqKey="Masaki S">Shinobu masaki</name></author><author><name sortKey="Shimada, Yasuhiro" uniqKey="Shimada Y">Yasuhiro shimada</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Stevens, K N" uniqKey="Stevens K">K.N stevens</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bernardoni, Nathalie Henrich" uniqKey="Bernardoni N">Nathalie Henrich Bernardoni</name></author><author><name sortKey="Smith, John" uniqKey="Smith J">John Smith</name></author><author><name sortKey="Wolfe, Joe" uniqKey="Wolfe J">Joe Wolfe</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Echternach, M" uniqKey="Echternach M">M Echternach</name></author><author><name sortKey="Sundberg, J" uniqKey="Sundberg J">J Sundberg</name></author><author><name sortKey="Arndt, S" uniqKey="Arndt S">S Arndt</name></author><author><name sortKey="Markl, M" uniqKey="Markl M">M Markl</name></author><author><name sortKey="Schumacher, M" uniqKey="Schumacher M">M Schumacher</name></author><author><name sortKey="Richter, B" uniqKey="Richter B">B Richter</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sundberg, J" uniqKey="Sundberg J">J Sundberg</name></author><author><name sortKey="La, F M B" uniqKey="La F">F. M. B La</name></author><author><name sortKey="Gill, B P" uniqKey="Gill B">B. P Gill</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sundberg, Johan" uniqKey="Sundberg J">Johan Sundberg</name></author><author><name sortKey="Filipa, M B" uniqKey="Filipa M">M. B Filipa</name></author><author><name sortKey="Gill, Brian P" uniqKey="Gill B">Brian P. Gill</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Boersma, P" uniqKey="Boersma P">P. Boersma</name></author><author><name sortKey="Weenink, D" uniqKey="Weenink D">D. Weenink</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Abou Abbas, L" uniqKey="Abou Abbas L">L Abou-Abbas</name></author><author><name sortKey="Tadj, C" uniqKey="Tadj C">C Tadj</name></author><author><name sortKey="Gargour, C" uniqKey="Gargour C">C Gargour</name></author><author><name sortKey="Montazeri, L" uniqKey="Montazeri L">L Montazeri</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Markel, J D" uniqKey="Markel J">J.D Markel</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lederman, D" uniqKey="Lederman D">D Lederman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Manfredi, C" uniqKey="Manfredi C">C Manfredi</name></author><author><name sortKey="Bocchi, L" uniqKey="Bocchi L">L Bocchi</name></author><author><name sortKey="Orlandi, S" uniqKey="Orlandi S">S Orlandi</name></author><author><name sortKey="Spaccaterra, L" uniqKey="Spaccaterra L">L Spaccaterra</name></author><author><name sortKey="Donzelli, Gp" uniqKey="Donzelli G">GP Donzelli</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Fort, A" uniqKey="Fort A">A. Fort</name></author><author><name sortKey="Ismaellit, A" uniqKey="Ismaellit A">A. Ismaellit</name></author><author><name sortKey="Manfredi, C" uniqKey="Manfredi C">C. Manfredi</name></author><author><name sortKey="Bruscaglionit, P" uniqKey="Bruscaglionit P">P. Bruscaglionit</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wermk, K" uniqKey="Wermk K">K Wermk</name></author><author><name sortKey="Mende, W" uniqKey="Mende W">W Mende</name></author><author><name sortKey="Kempf, A" uniqKey="Kempf A">A Kempf</name></author><author><name sortKey="Manfredi, C" uniqKey="Manfredi C">C Manfredi</name></author><author><name sortKey="Bruscaglioni, P" uniqKey="Bruscaglioni P">P Bruscaglioni</name></author><author><name sortKey="Stellzig Eisenhauer, A" uniqKey="Stellzig Eisenhauer A">A Stellzig- Eisenhauer</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sweeney, Walter P" uniqKey="Sweeney W">Walter P. Sweeney</name></author><author><name sortKey="Musavi, Mohamad T" uniqKey="Musavi M">Mohamad T. Musavi</name></author><author><name sortKey="Guidi, John N" uniqKey="Guidi J">John N. Guidi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Othman, Mohd Fauzi" uniqKey="Othman M">Mohd Fauzi Othman</name></author><author><name sortKey="Mohd Basri, Mohd Ariffanan" uniqKey="Mohd Basri M">Mohd Ariffanan Mohd Basri</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kusy, Maciej" uniqKey="Kusy M">Maciej Kusy</name></author><author><name sortKey="Zajdel, Roman" uniqKey="Zajdel R">Roman Zajdel</name></author></analytic></biblStruct><biblStruct></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">Biomed Signal Process Control</journal-id><journal-id journal-id-type="iso-abbrev">Biomed Signal Process Control</journal-id><journal-id journal-id-type="publisher-id">BSPC</journal-id><journal-title-group><journal-title>Biomedical Signal Processing and Control</journal-title></journal-title-group><issn pub-type="ppub">1746-8094</issn><publisher><publisher-name>Elsevier Ltd.</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="pmid">33281921</article-id><article-id pub-id-type="pmc">7672377</article-id><article-id pub-id-type="publisher-id">BSPC-50-035</article-id><article-id pub-id-type="doi">10.1016/j.bspc.2019.01.010</article-id><article-categories><subj-group subj-group-type="heading"><subject>Articles</subject></subj-group></article-categories><title-group><article-title>Identification of Diseases in Newborns Using Advanced Acoustic Features of Cry Signals</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>Kheddache</surname><given-names>Yasmina</given-names></name><xref ref-type="aff" rid="aff0001">*</xref><xref ref-type="corresp" rid="cor1"></xref></contrib><contrib contrib-type="author"><name><surname>Tadj</surname><given-names>Chakib</given-names></name><xref ref-type="aff" rid="aff0002">**</xref></contrib><aff id="aff0001"><label>*</label>Faculty of Science and Technology, Ziane Achour University, 3117 Djelfa, Algeria</aff><aff id="aff0002"><label>**</label>Department of Electrical Engineering, École de technologie supérieure, H3C 1K3 Montréal (Qc), Canada</aff></contrib-group><author-notes><corresp id="cor1"><bold><italic>Corresponding Author:</italic></bold> Yasmina Kheddache, e-mail: <email xlink:href="yasmina.kheddache.1@etsmtl.net">yasmina.kheddache.1@etsmtl.net</email> fax: + (001) 514 396-8684</corresp></author-notes><pub-date pub-type="epub"><day>01</day><month>4</month><year>2019</year></pub-date><pub-date pub-type="ppub"><year>2019</year></pub-date><volume>50</volume><fpage>35</fpage><lpage>44</lpage><permissions><copyright-statement>© 2019 The Author(s).</copyright-statement><copyright-year>2019</copyright-year><license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by-nc-nd/4.0/"><license-p>This manuscript version is made available under the CC-BY-NC-ND 4.0 license.</license-p></license></permissions><abstract><p>Our challenge in the current study is to extend research on the cries of newborns for the early diagnosis of different pathologies. This paper proposes a recognition system for healthy and pathological cries using a probabilistic neural network classifier. Two different kinds of features have been used to characterize newborn cry signals: 1) acoustic features such as fundamental frequency glide (F<sub>0glide</sub>) and resonance frequencies dysregulation (RFs<sub>dys</sub>); 2) conventional features such as mel-frequency cestrum coefficients.</p><p>This paper describes the automatic estimation of the proposed characteristics and the performance evaluation of these features in identifying pathological cries. The adopted methods for F<sub>0glides</sub> and RFs<sub>dys</sub> estimation are based on the derived function of the F0 contour and the jump “J” of the RFs between two subsequent tunings, respectively. The database used contains 3250 cry samples of full-term and preterm newborns, and includes healthy and pathologic cries.</p><p>The obtained results indicate the important association between the quantified features and some studied pathologies, and also an improvement in the identification of pathologic cries. The best result obtained is 88.71% for the correct identification of health status of preterm newborns, and 82% for the correct identification of full-term infants with a specific disease. We conclude that using the proposed characteristics improves the diagnosis of pathologies in newborns. Moreover, the method applied in the estimation of these characteristics allows us to extend this study to other uninvestigated pathologies.</p></abstract><kwd-group><kwd>Pathologic cry</kwd><kwd>Classification</kwd><kwd>Probabilistic neural network</kwd><kwd>Mel-frequency cestrum coefficients</kwd><kwd>RF dysregulations</kwd><kwd>F<sub>0</sub> glides</kwd></kwd-group></article-meta></front><body><sec sec-type="Intro" id="sec1"><title>1. Introduction</title><p>The majority of sick babies appear healthy at birth. Thus, early diagnosis of hidden pathologies for a quick and effective treatment during their first week of life is crucial, as it could save the lives of these babies. However, the realization of a systematic neonatal diagnosis procedure for all newborns demands high costs because it involves the participation of numerous health professionals and specialized equipment. Thus, our aim is to develop a low-cost diagnostic system that allows pediatricians to detect pathologies affecting newborns using spontaneous cry signals.</p><p>Cry signal analysis is a valuable tool for predicting neonatal diseases. It allows for the recognition of a sick infant when signs of illnesses are absent. In the previous studies of infant cries, two different approaches have been adopted: 1) an automatic recognition that consist of cry classification using advanced signal processing techniques [<xref rid="cit0001" ref-type="bibr">1</xref>, <xref rid="cit0002" ref-type="bibr">2</xref>, <xref rid="cit0003" ref-type="bibr">3</xref>, <xref rid="cit0004" ref-type="bibr">4</xref>, and <xref rid="cit0005" ref-type="bibr">5</xref>], and 2) a spectral cry analysis based on observing the spectrograms of cry signals and software tools [<xref rid="cit0006" ref-type="bibr">6</xref>,<xref rid="cit0007" ref-type="bibr">7</xref>,<xref rid="cit0008" ref-type="bibr">8</xref>,<xref rid="cit0009" ref-type="bibr">9</xref>, and <xref rid="cit0010" ref-type="bibr">10</xref>].</p><p>However, the cries of newborns provide important acoustic parameters that are not considered while monitoring the first days of infant life and in the standard measurements of the Apgar score (appearance, pulse, grimace, activity, and respiration), which is used to verify a baby’s health immediately after birth. Moreover, studies regarding the development of reliable estimation procedures for the most important acoustics characteristics as well as the identification of their pathological markers are scarce.</p><p>Unlike previous works, the aim of our study is to propose an automated tool that uses important characteristics of cry signals to support the diagnosis of diseases in newborn infants. These characteristics are described in <xref ref-type="sec" rid="sec4.2.1">Section 4.2.1</xref>. They include acoustic parameters that qualify the vocal tract, such as mel-frequency cestrum coefficients (MFCCs) and resonance frequencies dysregulation (RFs<sub>dys</sub>), and the vocal fold, such as fundamental frequency glide (F<sub>0glide</sub>).</p><p>F<sub>0glide</sub> has been associated with central nervous system diseases, asphyxia, and malformations of the orolaryngeal tract; RFs<sub>dys</sub> indicate poor neural control of the vocal tract and breathing, and has been associated with hyperbilirubinemia, and prenatal tobacco and cocaine exposure [<xref rid="cit0011" ref-type="bibr">11</xref>]. MFCCs can represent properly various models of cries [<xref rid="cit0002" ref-type="bibr">2</xref>], they allow decoupling between the features of the vocal tract and the features generated by the source of excitation.</p><p>To differentiate pathologic cries from healthy ones, we strengthened the previous studies with data based on cry analysis. Hence, the primary purpose of our work is to complete a detailed analysis of acoustic phenomena produced in newborn cries, focusing specifically on the prevalence of F<sub>0glide</sub> and RFs<sub>dys</sub>. Moreover, we study the effectiveness of their use as an input of the proposed newborn’s pathological cry identification system (PCIS).