Fast and efficient retinal blood vessel segmentation method based on deep learning network.
Identifieur interne : 000809 ( Main/Exploration ); précédent : 000808; suivant : 000810Fast and efficient retinal blood vessel segmentation method based on deep learning network.
Auteurs : Henda Boudegga [Tunisie] ; Yaroub Elloumi [Tunisie] ; Mohamed Akil [France] ; Mohamed Hedi Bedoui [Tunisie] ; Rostom Kachouri [France] ; Asma Ben Abdallah [Tunisie]Source :
- Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society [ 1879-0771 ] ; 2021.
Abstract
The segmentation of the retinal vascular tree presents a major step for detecting ocular pathologies. The clinical context expects higher segmentation performance with a reduced processing time. For higher accurate segmentation, several automated methods have been based on Deep Learning (DL) networks. However, the used convolutional layers bring to a higher computational complexity and so for execution times. For such need, this work presents a new DL based method for retinal vessel tree segmentation. Our main contribution consists in suggesting a new U-form DL architecture using lightweight convolution blocks in order to preserve a higher segmentation performance while reducing the computational complexity. As a second main contribution, preprocessing and data augmentation steps are proposed with respect to the retinal image and blood vessel characteristics. The proposed method is tested on DRIVE and STARE databases, which can achieve a better trade-off between the retinal blood vessel detection rate and the detection time with average accuracy of 0.978 and 0.98 in 0.59 s and 0.48 s per fundus image on GPU NVIDIA GTX 980 platforms, respectively for DRIVE and STARE database fundus images.
DOI: 10.1016/j.compmedimag.2021.101902
PubMed: 33892389
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PubMed, to step Corpus: 000016
- to stream PubMed, to step Curation: 000016
- to stream PubMed, to step Checkpoint: 000045
- to stream Main, to step Merge: 000809
- to stream Main, to step Curation: 000809
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Fast and efficient retinal blood vessel segmentation method based on deep learning network.</title><author><name sortKey="Boudegga, Henda" sort="Boudegga, Henda" uniqKey="Boudegga H" first="Henda" last="Boudegga">Henda Boudegga</name><affiliation wicri:level="4"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; ISITCom Hammam-Sousse, University of Sousse, Tunisia. Electronic address: hendaboudegga@gmail.com.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; ISITCom Hammam-Sousse, University of Sousse</wicri:regionArea><placeName><settlement type="city">Sousse</settlement><region type="region" nuts="2">Gouvernorat de Sousse</region></placeName><orgName type="university">Université de Sousse</orgName></affiliation></author><author><name sortKey="Elloumi, Yaroub" sort="Elloumi, Yaroub" uniqKey="Elloumi Y" first="Yaroub" last="Elloumi">Yaroub Elloumi</name><affiliation wicri:level="4"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France; ISITCom Hammam-Sousse, University of Sousse, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France; ISITCom Hammam-Sousse, University of Sousse</wicri:regionArea><placeName><settlement type="city">Sousse</settlement><region type="region" nuts="2">Gouvernorat de Sousse</region></placeName><orgName type="university">Université de Sousse</orgName></affiliation></author><author><name sortKey="Akil, Mohamed" sort="Akil, Mohamed" uniqKey="Akil M" first="Mohamed" last="Akil">Mohamed Akil</name><affiliation wicri:level="3"><nlm:affiliation>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France.</nlm:affiliation><country xml:lang="fr">France</country><wicri:regionArea>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée</wicri:regionArea><placeName><region type="region" nuts="2">Île-de-France</region><settlement type="city">Marne-la-Vallée</settlement></placeName></affiliation></author><author><name sortKey="Hedi Bedoui, Mohamed" sort="Hedi Bedoui, Mohamed" uniqKey="Hedi Bedoui M" first="Mohamed" last="Hedi Bedoui">Mohamed Hedi Bedoui</name><affiliation wicri:level="1"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir</wicri:regionArea><wicri:noRegion>University of Monastir</wicri:noRegion></affiliation></author><author><name sortKey="Kachouri, Rostom" sort="Kachouri, Rostom" uniqKey="Kachouri R" first="Rostom" last="Kachouri">Rostom Kachouri</name><affiliation wicri:level="3"><nlm:affiliation>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France.