Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program
Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. 25,326 gradable retinal images of patient...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | Raumviboonsuk, Paisan Krause, Jonathan Chotcomwongse, Peranut Sayres, Rory Raman, Rajiv Widner, Kasumi Campana, Bilson J L Phene, Sonia Hemarat, Kornwipa Tadarati, Mongkol Silpa-Acha, Sukhum Limwattanayingyong, Jirawut Rao, Chetan Kuruvilla, Oscar Jung, Jesse Tan, Jeffrey Orprayoon, Surapong Kangwanwongpaisan, Chawawat Sukulmalpaiboon, Ramase Luengchaichawang, Chainarong Fuangkaew, Jitumporn Kongsap, Pipat Chualinpha, Lamyong Saree, Sarawuth Kawinpanitan, Srirat Mitvongsa, Korntip Lawanasakol, Siriporn Thepchatri, Chaiyasit Wongpichedchai, Lalita Corrado, Greg S Peng, Lily Webster, Dale R |
| description | Deep learning algorithms have been used to detect diabetic retinopathy (DR)
with specialist-level accuracy. This study aims to validate one such algorithm
on a large-scale clinical population, and compare the algorithm performance
with that of human graders. 25,326 gradable retinal images of patients with
diabetes from the community-based, nation-wide screening program of DR in
Thailand were analyzed for DR severity and referable diabetic macular edema
(DME). Grades adjudicated by a panel of international retinal specialists
served as the reference standard. Across different severity levels of DR for
determining referable disease, deep learning significantly reduced the false
negative rate (by 23%) at the cost of slightly higher false positive rates
(2%). Deep learning algorithms may serve as a valuable tool for DR screening. |
| doi_str_mv | 10.48550/arxiv.1810.08290 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1810_08290</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1810_08290</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-90e28fec2e75334111ed1af62aa2be64b10ddb0d56968b1506c0416bd69b31b13</originalsourceid><addsrcrecordid>eNotkMtOwzAQRb1hgQofwIr5gQQ7DzdZohRapAoQrcQyGseTYimxIycEsufDSQObGeno3hnpMHYjeJhkacrv0H-bMRTZDHgW5fyS_WyIOtgTemvsCcY-hN1nixa2HjX5HmrnoWiw7009nRMHGsmbYZo7IzU9uBo2BhUNpoK3eVrX4fAxgbGAM8AmeHe-0fCMg3H2y2iCQ-WJlnev3p08tlfsosamp-v_vWLHx4djsQv2L9un4n4foFzzIOcUZTVVEa3TOE6EEKQF1jJCjBTJRAmuteI6lbnMlEi5rHgipNIyV7FQIl6x27-zi4ay86ZFP5VnHeWiI_4FWZpcCQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program</title><source>arXiv.org</source><creator>Raumviboonsuk, Paisan ; Krause, Jonathan ; Chotcomwongse, Peranut ; Sayres, Rory ; Raman, Rajiv ; Widner, Kasumi ; Campana, Bilson J L ; Phene, Sonia ; Hemarat, Kornwipa ; Tadarati, Mongkol ; Silpa-Acha, Sukhum ; Limwattanayingyong, Jirawut ; Rao, Chetan ; Kuruvilla, Oscar ; Jung, Jesse ; Tan, Jeffrey ; Orprayoon, Surapong ; Kangwanwongpaisan, Chawawat ; Sukulmalpaiboon, Ramase ; Luengchaichawang, Chainarong ; Fuangkaew, Jitumporn ; Kongsap, Pipat ; Chualinpha, Lamyong ; Saree, Sarawuth ; Kawinpanitan, Srirat ; Mitvongsa, Korntip ; Lawanasakol, Siriporn ; Thepchatri, Chaiyasit ; Wongpichedchai, Lalita ; Corrado, Greg S ; Peng, Lily ; Webster, Dale R</creator><creatorcontrib>Raumviboonsuk, Paisan ; Krause, Jonathan ; Chotcomwongse, Peranut ; Sayres, Rory ; Raman, Rajiv ; Widner, Kasumi ; Campana, Bilson J L ; Phene, Sonia ; Hemarat, Kornwipa ; Tadarati, Mongkol ; Silpa-Acha, Sukhum ; Limwattanayingyong, Jirawut ; Rao, Chetan ; Kuruvilla, Oscar ; Jung, Jesse ; Tan, Jeffrey ; Orprayoon, Surapong ; Kangwanwongpaisan, Chawawat ; Sukulmalpaiboon, Ramase ; Luengchaichawang, Chainarong ; Fuangkaew, Jitumporn ; Kongsap, Pipat ; Chualinpha, Lamyong ; Saree, Sarawuth ; Kawinpanitan, Srirat ; Mitvongsa, Korntip ; Lawanasakol, Siriporn ; Thepchatri, Chaiyasit ; Wongpichedchai, Lalita ; Corrado, Greg S ; Peng, Lily ; Webster, Dale R</creatorcontrib><description>Deep learning algorithms have been used to detect diabetic retinopathy (DR)