</p><p>We performed specific procedures to estimate the average percentage of F<sub>0glide</sub>, the rising and falling times of F<sub>0glide</sub> (T<sub>glide</sub>), and the percentage of RFs<sub>dys</sub> in the cry signals of healthy and sick infants. By implementing these procedures as described in <xref ref-type="sec" rid="sec4.2.2">Section 4.2.2</xref> and by using the median and interquartile range, we managed to 1) determine the quantitative relationship among these acoustic parameters and the various studied pathologies, 2) facilitate their use in the proposed diagnostic system.</p><p>This paper is organized as follows. <xref ref-type="sec" rid="sec2">Section 2</xref> provides an overview of previous works. In <xref ref-type="sec" rid="sec3">Section 3</xref>, the details regarding the database used are presented. In <xref ref-type="sec" rid="sec4">Section 4</xref>, we explain our methodology for cry signal feature extraction. <xref ref-type="sec" rid="sec5">Section 5</xref> describes the proposed PCIS. In <xref ref-type="sec" rid="sec6">Section 6</xref>, we provide a statistical analysis of the estimated characteristics and also an evaluation of the obtained classification results. Finally, the conclusion is presented in <xref ref-type="sec" rid="sec7">Section 7</xref>, which also summarizes the suggestions and future directions for further research.</p></sec><sec id="sec2"><title>2. Previous studies</title><p>Studies regarding neonatal cry had focused primarily on the classification of cries and also on the relationship between the characteristics of cry signals and diseases. These two research directions progressed separately. Hence, the characteristics of cries that are closely related to the pathologies of newborns have not been considered in the classification of cry studies.</p><p>In cry recognition studies, different types of classifiers have been used, such as artificial neural networks [<xref rid="cit0012" ref-type="bibr">12</xref>], Gaussian mixture model [<xref rid="cit0013" ref-type="bibr">13</xref>], hidden Markov model [<xref rid="cit0002" ref-type="bibr">2</xref>], probabilistic neural network (PNN) [<xref rid="cit0003" ref-type="bibr">3</xref>], support vector machine [<xref rid="cit0014" ref-type="bibr">14</xref>], and radial basis function networks[<xref rid="cit0015" ref-type="bibr">15</xref>]. The developed automatic recognition systems were applied on cry signals of normal infants and infants suffering from deafness, asphyxia, cleft lip, or cleft palate. The most used features were MFCCs, wavelet packet transform coefficients, and linear predictive coding coefficients.</p><p>According to our knowledge, none have considered the acoustic characteristics of cry signals in the classification of pathologic cries.</p><p>In this study, we are interested in the early diagnosis of newborn pathologies using the acoustic characteristics of cry signals. Most direct methods that have been addressed in different studies to assess the acoustic properties of cry signals were based on spectrographic analysis using visual techniques [<xref rid="cit0008" ref-type="bibr">8</xref>, <xref rid="cit0009" ref-type="bibr">9</xref>]. The studied parameters were manually measurable in the spectrogram of the cry signal. These studies were primarily focused on babies at risk with low neurological factors such as prematurity, hyperbilirubinemia, prenatal exposure to drugs, and also neurological damages such as Krabbe’s disease, Down syndrome, asphyxia, and meningitis [<xref rid="cit0016" ref-type="bibr">16</xref>]. Several characteristics have been associated with these diseases, such as hyperphonic cries, F<sub>0</sub> irregularity, extremely high-pitched cries, dysphonic cries, change in phonation mode, variability of F<sub>0</sub>, glide of F<sub>0</sub>, and dysregulation of RFs [<xref rid="cit0007" ref-type="bibr">7</xref>,<xref rid="cit0008" ref-type="bibr">8</xref>,<xref rid="cit0010" ref-type="bibr">10</xref>]. In our previous studies [<xref rid="cit0010" ref-type="bibr">10</xref>,<xref rid="cit0017" ref-type="bibr">17</xref>,<xref rid="cit0018" ref-type="bibr">18</xref>], we adopted an automatic approach for the evaluation of the prevalence of these features. The results obtained indicated that the studied characteristics depend on the pathology itself.</p><p>We have also provided the experimental results corresponding to a proposed PCIS to separate pathologic cries from healthy ones, with and without using the studied characteristics such as the average percentage of hyperphonic cries and F0 irregularity [<xref rid="cit0017" ref-type="bibr">17</xref>]. The results of cry classification were better when these features were used in the PCIS.</p></sec><sec id="sec3"><title>3. Database and cry recording</title><p>The acoustic analysis presented herein was performed on a real database. The recordings of cry signals were created specifically to study the possibilities of the early diagnosis of various pathologies using spontaneous cries during the first days of infant life. They were performed in the Pediatrics Department at the Sainte-Justine Hospital in Montreal, using a small recorder, at a distance of 10 cm from a baby’s mouth with a sampling rate of 44.1 kHz and a resolution of 16 bits. This database is similar to that used in [<xref rid="cit0017" ref-type="bibr">17</xref>], [<xref rid="cit0018" ref-type="bibr">18</xref>], and [<xref rid="cit0019" ref-type="bibr">19</xref>]. A description of the recording sessions was presented in a previous work [<xref rid="cit0013" ref-type="bibr">13</xref>]. The recordings were performed with background noise; they have been performed for healthy and pathological newborns cries for different kinds of cries, such as hunger, sampling blood, and change of diapers. The categories of health conditions considered in our cry database are as follows: healthy, heart problems, neurological disorders, respiratory diseases, and blood abnormalities. The constructed dataset contains 3250 cry samples of 1-s duration from 66 babies. Among them were preterm and full-term newborns, from 1-day to 1-month old. It is distributed by pathologies and gestational age, as illustrated in <xref rid="t0001" ref-type="table">Table 1</xref>. It is noteworthy that the age of the babies does not exceed one month because the infants acquire voluntary control of their vocal tracts beyond this age [<xref rid="cit0010" ref-type="bibr">10</xref>].</p><table-wrap id="t0001" orientation="portrait" position="float"><label>Table 1</label><caption><p>Studied pathologies for different gestational ages</p></caption><table frame="hsides" rules="groups"><thead valign="bottom"><tr><th align="left" rowspan="1" colspan="1">Gestational age</th><th align="left" rowspan="1" colspan="1">Pathology</th><th align="left" rowspan="1" colspan="1">Sample size</th></tr></thead><tbody valign="bottom"><tr><td align="center" rowspan="7" colspan="1">Full-term newborn (t)</td><td align="left" rowspan="1" colspan="1">Healthy</td><td align="right" rowspan="1" colspan="1">1010</td></tr><tr><td align="left" rowspan="1" colspan="1">Hyperbilirubinemia</td><td align="right" rowspan="1" colspan="1">250</td></tr><tr><td align="left" rowspan="1" colspan="1">Vena cava thrombosis</td><td align="right" rowspan="1" colspan="1">77</td></tr><tr><td align="left" rowspan="1" colspan="1">Meningitis</td><td align="right" rowspan="1" colspan="1">115</td></tr><tr><td align="left" rowspan="1" colspan="1">Peritonitis</td><td align="right" rowspan="1" colspan="1">20</td></tr><tr><td align="left" rowspan="1" colspan="1">Asphyxia</td><td align="right" rowspan="1" colspan="1">190</td></tr><tr><td align="left" rowspan="1" colspan="1">Lingual frenum</td><td align="right" rowspan="1" colspan="1">141</td></tr><tr><td align="left" rowspan="6" colspan="1">Preterm newborn (P)</td><td align="left" rowspan="1" colspan="1">Healthy</td><td align="right" rowspan="1" colspan="1">764</td></tr><tr><td align="left" rowspan="1" colspan="1">IUGR-microcephaly (in utero growth retardation)</td><td align="right" rowspan="1" colspan="1">78</td></tr><tr><td align="left" rowspan="1" colspan="1">Tetralogy of Fallot</td><td align="right" rowspan="1" colspan="1">53</td></tr><tr><td align="left" rowspan="1" colspan="1">Gastroschisis</td><td align="right" rowspan="1" colspan="1">134</td></tr><tr><td align="left" rowspan="1" colspan="1">IUGR-asphyxia (intra-uterine growth retardation)</td><td align="right" rowspan="1" colspan="1">148</td></tr><tr><td align="left" rowspan="1" colspan="1">RDS (respiratory distress syndrome)</td><td align="right" rowspan="1" colspan="1">270</td></tr></tbody></table></table-wrap><p>All the samples in the database have been used to investigate and analyze the proposed acoustic features in cry signals. However, because more healthy samples are available than pathology samples, and the number of cry samples associated with each of the diseases are different, only part of this database has been used in the study of the newborn’s PCIS, as shown in <xref rid="t0004" ref-type="table">Table 4</xref> in <xref ref-type="sec" rid="sec6.2">Section 6.2</xref>.</p></sec><sec id="sec4"><title>4. Cry signal features extraction</title><p>Feature extraction is a crucial step in the classification task; it affects the accuracy and reliability of pattern recognition. In our work, new features were extracted in addition to MFCCs. They were used to characterize different models of pathological cries and also cries from healthy newborns. These new high level features are the F<sub>0glide</sub>, T<sub>glide</sub>, and RFs<sub>dys</sub>.</p><sec id="sec4.1"><title>4.1 MFCC</title><p>MFCCs are the most used feature coefficients in automatic infant cry classification [<xref rid="cit0001" ref-type="bibr">1</xref>, <xref rid="cit0002" ref-type="bibr">2</xref>, <xref rid="cit0004" ref-type="bibr">4</xref>]. Considering the statistical stationary of cry signals in short periods of time, MFCCs were extracted using a frame duration of 23.2 ms (1024 samples) with 50% recovery and a sampling rate of 44.1 kHz.</p><p>For every cry signal, we extracted 13 MFCC parameters for each frame of 23.2 ms. We obtained an MFCC matrix of 13 lines × N columns. N corresponds to the total number of frames in the whole cry signal.</p><p>The calculation of MFCCs was performed as follows:</p><list list-type="bullet"><list-item><p>Perform a glottal inverse filtering to attenuate the influence of the vocal tract.</p></list-item><list-item><p>Multiply each frame by the Hamming window.</p></list-item><list-item><p>Estimate the power spectrum sequence using fast Fourier transform (FFT).</p></list-item><list-item><p>Convert the frequency outputs by the FFT onto the mel scale and obtain the logarithm of all filterbank energies.</p></list-item></list><p>Apply the discrete cosine transform on the log filterbank energies.</p><p>Hence, MFCCs are obtained from the following relation:</p><disp-formula id="eq1"><alternatives><mml:math id="M1"><mml:mrow><mml:msub><mml:mrow><mml:mi>c</mml:mi></mml:mrow><mml:mrow><mml:mi>n</mml:mi></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:munderover><mml:mrow><mml:mi mathvariant="normal">Σ</mml:mi></mml:mrow><mml:mrow><mml:mi>k</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn></mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:munderover><mml:mi>log</mml:mi><mml:mo></mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi>S</mml:mi></mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:msub><mml:mo>)</mml:mo><mml:mi>cos</mml:mi><mml:mo></mml:mo><mml:mo>[</mml:mo><mml:mi>i</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>k</mml:mi><mml:mo>−</mml:mo><mml:mfrac><mml:mrow><mml:mn>1</mml:mn></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:mfrac><mml:mo>)</mml:mo><mml:mfrac><mml:mrow><mml:mi>π</mml:mi></mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mfrac></mml:mrow><mml:mo>]</mml:mo></mml:mrow><mml:mo>,</mml:mo><mml:mi>i</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn><mml:mo>,</mml:mo><mml:mn>2</mml:mn><mml:mo>…</mml:mo><mml:mn>.