</nlm:affiliation><country xml:lang="fr">France</country><wicri:regionArea>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée</wicri:regionArea><placeName><region type="region" nuts="2">Île-de-France</region><settlement type="city">Marne-la-Vallée</settlement></placeName></affiliation></author><author><name sortKey="Abdallah, Asma Ben" sort="Abdallah, Asma Ben" uniqKey="Abdallah A" first="Asma Ben" last="Abdallah">Asma Ben Abdallah</name><affiliation wicri:level="1"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir</wicri:regionArea><wicri:noRegion>University of Monastir</wicri:noRegion></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2021">2021</date><idno type="RBID">pubmed:33892389</idno><idno type="pmid">33892389</idno><idno type="doi">10.1016/j.compmedimag.2021.101902</idno><idno type="wicri:Area/PubMed/Corpus">000016</idno><idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Corpus" wicri:corpus="PubMed">000016</idno><idno type="wicri:Area/PubMed/Curation">000016</idno><idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Curation">000016</idno><idno type="wicri:Area/PubMed/Checkpoint">000045</idno><idno type="wicri:explorRef" wicri:stream="Checkpoint" wicri:step="PubMed">000045</idno><idno type="wicri:Area/Main/Merge">000809</idno><idno type="wicri:Area/Main/Curation">000809</idno><idno type="wicri:Area/Main/Exploration">000809</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Fast and efficient retinal blood vessel segmentation method based on deep learning network.</title><author><name sortKey="Boudegga, Henda" sort="Boudegga, Henda" uniqKey="Boudegga H" first="Henda" last="Boudegga">Henda Boudegga</name><affiliation wicri:level="4"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; ISITCom Hammam-Sousse, University of Sousse, Tunisia. Electronic address: hendaboudegga@gmail.com.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; ISITCom Hammam-Sousse, University of Sousse</wicri:regionArea><placeName><settlement type="city">Sousse</settlement><region type="region" nuts="2">Gouvernorat de Sousse</region></placeName><orgName type="university">Université de Sousse</orgName></affiliation></author><author><name sortKey="Elloumi, Yaroub" sort="Elloumi, Yaroub" uniqKey="Elloumi Y" first="Yaroub" last="Elloumi">Yaroub Elloumi</name><affiliation wicri:level="4"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France; ISITCom Hammam-Sousse, University of Sousse, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia; LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France; ISITCom Hammam-Sousse, University of Sousse</wicri:regionArea><placeName><settlement type="city">Sousse</settlement><region type="region" nuts="2">Gouvernorat de Sousse</region></placeName><orgName type="university">Université de Sousse</orgName></affiliation></author><author><name sortKey="Akil, Mohamed" sort="Akil, Mohamed" uniqKey="Akil M" first="Mohamed" last="Akil">Mohamed Akil</name><affiliation wicri:level="3"><nlm:affiliation>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France.</nlm:affiliation><country xml:lang="fr">France</country><wicri:regionArea>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée</wicri:regionArea><placeName><region type="region" nuts="2">Île-de-France</region><settlement type="city">Marne-la-Vallée</settlement></placeName></affiliation></author><author><name sortKey="Hedi Bedoui, Mohamed" sort="Hedi Bedoui, Mohamed" uniqKey="Hedi Bedoui M" first="Mohamed" last="Hedi Bedoui">Mohamed Hedi Bedoui</name><affiliation wicri:level="1"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir</wicri:regionArea><wicri:noRegion>University of Monastir</wicri:noRegion></affiliation></author><author><name sortKey="Kachouri, Rostom" sort="Kachouri, Rostom" uniqKey="Kachouri R" first="Rostom" last="Kachouri">Rostom Kachouri</name><affiliation wicri:level="3"><nlm:affiliation>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée, France.