with specialist-level accuracy. This study aims to validate one such algorithm
on a large-scale clinical population, and compare the algorithm performance
with that of human graders. 25,326 gradable retinal images of patients with
diabetes from the community-based, nation-wide screening program of DR in
Thailand were analyzed for DR severity and referable diabetic macular edema
(DME). Grades adjudicated by a panel of international retinal specialists
served as the reference standard. Across different severity levels of DR for
determining referable disease, deep learning significantly reduced the false
negative rate (by 23%) at the cost of slightly higher false positive rates
(2%). Deep learning algorithms may serve as a valuable tool for DR screening.</description><identifier>DOI: 10.48550/arxiv.1810.08290</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2018-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1810.08290$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1810.08290$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Raumviboonsuk, Paisan</creatorcontrib><creatorcontrib>Krause, Jonathan</creatorcontrib><creatorcontrib>Chotcomwongse, Peranut</creatorcontrib><creatorcontrib>Sayres, Rory</creatorcontrib><creatorcontrib>Raman, Rajiv</creatorcontrib><creatorcontrib>Widner, Kasumi</creatorcontrib><creatorcontrib>Campana, Bilson J L</creatorcontrib><creatorcontrib>Phene, Sonia</creatorcontrib><creatorcontrib>Hemarat, Kornwipa</creatorcontrib><creatorcontrib>Tadarati, Mongkol</creatorcontrib><creatorcontrib>Silpa-Acha, Sukhum</creatorcontrib><creatorcontrib>Limwattanayingyong, Jirawut</creatorcontrib><creatorcontrib>Rao, Chetan</creatorcontrib><creatorcontrib>Kuruvilla, Oscar</creatorcontrib><creatorcontrib>Jung, Jesse</creatorcontrib><creatorcontrib>Tan, Jeffrey</creatorcontrib><creatorcontrib>Orprayoon, Surapong</creatorcontrib><creatorcontrib>Kangwanwongpaisan, Chawawat</creatorcontrib><creatorcontrib>Sukulmalpaiboon, Ramase</creatorcontrib><creatorcontrib>Luengchaichawang, Chainarong</creatorcontrib><creatorcontrib>Fuangkaew, Jitumporn</creatorcontrib><creatorcontrib>Kongsap, Pipat</creatorcontrib><creatorcontrib>Chualinpha, Lamyong</creatorcontrib><creatorcontrib>Saree, Sarawuth</creatorcontrib><creatorcontrib>Kawinpanitan, Srirat</creatorcontrib><creatorcontrib>Mitvongsa, Korntip</creatorcontrib><creatorcontrib>Lawanasakol, Siriporn</creatorcontrib><creatorcontrib>Thepchatri, Chaiyasit</creatorcontrib><creatorcontrib>Wongpichedchai, Lalita</creatorcontrib><creatorcontrib>Corrado, Greg S</creatorcontrib><creatorcontrib>Peng, Lily</creatorcontrib><creatorcontrib>Webster, Dale R</creatorcontrib><title>Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program</title><description>Deep learning algorithms have been used to detect diabetic retinopathy (DR)
with specialist-level accuracy. This study aims to validate one such algorithm
on a large-scale clinical population, and compare the algorithm performance
with that of human graders. 25,326 gradable retinal images of patients with
diabetes from the community-based, nation-wide screening program of DR in
Thailand were analyzed for DR severity and referable diabetic macular edema
(DME). Grades adjudicated by a panel of international retinal specialists
served as the reference standard. Across different severity levels of DR for
determining referable disease, deep learning significantly reduced the false
negative rate (by 23%) at the cost of slightly higher false positive rates