</mml:mn><mml:mo>,</mml:mo><mml:mi>M</mml:mi></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq1.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula><p>where K is the number of filterbanks, which is 20 in this study. M is the length of the cepstrum that is chosen to be 13, and S<sub>k</sub> represents the energy output of the kth triangular band pass filter.</p></sec><sec id="sec4.2"><title>4.2 Acoustic features</title><p>The set of acoustic parameters used in our study is presented in <xref rid="t0002" ref-type="table">Table 2</xref>. The first five characteristics were investigated previously in [<xref rid="cit0019" ref-type="bibr">19</xref>]. The automatic estimation of the remaining features are described below.</p><table-wrap id="t0002" orientation="portrait" position="float"><label>Table 2</label><caption><p>Studied cry characteristics</p></caption><table frame="hsides" rules="groups"><thead valign="bottom"><tr><th align="left" rowspan="1" colspan="1">Characteristics</th><th align="left" rowspan="1" colspan="1">Definitions.</th></tr></thead><tbody valign="bottom"><tr><td align="left" rowspan="1" colspan="1">Fundamental frequency (F<sub>0</sub>)</td><td align="left" rowspan="1" colspan="1">Vibration frequency (in Hz) of the vocal folds.</td></tr><tr><td align="left" rowspan="1" colspan="1">F<sub>0</sub> Harmonics</td><td align="left" rowspan="1" colspan="1">Multiples of fundamental frequency.</td></tr><tr><td align="left" rowspan="1" colspan="1">RF<sub>1</sub>, RF<sub>2</sub> (Hz)</td><td align="left" rowspan="1" colspan="1">The first and second vocal tract resonance frequencies.</td></tr><tr><td align="left" rowspan="1" colspan="1">Tuning (TUP)</td><td align="left" rowspan="1" colspan="1">Blocks of 1024 samples during which an RF remains close to the harmonics of F<sub>0</sub> (distance < 100 Hz).</td></tr><tr><td align="left" rowspan="1" colspan="1">Transition (TRP)</td><td align="left" rowspan="1" colspan="1">Blocks of 1024 samples between two subsequent TUPs.</td></tr><tr><td align="left" rowspan="1" colspan="1">F<sub>0glide</sub></td><td align="left" rowspan="1" colspan="1">Rapid increase or decrease in F0 of 600 Hz or more.</td></tr><tr><td align="left" rowspan="1" colspan="1">Glide duration (T<sub>glide</sub>)</td><td align="left" rowspan="1" colspan="1">Defined by the start time and finish time of F0 glide.</td></tr><tr><td align="left" rowspan="1" colspan="1">Dysregulation of RFs (RFs<sub>dys</sub>)</td><td align="left" rowspan="1" colspan="1">High frequency variability of RFs (RF<sub>1</sub>, RF<sub>2</sub>).</td></tr></tbody></table></table-wrap><sec id="sec4.2.1"><title>4.2.1 Features description</title><p>Research in acoustic phonetics was focused on the mechanisms of speech production rather than cry production in newborns. Moreover, the acoustic properties of vocal folds and many aspects of source-filter interaction have not been clearly defined in newborn cries. Therefore, we have investigated the acoustic properties of speech and singing to understand the F<sub>0glide</sub> and RFs<sub>dys</sub> phenomena in cry signals.</p><p>In vowel and voiced consonant production theory, the glottis operates independently of the vocal tract [<xref rid="cit0020" ref-type="bibr">20</xref>]. This theory concerns male speech more than female and child speeches. However, glide production is influenced by the interaction between the vocal tract and vocal folds [<xref rid="cit0021" ref-type="bibr">21</xref>]. When the oscillation frequency of the vocal fold F<sub>0</sub> approaches the vocal tract resonance RFs (RF<sub>1</sub>, RF<sub>2</sub>), a nonlinear source-filter system coupling occurs. This interaction is more important for female and child speeches and even more for singing [<xref rid="cit0022" ref-type="bibr">22</xref>]. It has been mentioned [<xref rid="cit0023" ref-type="bibr">23</xref>] that the rising and falling F0 occur with the upward and downward movement of the larynx respectively, and that the control mechanisms of F0 are influenced by the movements of the laryngeal frame and cervical spine.</p><p>Based on the average pressure decrease across a narrow constriction in normal voice, Stevens [<xref rid="cit0024" ref-type="bibr">24</xref>] defined glides as a class of consonants with a constriction that is not sufficiently narrow. Thus, the glides were produced with a greater degree of constriction in the vocal tract, and their categories were differentiated by the aerodynamics of different degrees of vocal tract constriction.</p><p>In cry signals, F<sub>0glide</sub> was defined as a rapid change in F<sub>0</sub> (> 600 Hz) in 0.1 s. It can be either rising or falling glides [<xref rid="cit0008" ref-type="bibr">8</xref>]. This characteristic is shown in <xref ref-type="fig" rid="f0001">Figure 1 (a)</xref> that represents a spectrogram of a hyperphonic and dysphonic cry for a newborn suffering from asphyxia. We can observe the upward glide of 1102-2390 Hz at 0.053-0.210 s from the F0 contour of this cry, and thus the duration of glide (0.157 s) is higher than 0.1 s.</p><fig id="f0001" orientation="portrait" position="float"><label>Figure 1</label><caption><p>Spectrograms of cry signals using PRAAT [<xref rid="cit0029" ref-type="bibr">29</xref>]. a) Rising F<sub>0glide</sub> in the cry of full-term infant suffering from asphyxia. b) RF<sub>1dys</sub> between two successive TUPs in the cry of preterm infant with gastroschisis.</p></caption><graphic xlink:href="BSPC-50-035-g001"></graphic></fig><p>Some studies have investigated the relationship between the glottal properties and vocal tract acoustics of an opera singer [<xref rid="cit0025" ref-type="bibr">25</xref>, <xref rid="cit0026" ref-type="bibr">26</xref>, <xref rid="cit0027" ref-type="bibr">27</xref>, and <xref rid="cit0028" ref-type="bibr">28</xref>]. According to [<xref rid="cit0025" ref-type="bibr">25</xref>], the increase or decrease in resonance frequencies is related to the possibility of changes in the configuration of organs involved in voice production as follows: glottal-opening duration, glottal vibratory amplitude, glottal area, laryngeal height, mouth opening, and tongue shape. Because the tuning (TUP) phenomenon is defined as an adjustment of the RF such that it is close to F<sub>0</sub> or one of its harmonics, it has been shown that professional singers use many resonance tuning strategies for vowels (RF<sub>1</sub>: 2F<sub>0</sub>, RF<sub>1</sub>: 3F<sub>0</sub>, RF<sub>2</sub>: 4F<sub>0</sub>, RF<sub>2</sub>: 5F<sub>0</sub>, RF<sub>2</sub>: 6F<sub>0</sub>, and RF<sub>2</sub>: 8F<sub>0</sub> tuning) [<xref rid="cit0025" ref-type="bibr">25</xref>].</p><p>In our previous work [<xref rid="cit0019" ref-type="bibr">19</xref>], an automated approach has been used to estimate and evaluate the duration and percentage of TUP between RFs and the harmonics of F0 for healthy and pathologic newborn cries. The obtained results encouraged us to further investigate the variability of RFs, termed RFs<sub>dys</sub> here. Furthermore, this characteristic has been associated with neonatal diseases [<xref rid="cit0010" ref-type="bibr">10</xref>]. However, to our best knowledge, none have evaluated the variation pattern of RFs<sub>dys</sub> whether in cry signals or in speech signals.</p><p>To quantify RFs<sub>dys</sub> and analyze their variation patterns according to the studied pathologies, we define this characteristic by the RF jumps between two successive TUPs. Therefore, during the transition period TRP as shown in <xref ref-type="fig" rid="f0001">Figure 1 (b)</xref>, we observed that RF<sub>1</sub> suddenly jumped downward from 3178 Hz to attain 2196 Hz at t = 0.031 s. Because F<sub>0</sub> = 450 Hz, this jump occurred between two successive tunings with the 7th and 5th harmonics of F<sub>0</sub>, (RF<sub>1</sub>:7F<sub>0</sub> and RF<sub>1</sub>:5F<sub>0</sub>). Thus, in this case, the jump “J” is equal to two.</p></sec><sec id="sec4.2.2"><title>4.2.2 Feature estimation</title><p>The adopted approach for the estimation of F<sub>0glide</sub>, T<sub>glide</sub>, and RFs<sub>dys</sub> is illustrated in <xref ref-type="fig" rid="f0002">Figure 2</xref>, and the principal steps are detailed below.</p><fig id="f0002" orientation="portrait" position="float"><label>Figure 2</label><caption><p>Flowchart of adopted approach.</p></caption><graphic xlink:href="BSPC-50-035-g002"></graphic></fig><p>In the first step, cry signals were processed manually using PRAAT (a freeware program for the analysis and reconstruction of acoustic speech signals) [<xref rid="cit0029" ref-type="bibr">29</xref>]. The automated preprocessing method is currently being investigated by other researchers studying the PICS [<xref rid="cit0030" ref-type="bibr">30</xref>]. Because the recorded sounds include the background noises, speech, sound of medical equipment, and silence, some of these parts may distort the results of the analysis. Therefore, this step consists of noise filtering and segmentation of recordings into useful and non-useful segments. Subsequently, the database was organized by pathology and gestational age.</p><p>In the second step, we applied the following procedure for acoustic characteristic measurement using self-developed functions in Matlab. This has been performed for healthy and pathologic cries by pathology and gestational age. Each cry segment of 1-s duration was divided into 50% overlapping frames of 1024 samples. Each frame was multiplied by the Hamming window. Thereafter, for each data frame, we estimated F0 and its 10 harmonics using the simple inverse filtering tracking algorithm [<xref rid="cit0031" ref-type="bibr">31</xref>, <xref rid="cit0032" ref-type="bibr">32</xref>]; further, we estimated the RFs (RF<sub>1</sub>, RF<sub>2</sub>) using the modified covariance method based on the autoregressive power spectral density (AR-PSD) [<xref rid="cit0033" ref-type="bibr">33</xref>, <xref rid="cit0034" ref-type="bibr">34</xref>]. The primary steps of the estimation algorithm of F<sub>0</sub> and RFs are detailed in [<xref rid="cit0018" ref-type="bibr">18</xref>]. The performances of these algorithms were tested on a real newborn cry database [<xref rid="cit0016" ref-type="bibr">16</xref>, <xref rid="cit0032" ref-type="bibr">32</xref>]. The derived function M(t) was estimated using F<sub>0</sub> contour measurements and subsequently used for F<sub>0glide</sub> detection.</p><p>The TUPs and the jumps of RFs were estimated using the harmonics of F<sub>0</sub> and RFs measurements. In contrast to a previous study [<xref rid="cit0035" ref-type="bibr">35</xref>], where the TUPs were investigated using a special visualization concept, the TUPs in our work were detected automatically according to the given definitions in <xref rid="t0002" ref-type="table">Table 2</xref> and subsequently used for RFs<sub>dys</sub> estimation.</p><p>For each cry sample, F<sub>0glides</sub>, T<sub>glide</sub>, and RFs<sub>dys</sub> (RF<sub>1dys</sub>, RF<sub>2dys</sub>) were calculated using the adopted algorithms detailed below.</p><sec id="sec4.2.2.1"><title>4.2.2.1 F<sub>0glide</sub> and T<sub>glide</sub> estimation</title><p>The following approach was adopted in this work to detect F<sub>0glide</sub> and its corresponding T<sub>glide</sub> in 1-s cry segments:</p><list list-type="simple"><list-item><label>(1)</label><p>Estimate F<sub>0</sub> and obtain the derived function M(t) of F<sub>0</sub>.</p><p><disp-formula id="eq2"><alternatives><mml:math id="M2"><mml:mrow><mml:mtext>M</mml:mtext><mml:mrow><mml:mo>(</mml:mo><mml:mtext>t</mml:mtext><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mtext>dF</mml:mtext></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mtext>t</mml:mtext><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mrow><mml:mtext>dt</mml:mtext></mml:mrow></mml:mfrac></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq2.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p></list-item><list-item><label>(2)</label><p>Obtain the time indexes “<italic>x</italic>” when M (t) is equal to zero.</p><p>M(<italic>x</italic>) = 0</p></list-item><list-item><label>(3)</label><p>Calculate the corresponding F<sub>0</sub>(<italic>x</italic>).</p></list-item><list-item><label>(4)</label><p>Identify the glides of F<sub>0</sub> by the following condition:
<disp-formula id="eq3"><alternatives><mml:math id="M3"><mml:mrow><mml:mo>|</mml:mo><mml:msub><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>i</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo><mml:mo>−</mml:mo><mml:msub><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>i</mml:mi><mml:mo>+</mml:mo><mml:mn>1</mml:mn><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mo>|</mml:mo><mml:mo>≥</mml:mo><mml:mn>600</mml:mn><mml:mi>H</mml:mi><mml:mi>z</mml:mi><mml:mo>,</mml:mo><mml:mi>i</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mo>…</mml:mo><mml:mn>.</mml:mn><mml:mi>N</mml:mi><mml:mo>−</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq3.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p><italic>N:</italic> number of zero values at the function M(t)</p></list-item><list-item><label>(5)</label><p>Obtain the duration T<sub>glide</sub> of each F<sub>0glide</sub>, by the following formula:
<disp-formula id="eq4"><alternatives><mml:math id="M4"><mml:mrow><mml:msub><mml:mrow><mml:mtext>T</mml:mtext></mml:mrow><mml:mrow><mml:mtext>glide</mml:mtext></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:mtext>T</mml:mtext><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>i</mml:mi><mml:mo>+</mml:mo><mml:mn>1</mml:mn><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>−</mml:mo><mml:mtext>T</mml:mtext><mml:mrow><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mi>i</mml:mi><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq4.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p>T(<italic>x</italic>): the times at index « <italic>x</italic>».</p></list-item><list-item><label>(6)</label><p>Calculate the average percentage of F<sub>0glide</sub> (P<sub>F0glide</sub>) and average of T<sub>glide</sub> (A<sub>Tglide</sub>) by pathology and gestational age, using these formulas:
<disp-formula id="eq5"><alternatives><mml:math id="M5"><mml:mrow><mml:mtable><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mtext>P</mml:mtext></mml:mrow><mml:mrow><mml:mtext>F</mml:mtext><mml:mn>0</mml:mn><mml:mtext>glide</mml:mtext></mml:mrow></mml:msub></mml:mtd><mml:mtd><mml:mo>=</mml:mo></mml:mtd><mml:mtd><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mtext>N</mml:mtext></mml:mrow><mml:mrow><mml:mtext>glide</mml:mtext></mml:mrow></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mtext>N</mml:mtext></mml:mrow><mml:mrow><mml:mtext>Total</mml:mtext></mml:mrow></mml:msub></mml:mrow></mml:mfrac><mml:mo>,</mml:mo></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mtext>A</mml:mtext></mml:mrow><mml:mrow><mml:mtext>Tglide</mml:mtext></mml:mrow></mml:msub></mml:mtd><mml:mtd><mml:mo>=</mml:mo></mml:mtd><mml:mtd><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">Σ</mml:mi><mml:mtext>T</mml:mtext></mml:mrow><mml:mrow><mml:mtext>glide</mml:mtext></mml:mrow></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mtext>N</mml:mtext></mml:mrow><mml:mrow><mml:mtext>glide</mml:mtext></mml:mrow></mml:msub></mml:mrow></mml:mfrac></mml:mtd></mml:mtr></mml:mtable></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq5.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p>N<sub>Total</sub> : Total number of segments in cry samples of 1-s duration.</p><p>N<sub>glide</sub>: Number of F<sub>0glide</sub> that occurred.</p></list-item><list-item><label>(7)</label><p>Calculate the median and interquartile range of the evaluated characteristics P<sub>F0glide</sub> and A<sub>Tglide</sub>.</p></list-item></list><p>Herein, we did not consider the time constraint of 0.1 s for the durations of F<sub>0glide</sub>. Thus, T<sub>glide</sub>, which could be of any length, was estimated and analyzed according to the pathologies.</p></sec><sec id="sec4.2.2.2"><title>4.2.2.2 RFs<sub>dys</sub> estimation</title><p>The estimation of RFs<sub>dys</sub> was performed by computing the jump “J” of the RFs between two subsequent TUPs or during the transition duration (TRP). “Jump” represents the number of harmonics between two subsequent TUPs. This study examines the Js in the range of 1 to 9. In our work, the TUPs and TRPs were detected automatically and used for the evaluation of RFs<sub>dys</sub>. RFs<sub>dys</sub> were separately investigated for RF<sub>1dys</sub> and RF<sub>2dys</sub> with the same procedure for each data frame in 1-s cry segments as follow:</p><list list-type="simple"><list-item><label>(1)</label><p>Estimate F<sub>0</sub> and its 10 harmonics.</p></list-item><list-item><label>(2)</label><p>Estimate RF<sub>1</sub> and RF<sub>2</sub>.</p></list-item><list-item><label>(3)</label><p>Identify TUPs separately for RF<sub>1</sub> and RF<sub>2</sub> by the following conditions.<disp-formula id="eq6"><alternatives><mml:math id="M6"><mml:mrow><mml:mtable><mml:mtr><mml:mtd><mml:mo>|</mml:mo><mml:mtext>n</mml:mtext><mml:mn>.</mml:mn><mml:msub><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>p</mml:mi><mml:mo>)</mml:mo><mml:mo>−</mml:mo><mml:msub><mml:mrow><mml:mtext>RF</mml:mtext></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>p</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mo>|</mml:mo><mml:mo>≤</mml:mo><mml:mn>100</mml:mn><mml:mo>,</mml:mo><mml:mi>n</mml:mi><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>,</mml:mo><mml:mo>…</mml:mo><mml:mn>9</mml:mn><mml:mo>]</mml:mo><mml:mo>,</mml:mo><mml:mi>p</mml:mi><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mn>..</mml:mn><mml:msub><mml:mrow><mml:mtext>N</mml:mtext></mml:mrow><mml:mrow><mml:mi>T</mml:mi><mml:mi>o</mml:mi><mml:mi>t</mml:mi><mml:mi>a</mml:mi><mml:mi>l</mml:mi></mml:mrow></mml:msub><mml:mo>]</mml:mo></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:mo>|</mml:mo><mml:mtext>n</mml:mtext><mml:mn>.</mml:mn><mml:msub><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>p</mml:mi><mml:mo>)</mml:mo><mml:mo>−</mml:mo><mml:msub><mml:mrow><mml:mtext>RF</mml:mtext></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mi>p</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow><mml:mo>|</mml:mo><mml:mo>≤</mml:mo><mml:mn>100</mml:mn><mml:mo>,</mml:mo><mml:mi>n</mml:mi><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>,</mml:mo><mml:mo>…</mml:mo><mml:mn>9</mml:mn><mml:mo>]</mml:mo><mml:mo>,</mml:mo><mml:mi>p</mml:mi><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mn>..</mml:mn><mml:msub><mml:mrow><mml:mtext>N</mml:mtext></mml:mrow><mml:mrow><mml:mi>T</mml:mi><mml:mi>o</mml:mi><mml:mi>t</mml:mi><mml:mi>a</mml:mi><mml:mi>l</mml:mi></mml:mrow></mml:msub><mml:mo>]</mml:mo></mml:mtd></mml:mtr></mml:mtable></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq6.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p>n: order of harmonic (1 to 9).</p><p><italic>p:</italic> index of frame in cry segment</p></list-item><list-item><label>(4)</label><p>Find the time indexes of the start and the end of TRPs, i.e., (s<sub>1</sub>, e<sub>1</sub>) and (s<sub>2</sub>, e<sub>2</sub>), for RF<sub>1</sub> and RF<sub>2</sub>, respectively.</p></list-item><list-item><label>(5)</label><p>Calculate the corresponding RF<sub>1</sub> (s<sub>1</sub>), RF<sub>1</sub> (e<sub>1</sub>), RF<sub>2</sub> (s<sub>2</sub>), and RF<sub>2</sub> (e<sub>2</sub>).</p></list-item><list-item><label>(6)</label><p>Obtain the jumps J<sub>1</sub> and J<sub>2</sub> for RF<sub>1</sub> and RF<sub>2</sub> ,respectively, using the following formulas, respectively:
<disp-formula id="eq7"><alternatives><mml:math id="M7"><mml:mrow><mml:mtable><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mi mathvariant="normal">F</mml:mi><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mi mathvariant="normal">R</mml:mi><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi>s</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>−</mml:mo><mml:mi mathvariant="normal">R</mml:mi><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi>e</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>×</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mo>…</mml:mo><mml:mn>..</mml:mn><mml:msub><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn><mml:mi>T</mml:mi><mml:mi>R</mml:mi><mml:mi>P</mml:mi></mml:mrow></mml:msub><mml:mo>]</mml:mo></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mi mathvariant="normal">F</mml:mi><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mi mathvariant="normal">R</mml:mi><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi>s</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>−</mml:mo><mml:mi mathvariant="normal">R</mml:mi><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi>e</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>×</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn>0</mml:mn></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mi mathvariant="normal">z</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mo>…</mml:mo><mml:mn>..</mml:mn><mml:msub><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn><mml:mi>T</mml:mi><mml:mi>R</mml:mi><mml:mi>P</mml:mi></mml:mrow></mml:msub><mml:mo>]</mml:mo></mml:mtd></mml:mtr></mml:mtable></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq7.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p>where C<sub>RF1</sub>, C<sub>RF2</sub> are the differences in RFs, i.e., RF<sub>1</sub>, RF<sub>2</sub>, respectively, during the TRPs. <italic>N<sub>1TRP</sub></italic>, <italic>N<sub>2TRP</sub></italic> represent the number of TRPs for RF<sub>1</sub>, RF<sub>2</sub>, respectively.</p><p>Thereafter, for each jump J<sub>1</sub> and J<sub>2</sub> in the range of 1 to 9, the average percentages of RF<sub>1dys</sub> (P<sub>RF1dys</sub>) and of RF<sub>2dys</sub> (P<sub>RF2dys</sub>) were estimated for each pathology and gestational age, using these formulas, respectively:
<disp-formula id="eq8"><alternatives><mml:math id="M8"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">P</mml:mi></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mi mathvariant="normal">d</mml:mi><mml:mi mathvariant="normal">y</mml:mi><mml:mi mathvariant="normal">s</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">N</mml:mi></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msub></mml:mrow></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">N</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">l</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:mfrac><mml:mo>,</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>1</mml:mn></mml:mrow></mml:msub><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mn>..9</mml:mn><mml:mo>]</mml:mo><mml:mo>,</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">P</mml:mi></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mi mathvariant="normal">d</mml:mi><mml:mi mathvariant="normal">y</mml:mi><mml:mi mathvariant="normal">s</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">N</mml:mi></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msub></mml:mrow></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">N</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">l</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:mfrac><mml:mo>,</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="normal">J</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo>∈</mml:mo><mml:mo>[</mml:mo><mml:mn>1</mml:mn><mml:mo>…</mml:mo><mml:mn>..9</mml:mn><mml:mo>]</mml:mo><mml:mo>,</mml:mo></mml:mrow></mml:math><graphic xlink:href="BSPC-50-035-eq8.jpg" position="float" orientation="portrait"></graphic></alternatives></disp-formula></p><p>N<sub>J1</sub>, N<sub>J2</sub> are the number of jumps J<sub>1</sub> and J<sub>2</sub>, respectively, in cry samples of 1-s duration.</p><p>N<sub>Total</sub>: Total number of segments in cry samples of 1-s duration.</p></list-item><list-item><label>(7)</label><p>Calculate the median and interquartile range of the evaluated characteristics P<sub>RF1dys</sub> and P<sub>RF2dys</sub>.