</nlm:affiliation><country xml:lang="fr">France</country><wicri:regionArea>LIGM, Univ Gustave Eiffel, CNRS, ESIEE Paris, F-77454 Marne-la-Vallée</wicri:regionArea><placeName><region type="region" nuts="2">Île-de-France</region><settlement type="city">Marne-la-Vallée</settlement></placeName></affiliation></author><author><name sortKey="Abdallah, Asma Ben" sort="Abdallah, Asma Ben" uniqKey="Abdallah A" first="Asma Ben" last="Abdallah">Asma Ben Abdallah</name><affiliation wicri:level="1"><nlm:affiliation>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir, Tunisia.</nlm:affiliation><country xml:lang="fr">Tunisie</country><wicri:regionArea>Medical Technology and Image Processing Laboratory, Faculty of Medicine, University of Monastir</wicri:regionArea><wicri:noRegion>University of Monastir</wicri:noRegion></affiliation></author></analytic><series><title level="j">Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society</title><idno type="eISSN">1879-0771</idno><imprint><date when="2021" type="published">2021</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The segmentation of the retinal vascular tree presents a major step for detecting ocular pathologies. The clinical context expects higher segmentation performance with a reduced processing time. For higher accurate segmentation, several automated methods have been based on Deep Learning (DL) networks. However, the used convolutional layers bring to a higher computational complexity and so for execution times. For such need, this work presents a new DL based method for retinal vessel tree segmentation. Our main contribution consists in suggesting a new U-form DL architecture using lightweight convolution blocks in order to preserve a higher segmentation performance while reducing the computational complexity. As a second main contribution, preprocessing and data augmentation steps are proposed with respect to the retinal image and blood vessel characteristics. The proposed method is tested on DRIVE and STARE databases, which can achieve a better trade-off between the retinal blood vessel detection rate and the detection time with average accuracy of 0.978 and 0.98 in 0.59 s and 0.48 s per fundus image on GPU NVIDIA GTX 980 platforms, respectively for DRIVE and STARE database fundus images.</div></front></TEI><affiliations><list><country><li>France</li><li>Tunisie</li></country><region><li>Gouvernorat de Sousse</li><li>Île-de-France</li></region><settlement><li>Marne-la-Vallée</li><li>Sousse</li></settlement><orgName><li>Université de Sousse</li></orgName></list><tree><country name="Tunisie"><region name="Gouvernorat de Sousse"><name sortKey="Boudegga, Henda" sort="Boudegga, Henda" uniqKey="Boudegga H" first="Henda" last="Boudegga">Henda Boudegga</name></region><name sortKey="Abdallah, Asma Ben" sort="Abdallah, Asma Ben" uniqKey="Abdallah A" first="Asma Ben" last="Abdallah">Asma Ben Abdallah</name><name sortKey="Elloumi, Yaroub" sort="Elloumi, Yaroub" uniqKey="Elloumi Y" first="Yaroub" last="Elloumi">Yaroub Elloumi</name><name sortKey="Hedi Bedoui, Mohamed" sort="Hedi Bedoui, Mohamed" uniqKey="Hedi Bedoui M" first="Mohamed" last="Hedi Bedoui">Mohamed Hedi Bedoui</name></country><country name="France"><region name="Île-de-France"><name sortKey="Akil, Mohamed" sort="Akil, Mohamed" uniqKey="Akil M" first="Mohamed" last="Akil">Mohamed Akil</name></region><name sortKey="Kachouri, Rostom" sort="Kachouri, Rostom" uniqKey="Kachouri R" first="Rostom" last="Kachouri">Rostom Kachouri</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sante/explor/MaghrebDataLibMedV2/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000809 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000809 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Sante
|area= MaghrebDataLibMedV2
|flux= Main
|étape= Exploration
|type= RBID
|clé= pubmed:33892389
|texte= Fast and efficient retinal blood vessel segmentation method based on deep learning network.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Exploration/RBID.i -Sk "pubmed:33892389" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd \
| NlmPubMed2Wicri -a MaghrebDataLibMedV2
| This area was generated with Dilib version V0.6.38. |  |