</p></list-item></list></sec></sec></sec></sec><sec sec-type="description" id="sec5"><title>5. PCIS description</title><p>The proposed diagnostic system was built around an efficient PNN classifier and significant infant cry features. In this study, we investigate the effectiveness of the extracted characteristics (P<sub>F0glide</sub>+A<sub>Tglide</sub>+P<sub>RF1dys</sub>+P<sub>RF2dys</sub>) concatenated with MFCC features for the classification of healthy and sick newborn infants according to their cries. Thus, the sets of characteristics once obtained were used as inputs to the PNN classifier.</p><p>The PNN classifier is used widely for classification problems in the medical domain [<xref rid="cit0036" ref-type="bibr">36</xref>, <xref rid="cit0037" ref-type="bibr">37</xref>]. It is ideal for real-time applications and is computationally inexpensive. It can learn using the conjugate gradient method, new incoming training data without having to repeat the whole training process [<xref rid="cit0038" ref-type="bibr">38</xref>]. The PNN classifier implements the Bayesian decision rule. Its network architecture is based on three layers: (1) the input layer that computes the distances between the input vector and the training input; (2) radial basis layer that produces a vector of probabilities; (3) competitive layer that selects the maximum of these probabilities. The network classifies the input vector, the assigned class as those with the highest probability [<xref rid="cit0036" ref-type="bibr">36</xref>].</p><p>Matlab was used for the development of the test system. The primary steps are as follows:</p><list list-type="bullet"><list-item><p>wav files recovery from each folder of the studied pathologies.</p></list-item><list-item><p>division of 1-s cry signals into overlapping frames of 1024 samples, with 50% recovery.</p></list-item></list><p>For each 1-s cry segment, the following were performed:</p><list list-type="bullet"><list-item><p>estimation of P<sub>F0glide</sub>, AT<sub>glide</sub>, P<sub>RF1dys</sub>, and P<sub>RF2dys</sub>.</p></list-item><list-item><p>extraction of 13 MFCCs for each interlaced frame.</p></list-item><list-item><p>conversion of each matrix of the MFCCs specific to a cry sample into only one vector.</p></list-item><list-item><p>concatenation of each vector with its corresponding P<sub>F0glide</sub>, AT<sub>glide</sub>, P<sub>RF1dys</sub>, and P<sub>RF2dys</sub>. The set of obtained vectors were selected as the final inputs to the PNN classifier.</p></list-item><list-item><p>application of five-fold cross validation; four folds for training set and one fold for testing set.</p></list-item><list-item><p>data classification using the PNN classifier; the strategy proposed by [<xref rid="cit0036" ref-type="bibr">36</xref>] paper and the function newpnn () in Matlab [<xref rid="cit0039" ref-type="bibr">39</xref>] were used.</p></list-item><list-item><p>calculation of the correct identification rate namely, accuracy.</p></list-item></list></sec><sec sec-type="results" id="sec6"><title>6. Results and discussion</title><sec id="sec6.1"><title>6.1 Acoustic and statistical analysis</title><p>The cry signals were analyzed using the studied characteristics (P<sub>F0glide</sub>, A<sub>Tglide</sub>, P<sub>RF1dys</sub>, and P<sub>RF2dys</sub>). Moreover, they have been investigated according to the gestational age (full-term and preterm) and diseases afflicting newborns. The cry features given in <xref rid="t0002" ref-type="table">Table 2</xref> were estimated for the complete database presented in <xref rid="t0001" ref-type="table">Table 1</xref>.</p><p>An example of F<sub>0glide</sub> is shown in <xref ref-type="fig" rid="f0003">Figure 3</xref>; it represents an estimate of F<sub>0</sub> for a 1-s cry of a newborn suffering from asphyxia. The irregular contour of F0 presents the rising glide between 0.21 s and 0.24 s and the falling glide between 0.58 s and 0.63 s.</p><fig id="f0003" orientation="portrait" position="float"><label>Figure 3</label><caption><p>Estimated F<sub>0</sub> for a cry signal of a newborn suffering from asphyxia; rising and falling F<sub>0glide</sub> (red ellipses).</p></caption><graphic xlink:href="BSPC-50-035-g003"></graphic></fig><p>The average percentage of F<sub>0glide</sub> (P<sub>F0glide</sub>) and the average durations of the glides (AT<sub>glide</sub>) are given in <xref rid="t0003" ref-type="table">Table 3</xref> and presented by pathology and gestational age in <xref ref-type="fig" rid="f0004">Figures 4(a)</xref> and <xref ref-type="fig" rid="f0004">4(b)</xref>, respectively.</p><fig id="f0004" orientation="portrait" position="float"><label>Figure 4</label><caption><p>Box-and-whiskers plots for a) P<sub>F0glide</sub>, b)T<sub>glide</sub>, by pathology.</p></caption><graphic xlink:href="BSPC-50-035-g004"></graphic></fig><table-wrap id="t0003" orientation="portrait" position="float"><label>Table 3</label><caption><p>P<sub>F0glide</sub>, and AT<sub>glide</sub> estimation results</p></caption><table frame="hsides" rules="groups"><thead valign="bottom"><tr><th align="left" rowspan="2" colspan="1">Pathology</th><th align="center" colspan="2" rowspan="1">P<sub>F0glide</sub> (%)</th><th colspan="2" align="center" rowspan="1">AT<sub>glide</sub> (s)</th></tr><tr><th align="left" rowspan="1" colspan="1"><italic>Median</italic></th><th align="left" rowspan="1" colspan="1"><italic>Interquartile Range</italic></th><th align="left" rowspan="1" colspan="1"><italic>Median</italic></th><th align="left" rowspan="1" colspan="1"><italic>Interquartile Range</italic></th></tr></thead><tbody valign="bottom"><tr><td align="left" rowspan="1" colspan="1">Healthy (t)</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td></tr><tr><td align="left" rowspan="1" colspan="1">Hyperbiluribinemia(t)</td><td align="center" rowspan="1" colspan="1">8.33</td><td align="center" rowspan="1" colspan="1">3.9</td><td align="center" rowspan="1" colspan="1">0.16</td><td align="center" rowspan="1" colspan="1">0.10</td></tr><tr><td align="left" rowspan="1" colspan="1">Vena cava thrombosis (t)</td><td align="center" rowspan="1" colspan="1">7.14</td><td align="center" rowspan="1" colspan="1">4.87</td><td align="center" rowspan="1" colspan="1">0.25</td><td align="center" rowspan="1" colspan="1">0.11</td></tr><tr><td align="left" rowspan="1" colspan="1">Meningitis (t)</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0.00</td></tr><tr><td align="left" rowspan="1" colspan="1">Peritonitis (t)</td><td align="center" rowspan="1" colspan="1">12.5</td><td align="center" rowspan="1" colspan="1">6.52</td><td align="center" rowspan="1" colspan="1">0.22</td><td align="center" rowspan="1" colspan="1">0.12</td></tr><tr><td align="left" rowspan="1" colspan="1">Asphyxia (t)</td><td align="center" rowspan="1" colspan="1">8</td><td align="center" rowspan="1" colspan="1">4.17</td><td align="center" rowspan="1" colspan="1">0.15</td><td align="center" rowspan="1" colspan="1">0.09</td></tr><tr><td align="left" rowspan="1" colspan="1">Lingual frenum (t)</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0.00</td></tr><tr><td align="left" rowspan="1" colspan="1">Healthy (P)</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0</td><td align="center" rowspan="1" colspan="1">0.00</td></tr><tr><td align="left" rowspan="1" colspan="1">IUGR-microcephaly (P)</td><td align="center" rowspan="1" colspan="1">18.46</td><td align="center" rowspan="1" colspan="1">12.5</td><td align="center" rowspan="1" colspan="1">0.28</td><td align="center" rowspan="1" colspan="1">0.10</td></tr><tr><td align="left" rowspan="1" colspan="1">Tetralogy of Fallot (P)</td><td align="center" rowspan="1" colspan="1">10</td><td align="center" rowspan="1" colspan="1">4.91</td><td align="center" rowspan="1" colspan="1">0.31</td><td align="center" rowspan="1" colspan="1">0.17</td></tr><tr><td align="left" rowspan="1" colspan="1">Gastroschisis (P)</td><td align="center" rowspan="1" colspan="1">16.66</td><td align="center" rowspan="1" colspan="1">8</td><td align="center" rowspan="1" colspan="1">0.27</td><td align="center" rowspan="1" colspan="1">0.11</td></tr><tr><td align="left" rowspan="1" colspan="1">IUGR-asphyxia (P)</td><td align="center" rowspan="1" colspan="1">11.11</td><td align="center" rowspan="1" colspan="1">4.95</td><td align="center" rowspan="1" colspan="1">0.23</td><td align="center" rowspan="1" colspan="1">0.12</td></tr><tr><td align="left" rowspan="1" colspan="1">RDS (P)</td><td align="center" rowspan="1" colspan="1">10</td><td align="center" rowspan="1" colspan="1">5.71</td><td align="center" rowspan="1" colspan="1">0.18</td><td align="center" rowspan="1" colspan="1">0.11</td></tr></tbody></table></table-wrap><p>The obtained results indicate that the cry signals do not present the glides of F0 in both healthy premature and full-term newborns, and also in the studied cases of lingual frenum and meningitis. They also indicate that, in a group of full-term newborns, the highest median and interquartile range of PF0glide are found in peritonitis with average glide durations compared to other diseases. The lowest percentage is found in the case of vena cava thrombosis with longer durations of F0glide. In the same group of full-term newborns, a small difference occurred in the occurrence of F0glide and also ATglide between the cries of hyperbilirubinemia and asphyxia diseases.</p><p>The highest median and interquartile ranges of PF<sub>0glide</sub> among the two groups of newborns were observed in the cries of preterm newborns affected by IUGR-microcephaly disease with rather long periods of glides. However, <xref ref-type="fig" rid="f0004">Figure 4</xref> and <xref rid="t0003" ref-type="table">Table 3</xref> show that the highest median and interquartile range of A<sub>Tglide</sub> characterize the tetralogy of Fallot disease, which presents the same P<sub>F0glide</sub> as RDS disease compared to other pathologies.</p><p>Concerning the RFs<sub>dys</sub>, examples of this feature are shown in <xref ref-type="fig" rid="f0005">Figure 5</xref>, which represents the estimates of F<sub>0</sub>, their harmonics, RF<sub>1</sub>, and RF<sub>2</sub> for the 1-s cry of a newborn suffering from hyperbilirubinemia. An RF<sub>1dys</sub> was observed at approximately 0.38 s between the third and sixth harmonics, “J = 3” and also an RF<sub>2dys</sub> at approximately 0.26 s between the seventh and ninth harmonics “J = 2”.</p><fig id="f0005" orientation="portrait" position="float"><label>Figure 5</label><caption><p>Full-term newborn suffering from hyperbilirubinemia, F<sub>0</sub> and their harmonics (red lines), RFs: RF<sub>1</sub> (black dots) and RF2 (blue dots), RFs<sub>dys</sub> between TUPs (green rectangles).</p></caption><graphic xlink:href="BSPC-50-035-g005"></graphic></fig><p>Comparative results are presented in <xref ref-type="fig" rid="f0006">Figure 6</xref>. They exhibit the variation patterns of RFs<sub>dys</sub> by pathology. This dysregulation is expressed by the percentage of jump “J” previously defined in Section 3.4. <xref ref-type="fig" rid="f0006">Figures 6(a) and (b)</xref> show that the majority of cries of preterm newborns are characterized by more RF<sub>1dys</sub> when J = 1 than cries of full-term newborns. The percentage of RF<sub>1</sub> jumps is more important for all jumps J = 1 to 9 in the cries of healthy preterm newborns than in the cries of healthy full-term newborns.</p><fig id="f0006" orientation="portrait" position="float"><label>Figure 6</label><caption><p>Average percentage of RFs<sub>dys</sub> by jumps “J”. a) RF<sub>1dys</sub> for full-term newborns, b) RF<sub>1dys</sub> for preterm newborns, c) RF<sub>2dys</sub> for full-term newborns, d) RF<sub>2dys</sub> preterm newborns.</p></caption><graphic xlink:href="BSPC-50-035-g006"></graphic></fig><p>According to <xref ref-type="fig" rid="f0006">Figure 6(a)</xref>, the cry signals of vena cava thrombosis and asphyxia diseases present the most RF<sub>1dys</sub>, with the highest percentage of jump for “J” from 1 to 9, except for “J” of 3 and 4 jumps, where the largest percentage is found in cry signals of lingual frenum pathology. In the case of hyperbilirubinemia and for “J” equal to 1, 2, and 3 jumps, the cry signals were qualified by more RF<sub>1dys</sub> than those of healthy newborns. Further, in peritonitis disease, the percentage of jumps for “J” of 1, 2, 5, and 6 is higher than those of healthy newborns. Unlike these latest pathologies, the cry signals in case of meningitis present less RF<sub>1dys</sub> for all jumps “J” (1 to 9) than in healthy newborn. From <xref ref-type="fig" rid="f0006">Figure 6 (b)</xref>, we can observe the highest RF<sub>1dys</sub> for jumps J = 1 to 6 in the case of tetralogy of Fallot disease compared to the other studied cases. We also noticed a higher percentage of RF<sub>1</sub> jumps for J = 1 in both cases of RDS and gastroschisis, and less RF<sub>1dys</sub> was observed in the cry signals of IUGR–microcephaly and IUGR–asphyxia compared to the cries of healthy preterm newborns. The variation patterns of RF<sub>2dys</sub> for full-term and preterm newborn cry signals are shown in <xref ref-type="fig" rid="f0006">Figure 6</xref> (c and d, respectively). We observed that the percentage of RF<sub>2</sub> jumps depends on the studied pathologies. According to the jumps “J”, it can be less or more important than in those in healthy newborn cries. In the cases of full and preterm newborns, all studied diseases are characterized by low difference in the percentage of RF<sub>2</sub> jumps compared to healthy preterm newborn cries, except for the cry signals of tetralogy of Fallot disease that presents a greater percentage of RF<sub>2dys</sub> than the healthy ones for J = 1 to 5 jumps.</p></sec><sec id="sec6.2"><title>6.2 Classification results</title><p>In this work, we present the results of healthy and sick (with specific diseases, see <xref rid="t0001" ref-type="table">Table 1</xref>) infant identification systems. To test and validate the use of the studied characteristics, two experiments were performed in both cases of full-term and preterm infants: 1) separate cries by health status of infant (pathologic cries from healthy ones); 2) separate cries into those of healthy infants and those sick with specific diseases. The effectiveness of the studied characteristics (P<sub>F0glide</sub>, A<sub>Tglide</sub>, P<sub>RF1dys</sub>, and P<sub>RF2dys</sub>) in the recognition of pathological cries was compared to the use of MFCC parameters as the input to the PCIS. The performance of the PCIS was evaluated based on the rate of correct identification (overall accuracy) [<xref rid="cit0004" ref-type="bibr">4</xref>].</p><p>The distribution of cry samples according to classes and pathologies are shown in <xref rid="t0004" ref-type="table">Table 4</xref>. In the first experiment, the cry samples are distributed into two classes: healthy cries and pathological cries. In the second experiment, the cry samples are distributed in five classes according to the used pathologies. There are many healthy samples in our database and limited numbers of cry samples from sick infants. However, the five-fold cross validation has been used, in which each fold is of equal size and contains the same percentage of samples as that of each target class.</p><table-wrap id="t0004" orientation="portrait" position="float"><label>Table 4</label><caption><p>Distribution of cry samples per class</p></caption><table frame="hsides" rules="groups"><thead valign="bottom"><tr><th align="left" rowspan="1" colspan="1">Experiments</th><th align="left" rowspan="1" colspan="1">Number of Classes</th><th align="left" rowspan="1" colspan="1">Gestational age</th><th align="left" rowspan="1" colspan="1">Samples number Per class</th><th align="center" rowspan="1" colspan="1">Classes</th></tr></thead><tbody valign="bottom"><tr><td align="left" rowspan="2" colspan="1">First</td><td align="left" rowspan="2" colspan="1">2 classes</td><td align="left" rowspan="1" colspan="1">Preterm</td><td align="center" rowspan="1" colspan="1">50×5/250</td><td align="center" rowspan="1" colspan="1">Pathologic (Tetralogy of Fallot, Gastroschisis, IUGR-microcephaly, RDS, IUGR-asphyxia) /healthy</td></tr><tr><td align="left" rowspan="1" colspan="1">Full term</td><td align="center" rowspan="1" colspan="1">50×5/250</td><td align="center" rowspan="1" colspan="1">Pathologic (hyperbilirubinemia,lingual frenum, vena cava thrombosis, asphyxia, meningitis) /healthy</td></tr><tr><td align="left" rowspan="2" colspan="1">Second</td><td align="left" rowspan="2" colspan="1">5 classes</td><td align="center" rowspan="1" colspan="1">Preterm</td><td align="center" rowspan="1" colspan="1">50×5</td><td align="center" rowspan="1" colspan="1">Tetralogy of Fallo /gastroschisis /RDS /IUGR-asphyxia/healthy</td></tr><tr><td align="left" rowspan="1" colspan="1">Full-term</td><td align="center" rowspan="1" colspan="1">50×5</td><td align="center" rowspan="1" colspan="1">Hyperbiluribinemia/vena cava thrombosis/asphyxia/meningitis/healthy</td></tr></tbody></table></table-wrap><p>The best results of the correct identification rate obtained for all implemented experiments are shown in <xref rid="t0005" ref-type="table">Table 5</xref>. The use of MFCCs as feature characteristics is tested for the two experiments. The best performance obtained (72%) is found in the identification of full-term infants pathologies. The second test corresponds to the use of the studied features in addition to MFCCs. This test provides a better identification rate (88.71%) of health status of preterm infants.</p><table-wrap id="t0005" orientation="portrait" position="float"><label>Table 5</label><caption><p>Overall accuracy results</p></caption><table frame="hsides" rules="groups"><thead valign="bottom"><tr><th rowspan="2" colspan="1">Experiments</th><th rowspan="2" colspan="1">Classes number</th><th align="center" rowspan="2" colspan="1">Gestational age</th><th align="center" colspan="2" rowspan="1">Features</th></tr><tr><th align="center" rowspan="1" colspan="1"><italic>MFCC</italic></th><th align="center" rowspan="1" colspan="1"><italic>P<sub>F0glide</sub> + AT<sub>glide</sub> + P<sub>RF1dys</sub>+ P<sub>RF2dys</sub> + MFCC</italic></th></tr></thead><tbody valign="bottom"><tr><td align="left" rowspan="2" colspan="1">First</td><td align="left" rowspan="2" colspan="1">2 classes</td><td align="left" rowspan="1" colspan="1">Preterm</td><td align="center" rowspan="1" colspan="1">70.16 %</td><td align="center" rowspan="1" colspan="1">88.71 %</td></tr><tr><td align="left" rowspan="1" colspan="1">Full term</td><td align="center" rowspan="1" colspan="1">60.00 %</td><td align="center" rowspan="1" colspan="1">67.00 %</td></tr><tr><td align="left" rowspan="2" colspan="1">Second</td><td align="left" rowspan="2" colspan="1">5 classes</td><td align="left" rowspan="1" colspan="1">Preterm</td><td align="center" rowspan="1" colspan="1">60.00 %</td><td align="center" rowspan="1" colspan="1">72.00 %</td></tr><tr><td align="left" rowspan="1" colspan="1">Full term</td><td align="center" rowspan="1" colspan="1">72.00 %</td><td align="center" rowspan="1" colspan="1">82.00 %</td></tr></tbody></table></table-wrap><p>According to the comparative results presented in <xref rid="t0005" ref-type="table">Table 5</xref>, the use of the studied characteristics (P<sub>F0glide</sub>, A<sub>Tglide</sub>, P<sub>RF1dys</sub>, and P<sub>RF2dys</sub>) in addition to MFCCs improves the performance of the PCIS compared to the use of MFCCs without the studied characteristics compared to the MFCC features only; in the first experiment, the overall accuracy had improved by 18.5% points for the identification of health status of preterm infants and by 7% points for the identification of the health status of full-term infants. In the second experiment, the overall accuracy improved by 12% points for the identification of preterm infants with a specific disease, and 10% points for the identification of full-term infants with a specific disease.</p></sec></sec><sec sec-type="Conclusion" id="sec7"><title>7. Conclusion</title><p>In this work, we proposed an automated classification method for different cry patterns. We used new features that improved the accuracy of the recognition of pathological cries. This study focused on the following: 1) improvement of PCIS performances using studied features, 2) description of our method for estimating the proposed features, and 3) description of associations between infant medical conditions and the considered characteristics.</p><p>The proposed methods to estimate these features permitted the analysis of a considerable number of pathologic cry signals in a short time, and the extension of the study for other pathologies not considered. Hence, we are recording a larger database with a greater variety of pathologies and more subjects for each pathology.</p><p>We successfully identified the occurrence of the proposed characteristics in cry signals according to the studied pathologies. The glide of F0 was found absent in healthy newborn cry signals and also in some of the pathologies. The obtained results suggested that the prevalence of relatively slow glides increased in sick preterm infants, especially in babies with diseases affecting the CNS such as IUGR-microcephaly. Our results indicated a complex pattern variation of RF<sub>2dys</sub> in the pathologic cry signals compared to that of RF<sub>1dys</sub>.</p><p>From the presented work and results obtained in this study, we conclude that, occurrences of F<sub>glides</sub> and RF<sub>1dys</sub> in addition to other characteristics investigated in our previous works were found to be useful in the diagnosis of the studied pathologies.</p><p>The encouraging classification results indicated the highly discriminatory nature of the proposed features. These characteristics can be explored as inputs to a diagnostic system using other modeling and classification methods to ultimately provide a basis for alerting healthcare workers to intervene. We also expect to improve the results by studying other classifiers.</p><p>The results obtained using the PNN classifier and the proposed acoustic features were not compared with other results of previous studies. It is noteworthy that in this study, cases of full-term and premature babies, both with and without pathologies were investigated, thus increased the complexity of a PCIS significantly. For example, the cry signals of neonatal asphyxiate and healthy babies contain significant differences in the acoustic signals, thus easing the identification process.</p><p>In conclusion, our research underlined the importance of using acoustic features in the task of cry recognition and provided support and additional results to investigations regarding neonatal cries.</p></sec></body><back><ack><title>Acknowledgments</title><p>We thank Dr. Barrington and members of the Neonatology group at Saint-Justine Hospital in Montreal (QC) for helping us collect the infant cry database. This research has been funded by a grant from the Bill & Melinda Gates Foundation through the Grand Challenges Explorations Initiative.</p></ack><ref-list id="references"><title>References</title><ref id="cit0001"><label>[1]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Fausto</surname><given-names>O.</given-names></name>, <name name-style="western"><surname>Galaviz</surname><given-names>R.</given-names></name>, <name name-style="western"><surname>Garcia</surname><given-names>C.A. Reyes</given-names></name></person-group>, <source>Infant cry classification to identify hypo acoustics and asphyxia comparing an evolutionary-neural system with a network system</source>, <publisher-name>Springer-Verlag</publisher-name><publisher-loc>Berlin Heidelberg</publisher-loc><year>2005</year>, <fpage>949</fpage>-<lpage>958</lpage>.</mixed-citation></ref><ref id="cit0002"><label>[2]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Lederman</surname><given-names>D.</given-names></name>, <name name-style="western"><surname>Zmora</surname><given-names>E.</given-names></name>, <name name-style="western"><surname>Hauschildt</surname><given-names>S.</given-names></name>, <name name-style="western"><surname>Stellzig-Eisenhauer</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Wermke</surname><given-names>K.</given-names></name></person-group>, <article-title>Classification of cries of infants with cleft-palate using parallel hidden markov models</article-title>, <source>Med Bio Eng Comput</source>. <year>2008</year>, <fpage>965</fpage>-<lpage>975</lpage>.<pub-id pub-id-type="pmid">18368431</pub-id></mixed-citation></ref><ref id="cit0003"><label>[3]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Hariharan</surname><given-names>M.</given-names></name>, <name name-style="western"><surname>Yaacob</surname><given-names>S.</given-names></name>, <name name-style="western"><surname>Awang</surname><given-names>S. Ardeena awatie</given-names></name></person-group>, <article-title>Pathological infant cry analysis using wavelet packet transform and probabilistic neural network</article-title>, <source>Expert Systems with Applications</source>. <year>2011</year>, <volume>38</volume>, <fpage>15377</fpage>-<lpage>15382</lpage>.</mixed-citation></ref><ref id="cit0004"><label>[4]</label><mixed-citation publication-type="confproc"><person-group person-group-type="author"><name name-style="western"><surname>Kheddache</surname><given-names>Y</given-names></name>, <name name-style="western"><surname>Tadj</surname><given-names>C</given-names></name></person-group>, <article-title>“Newborn’s pathological cry identification system”</article-title>, <conf-name>The 11 th international conference on information sciences, signal processing and their applications</conf-name><year>2012</year> pp <fpage>1051</fpage>-<lpage>1056</lpage></mixed-citation></ref><ref id="cit0005"><label>[5]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Manfredi</surname><given-names>C</given-names></name>, <name name-style="western"><surname>Andrea</surname><given-names>B</given-names></name>, <name name-style="western"><surname>Melino</surname><given-names>D</given-names></name>, <name name-style="western"><surname>Viellevoyec</surname><given-names>R</given-names></name>, <name name-style="western"><surname>Masendu</surname><given-names>K</given-names></name>, <name name-style="western"><surname>Silvia</surname><given-names>O</given-names></name></person-group>, <article-title>Automated detection and classification of basic shapes of newborn cry melody</article-title>, <source>Biomedical Signal Processing and Control</source>, <year>2018</year>, pp <fpage>174</fpage>-<lpage>181</lpage>,</mixed-citation></ref><ref id="cit0006"><label>[6]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>LaGasse</surname><given-names>LL</given-names></name>, <name name-style="western"><surname>Neal</surname><given-names>AR</given-names></name>, <name name-style="western"><surname>Lester</surname><given-names>BM</given-names></name></person-group>, <article-title>Assessment of the infant cry: acoustic cry analysis and parental perception</article-title>, <source>Ment Retard Dev Disabil Res Rev</source>. <year>2005</year>, <volume>11</volume>, <fpage>83</fpage>–<lpage>93</lpage>.<pub-id pub-id-type="pmid">15856439</pub-id></mixed-citation></ref><ref id="cit0007"><label>[7]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Orlandia</surname><given-names>S.</given-names></name>, <name name-style="western"><surname>Bandiniab</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Fiaschia</surname><given-names>F.F.</given-names></name>, <name name-style="western"><surname>Manfredia</surname><given-names>C.</given-names></name></person-group>, <article-title>Testing software tools for newborn cry analysis using synthetic signals</article-title>, <source>Biomedical Signal Processing and Control</source>. <year>2017</year>, pp <fpage>16</fpage>-<lpage>22</lpage>.</mixed-citation></ref><ref id="cit0008"><label>[8]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Wasz-Hockert</surname><given-names>O</given-names></name>, <name name-style="western"><surname>Michelsson</surname><given-names>K</given-names></name>, <name name-style="western"><surname>Lind</surname><given-names>J</given-names></name></person-group><chapter-title>Twenty-five years of scandinavian cry research</chapter-title>, In: <person-group person-group-type="editor"><name name-style="western"><surname>Lester</surname><given-names>BM</given-names></name>, <name name-style="western"><surname>Boukydis</surname><given-names>CFZ</given-names></name></person-group>, eds. <source>Infant crying</source>. <publisher-loc>New York, NY</publisher-loc>: <publisher-name>Plenum</publisher-name>; <year>1985</year>, <fpage>83</fpage>–<lpage>104</lpage></mixed-citation></ref><ref id="cit0009"><label>[9]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Michelson</surname><given-names>K</given-names></name>, <name name-style="western"><surname>Todd de Barra</surname><given-names>H</given-names></name>, <name name-style="western"><surname>Michelson</surname><given-names>O</given-names></name></person-group>, <chapter-title>Sound Spectrographic cry analysis and mothers perception of their infant’s crying</chapter-title> In: <person-group person-group-type="editor"><name name-style="western"><surname>Lewis</surname><given-names>FR</given-names></name></person-group>, ed. <source>Focus on nonverbal communication research</source>. <publisher-loc>New York, NY</publisher-loc>: <publisher-name>Nova Science Publishers</publisher-name><year>2007</year>, pp <fpage>31</fpage>–<lpage>64</lpage>.</mixed-citation></ref><ref id="cit0010"><label>[10]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Lester</surname><given-names>BM</given-names></name>, <name name-style="western"><surname>LaGasse</surname><given-names>LL</given-names></name></person-group><chapter-title>Crying</chapter-title> In: <person-group person-group-type="editor"><name name-style="western"><surname>Haith</surname><given-names>MM</given-names></name>, <name name-style="western"><surname>Benson</surname><given-names>JB</given-names></name></person-group>, eds. <source>Encyclopedia of infant and early childhood development</source>. <publisher-loc>San Diego, CA</publisher-loc>: <publisher-name>Academic Press</publisher-name>; <year>2008</year>, pp <fpage>332</fpage>–<lpage>343</lpage>.</mixed-citation></ref><ref id="cit0011"><label>[11]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Corwin</surname><given-names>Michael J</given-names></name>, <name name-style="western"><surname>Kayne</surname><given-names>Herbert</given-names></name>, <name name-style="western"><surname>Lester</surname><given-names>Barry M</given-names></name>, <name name-style="western"><surname>Sepkoski</surname><given-names>Carol</given-names></name>, <name name-style="western"><surname>McLaughlin</surname><given-names>Sarah</given-names></name> et <name name-style="western"><surname>Golub</surname><given-names>Howard L</given-names></name></person-group>. <article-title>Effects of in utero cocaine exposure on newborn acoustical cry characteristics »</article-title>. <year>1992</year><source>Pediatrics</source>, vol. <volume>89</volume>, no <issue>6</issue>, pp <fpage>1199</fpage>-<lpage>1203</lpage>.<pub-id pub-id-type="pmid">1594377</pub-id></mixed-citation></ref><ref id="cit0012"><label>[12]</label><mixed-citation publication-type="confproc"><person-group person-group-type="author"><name name-style="western"><surname>Orozco</surname><given-names>J</given-names></name>., <name name-style="western"><surname>Garcia</surname><given-names>C.A.R</given-names></name>.</person-group>,. <article-title>Detecting pathologies from infant cry applying scale conjugate gradient neural networks</article-title>. In: <conf-name>presented at the Europe an Symposium on Artificial Neural Networks</conf-name><conf-loc>Bruges-Belgium</conf-loc><year>2003</year></mixed-citation></ref><ref id="cit0013"><label>[13]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Hesam Farsaie</surname><given-names>Alaie</given-names></name>, et <name name-style="western"><surname>Chakib</surname><given-names>Tadj</given-names></name></person-group><article-title>« Cry-based classification of healthy and sick infants using adapted boosting mixture learning method for gaussian mixture models »</article-title>. <source>Modelling and Simulation in Engineering</source>, vol. <volume>2012</volume>, p. <fpage>55</fpage>.</mixed-citation></ref><ref id="cit0014"><label>[14]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Amaro-Camargo</surname><given-names>E</given-names></name>., <name name-style="western"><surname>Reyes García</surname><given-names>C</given-names></name>.</person-group><chapter-title>Applying statistical vectors of acoustic characteristics for the automatic classification of infant cry</chapter-title> In: <person-group person-group-type="editor"><name name-style="western"><surname>Huang</surname><given-names>D.-S</given-names></name>., <name name-style="western"><surname>Heutte</surname><given-names>L</given-names></name>., <name name-style="western"><surname>Loog</surname><given-names>M</given-names></name>.</person-group> (Eds.), <source>Advanced Intelligent Computing Theories and Applications. With Aspects of Theoretical and Methodological Issues, 4681</source>. <publisher-name>Springer</publisher-name>, <publisher-loc>Berlin/Heidelberg</publisher-loc>, <year>2007</year> pp. <fpage>1078</fpage>–<lpage>1085</lpage>.</mixed-citation></ref><ref id="cit0015"><label>[15]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Cano Ortiz</surname><given-names>Sergio D</given-names></name>, <name name-style="western"><surname>Beceiro</surname><given-names>Daniel I Escobedo</given-names></name> et <name name-style="western"><surname>Ekkel</surname><given-names>Taco</given-names></name></person-group>. <article-title>« A Radial Basis Function Network Oriented for Infant Cry Classification »</article-title>. In <source>Progress in Pattern Recognition, Image Analysis and Applications</source>,. <year>2004</year> Vol. <volume>3287</volume>, pp <fpage>374</fpage>-<lpage>380</lpage>.</mixed-citation></ref><ref id="cit0016"><label>[16]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Fisichelli</surname><given-names>Vincent R</given-names></name>., <name name-style="western"><surname>Karelitz</surname><given-names>S.</given-names></name></person-group>. <article-title>Frequency spectra of the cries of normal infants and those with down’s syndrome</article-title>. <source>Psychonomic Science</source>. <year>1966</year>, <volume>6</volume>, pp <fpage>195</fpage>-<lpage>196</lpage>.</mixed-citation></ref><ref id="cit0017"><label>[17]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Kheddache</surname><given-names>Y</given-names></name>, <name name-style="western"><surname>Tadj</surname><given-names>C</given-names></name>.</person-group><article-title>Frequential characterization of healthy and pathologic newborn cries</article-title>. <source>Am J Biomed Eng</source>. <year>2013</year>, <volume>3</volume>, pp <fpage>182</fpage>–<lpage>193</lpage>.</mixed-citation></ref><ref id="cit0018"><label>[18]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Kheddache</surname><given-names>Y</given-names></name>, <name name-style="western"><surname>Tadj</surname><given-names>C</given-names></name></person-group>, <article-title>Acoustic measures of the cry characteristics of healthy newborns and newborns with pathologies</article-title>, <source>J Biomed Sci Eng</source>. <year>2013</year>, <volume>6</volume>, pp <fpage>796</fpage>–<lpage>804</lpage>.</mixed-citation></ref><ref id="cit0019"><label>[19]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Kheddache</surname><given-names>Y</given-names></name>, <name name-style="western"><surname>Tadj</surname><given-names>C</given-names></name></person-group>, <article-title>Resonance frequencies behaviour in pathologic cries of newborns</article-title>, <source>Journal of Voice</source>, <year>2015</year>, <volume>29</volume>, pp <fpage>1</fpage>-<lpage>12</lpage><pub-id pub-id-type="pmid">25175781</pub-id></mixed-citation></ref><ref id="cit0020"><label>[20]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Aalto</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Aalto</surname><given-names>D.</given-names></name>, <name name-style="western"><surname>Malinen</surname><given-names>J.</given-names></name>, <name name-style="western"><surname>Vainio</surname><given-names>M</given-names></name></person-group>, <chapter-title>Interaction of vocal fold and vocal tract oscillations</chapter-title>, <source>Proceedings of the 24th Nordic Seminar on Computational Mechanics</source>, <person-group person-group-type="editor"><name name-style="western"><surname>Freund</surname><given-names>J.</given-names></name> and <name name-style="western"><surname>Kouhia</surname><given-names>R.</given-names></name></person-group> (Eds.) <publisher-name>Aalto University</publisher-name>, <year>2011</year>, pp. <fpage>1</fpage>-<lpage>4</lpage></mixed-citation></ref><ref id="cit0021"><label>[21]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Aalto</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Aalto</surname><given-names>D.</given-names></name>, <name name-style="western"><surname>Malinen</surname><given-names>J.</given-names></name>, <name name-style="western"><surname>Vainio</surname><given-names>M</given-names></name></person-group>, <article-title>Modal locking between vocal fold and vocal tract ascillations</article-title>, <comment>arXiv preprint arXiv:1506:01395</comment></mixed-citation></ref><ref id="cit0022"><label>[22]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Titze</surname><given-names>Ingo R.</given-names></name></person-group>, <article-title>Nonlinear source–filter coupling in phonation: Theory</article-title>, <source>J. Acoust. Soc. Am</source>, <year>2008</year>, pp <fpage>1902</fpage>-<lpage>1915</lpage>.<pub-id pub-id-type="pmid">18396999</pub-id></mixed-citation></ref><ref id="cit0023"><label>[23]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>honda</surname><given-names>Kiyoshi</given-names></name>, <name name-style="western"><surname>hirai</surname><given-names>Hiroyuki</given-names></name>, <name name-style="western"><surname>masaki</surname><given-names>Shinobu</given-names></name>, <name name-style="western"><surname>shimada</surname><given-names>Yasuhiro</given-names></name></person-group>, <article-title>Role of vertical larynx movement and cervical lordosis in F<sub>0</sub> control, language and speech</article-title>, <year>1999</year>, pp <fpage>401</fpage>–<lpage>411</lpage></mixed-citation></ref><ref id="cit0024"><label>[24]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>stevens</surname><given-names>K.N</given-names></name>.</person-group><source>Acoustic phonetics</source>, <publisher-loc>Cambridge, Massachusetts</publisher-loc>: <publisher-name>MIT Press</publisher-name>, (<year>1998</year>).</mixed-citation></ref><ref id="cit0025"><label>[25]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Bernardoni</surname><given-names>Nathalie Henrich</given-names></name>, <name name-style="western"><surname>Smith</surname><given-names>John</given-names></name>, <name name-style="western"><surname>Wolfe</surname><given-names>Joe</given-names></name></person-group>, <article-title>Vocal tract resonances in singing: variation with laryngeal mechanism for male operatic singers in chest and falsetto registers</article-title>, <source>J. Acoust. Soc. Am</source>, <year>2014</year>, pp <fpage>491</fpage>-<lpage>501</lpage>.<pub-id pub-id-type="pmid">24437789</pub-id></mixed-citation></ref><ref id="cit0026"><label>[26]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Echternach</surname><given-names>M</given-names></name>., <name name-style="western"><surname>Sundberg</surname><given-names>J</given-names></name>., <name name-style="western"><surname>Arndt</surname><given-names>S</given-names></name>., <name name-style="western"><surname>Markl</surname><given-names>M</given-names></name>., <name name-style="western"><surname>Schumacher</surname><given-names>M</given-names></name>., and <name name-style="western"><surname>Richter</surname><given-names>B</given-names></name></person-group>, <article-title>Vocal tract in female registers–A dynamic real-time MRI study</article-title>, <source>J. Voice</source>, <year>2010</year>, pp <fpage>133</fpage>–<lpage>139</lpage>.<pub-id pub-id-type="pmid">19185452</pub-id></mixed-citation></ref><ref id="cit0027"><label>[27]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Sundberg</surname><given-names>J</given-names></name>., <name name-style="western"><surname>La</surname><given-names>F. M. B</given-names></name>., and <name name-style="western"><surname>Gill</surname><given-names>B. P</given-names></name></person-group>, <article-title>Professional male singers’ formant tuning strategies for the vowel /a/,”</article-title><source>Logoped. Phoniatr. Vocol</source>. <year>2011</year>, pp <fpage>156</fpage>–<lpage>167</lpage>.<pub-id pub-id-type="pmid">21756222</pub-id></mixed-citation></ref><ref id="cit0028"><label>[28]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Sundberg</surname><given-names>Johan</given-names></name>, <name name-style="western"><surname>Filipa</surname><given-names>M. B</given-names></name>. <name name-style="western"><surname>Gill</surname><given-names>Brian P.</given-names></name></person-group>, <article-title>Formant Tuning Strategies in Professional Male Opera Singers</article-title><source>Journal of Voice</source>, <year>2013</year> pp <fpage>278</fpage>-<lpage>288</lpage>.<pub-id pub-id-type="pmid">23453594</pub-id></mixed-citation></ref><ref id="cit0029"><label>[29]</label><mixed-citation publication-type="web"><person-group person-group-type="author"><name name-style="western"><surname>Boersma</surname><given-names>P.</given-names></name>, <name name-style="western"><surname>Weenink</surname><given-names>D.</given-names></name></person-group>, <article-title>Praat: doing phonetics by computer</article-title>. <comment>[Online]. Available: <ext-link ext-link-type="uri" xlink:href="http://www.praat.org/">http://www.praat.org/</ext-link></comment></mixed-citation></ref><ref id="cit0030"><label>[30]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Abou-Abbas</surname><given-names>L</given-names></name>,. <name name-style="western"><surname>Tadj</surname><given-names>C</given-names></name>, <name name-style="western"><surname>Gargour</surname><given-names>C</given-names></name>, <name name-style="western"><surname>Montazeri</surname><given-names>L</given-names></name></person-group>, <article-title>Expiratory and Inspiratory Cries Detection Using Different Signals’ Decomposition Techniques</article-title>, <source>Journal of Voice</source>, <year>2017</year><volume>259</volume> pp <fpage>13</fpage>–<lpage>28</lpage></mixed-citation></ref><ref id="cit0031"><label>[31]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Markel</surname><given-names>J.D</given-names></name></person-group>, <article-title>The SIFT algorithm for fundamental frequency estimation</article-title>, <source>IEEE Transactions on Audio Electroacoustic</source>, <year>1972</year>, pp <fpage>367</fpage>-<lpage>377</lpage>.</mixed-citation></ref><ref id="cit0032"><label>[32]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Lederman</surname><given-names>D</given-names></name></person-group>, <source>Estimation of infants’ cry fundamental frequency using a modified SIFT algorithm</source>, <year>2010</year>, <comment>arXiv: 1009.2796</comment>.</mixed-citation></ref><ref id="cit0033"><label>[33]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Manfredi</surname><given-names>C</given-names></name>, <name name-style="western"><surname>Bocchi</surname><given-names>L</given-names></name>, <name name-style="western"><surname>Orlandi</surname><given-names>S</given-names></name>, <name name-style="western"><surname>Spaccaterra</surname><given-names>L</given-names></name>, <name name-style="western"><surname>Donzelli</surname><given-names>GP</given-names></name></person-group>, <article-title>High resolution cry analysis in preterm newborn infants</article-title>, <source>Med Eng Phys</source>, <year>2009</year>, pp <fpage>528</fpage>–<lpage>532</lpage>.</mixed-citation></ref><ref id="cit0034"><label>[34]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Fort</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Ismaellit</surname><given-names>A.</given-names></name>, <name name-style="western"><surname>Manfredi</surname><given-names>C.</given-names></name>, <name name-style="western"><surname>Bruscaglionit</surname><given-names>P.</given-names></name></person-group>, <article-title>Parametric and non-parametric estimation of speech formants: application to infant cry</article-title>, <source>Med. Eng. phys</source>. <year>1996</year>, pp <fpage>677</fpage>-<lpage>691</lpage><pub-id pub-id-type="pmid">8953561</pub-id></mixed-citation></ref><ref id="cit0035"><label>[35]</label><mixed-citation publication-type="confproc"><person-group person-group-type="author"><name name-style="western"><surname>Wermk</surname><given-names>K</given-names></name>, <name name-style="western"><surname>Mende</surname><given-names>W</given-names></name>, <name name-style="western"><surname>Kempf</surname><given-names>A</given-names></name>, <name name-style="western"><surname>Manfredi</surname><given-names>C</given-names></name>, <name name-style="western"><surname>Bruscaglioni</surname><given-names>P</given-names></name>, <name name-style="western"><surname>Stellzig- Eisenhauer</surname><given-names>A</given-names></name></person-group>, <article-title>Interaction patterns between melodies and resonance frequencies in infants’ pre-speech utterances</article-title>, In <conf-name>Proceedings of the 4th International Workshop. Models and Analysis of Vocal Emissions for Biomedical Applications</conf-name>, <year>2005</year> pp<fpage>187</fpage>–<lpage>190</lpage>,.</mixed-citation></ref><ref id="cit0036"><label>[36]</label><mixed-citation publication-type="book"><person-group person-group-type="author"><name name-style="western"><surname>Sweeney</surname><given-names>Walter P.</given-names><suffix>Jr.</suffix></name>, <name name-style="western"><surname>Musavi</surname><given-names>Mohamad T.</given-names></name>, <name name-style="western"><surname>Guidi</surname><given-names>John N.</given-names></name></person-group><source>Classification of chromosomes using a probabilistic neural network</source>, <publisher-name>Wiley-Liss, Inc.</publisher-name><year>1994</year> pp. <fpage>17</fpage>-<lpage>24</lpage>,.</mixed-citation></ref><ref id="cit0037"><label>[37]</label><mixed-citation publication-type="confproc"><person-group person-group-type="author"><name name-style="western"><surname>Othman</surname><given-names>Mohd Fauzi</given-names></name>, <name name-style="western"><surname>Mohd Basri</surname><given-names>Mohd Ariffanan</given-names></name></person-group> . <article-title>Probabilistic neural network for brain tumor classification</article-title>. <conf-name>Second International Conference on Intelligent Systems, Modelling and Simulation</conf-name>, <year>2011</year> pp.<fpage>136</fpage>-<lpage>138,</lpage>.</mixed-citation></ref><ref id="cit0038"><label>[38]</label><mixed-citation publication-type="journal"><person-group person-group-type="author"><name name-style="western"><surname>Kusy</surname><given-names>Maciej</given-names></name> Email author, <name name-style="western"><surname>Zajdel</surname><given-names>Roman</given-names></name></person-group><article-title>“ Probabilistic neural network training procedure based on Q(0)-learning algorithm in medical data classification”</article-title>, <source>Applied Intelligence</source>, <year>2014</year> pp. <fpage>837</fpage>–<lpage>854</lpage>.</mixed-citation></ref><ref id="cit0039"><label>[39]</label><mixed-citation publication-type="book"><article-title>Neural Network Toolbox™ 6 User’s Guide</article-title>.</mixed-citation></ref></ref-list></back></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 0001670 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd -nk 0001670 | 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. |  |