(2%). Deep learning algorithms may serve as a valuable tool for DR screening.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkMtOwzAQRb1hgQofwIr5gQQ7DzdZohRapAoQrcQyGseTYimxIycEsufDSQObGeno3hnpMHYjeJhkacrv0H-bMRTZDHgW5fyS_WyIOtgTemvsCcY-hN1nixa2HjX5HmrnoWiw7009nRMHGsmbYZo7IzU9uBo2BhUNpoK3eVrX4fAxgbGAM8AmeHe-0fCMg3H2y2iCQ-WJlnev3p08tlfsosamp-v_vWLHx4djsQv2L9un4n4foFzzIOcUZTVVEa3TOE6EEKQF1jJCjBTJRAmuteI6lbnMlEi5rHgipNIyV7FQIl6x27-zi4ay86ZFP5VnHeWiI_4FWZpcCQ</recordid><startdate>20181018</startdate><enddate>20181018</enddate><creator>Raumviboonsuk, Paisan</creator><creator>Krause, Jonathan</creator><creator>Chotcomwongse, Peranut</creator><creator>Sayres, Rory</creator><creator>Raman, Rajiv</creator><creator>Widner, Kasumi</creator><creator>Campana, Bilson J L</creator><creator>Phene, Sonia</creator><creator>Hemarat, Kornwipa</creator><creator>Tadarati, Mongkol</creator><creator>Silpa-Acha, Sukhum</creator><creator>Limwattanayingyong, Jirawut</creator><creator>Rao, Chetan</creator><creator>Kuruvilla, Oscar</creator><creator>Jung, Jesse</creator><creator>Tan, Jeffrey</creator><creator>Orprayoon, Surapong</creator><creator>Kangwanwongpaisan, Chawawat</creator><creator>Sukulmalpaiboon, Ramase</creator><creator>Luengchaichawang, Chainarong</creator><creator>Fuangkaew, Jitumporn</creator><creator>Kongsap, Pipat</creator><creator>Chualinpha, Lamyong</creator><creator>Saree, Sarawuth</creator><creator>Kawinpanitan, Srirat</creator><creator>Mitvongsa, Korntip</creator><creator>Lawanasakol, Siriporn</creator><creator>Thepchatri, Chaiyasit</creator><creator>Wongpichedchai, Lalita</creator><creator>Corrado, Greg S</creator><creator>Peng, Lily</creator><creator>Webster, Dale R</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20181018</creationdate><title>Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program</title><author>Raumviboonsuk, Paisan ; Krause, Jonathan ; Chotcomwongse, Peranut ; Sayres, Rory ; Raman, Rajiv ; Widner, Kasumi ; Campana, Bilson J L ; Phene, Sonia ; Hemarat, Kornwipa ; Tadarati, Mongkol ; Silpa-Acha, Sukhum ; Limwattanayingyong, Jirawut ; Rao, Chetan ; Kuruvilla, Oscar ; Jung, Jesse ; Tan, Jeffrey ; Orprayoon, Surapong ; Kangwanwongpaisan, Chawawat ; Sukulmalpaiboon, Ramase ; Luengchaichawang, Chainarong ; Fuangkaew, Jitumporn ; Kongsap, Pipat ; Chualinpha, Lamyong ; Saree, Sarawuth ; Kawinpanitan, Srirat ; Mitvongsa, Korntip ; Lawanasakol, Siriporn ; Thepchatri, Chaiyasit ; Wongpichedchai, Lalita ; Corrado, Greg S ; Peng, Lily ; Webster, Dale R</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-90e28fec2e75334111ed1af62aa2be64b10ddb0d56968b1506c0416bd69b31b13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Raumviboonsuk, Paisan</creatorcontrib><creatorcontrib>Krause, Jonathan</creatorcontrib><creatorcontrib>Chotcomwongse, Peranut</creatorcontrib><creatorcontrib>Sayres, Rory</creatorcontrib><creatorcontrib>Raman, Rajiv</creatorcontrib><creatorcontrib>Widner, Kasumi</creatorcontrib><creatorcontrib>Campana, Bilson J L</creatorcontrib><creatorcontrib>Phene, Sonia</creatorcontrib><creatorcontrib>Hemarat, Kornwipa</creatorcontrib><creatorcontrib>Tadarati, Mongkol</creatorcontrib><creatorcontrib>Silpa-Acha, Sukhum</creatorcontrib><creatorcontrib>Limwattanayingyong, Jirawut</creatorcontrib><creatorcontrib>Rao, Chetan</creatorcontrib><creatorcontrib>Kuruvilla, Oscar</creatorcontrib><creatorcontrib>Jung, Jesse</creatorcontrib><creatorcontrib>Tan, Jeffrey</creatorcontrib><creatorcontrib>Orprayoon, Surapong</creatorcontrib><creatorcontrib>Kangwanwongpaisan, Chawawat</creatorcontrib><creatorcontrib>Sukulmalpaiboon, Ramase</creatorcontrib><creatorcontrib>Luengchaichawang, Chainarong</creatorcontrib><creatorcontrib>Fuangkaew, Jitumporn</creatorcontrib><creatorcontrib>Kongsap, Pipat</creatorcontrib><creatorcontrib>Chualinpha, Lamyong</creatorcontrib><creatorcontrib>Saree, Sarawuth</creatorcontrib><creatorcontrib>Kawinpanitan, Srirat</creatorcontrib><creatorcontrib>Mitvongsa, Korntip</creatorcontrib><creatorcontrib>Lawanasakol, Siriporn</creatorcontrib><creatorcontrib>Thepchatri, Chaiyasit</creatorcontrib><creatorcontrib>Wongpichedchai, Lalita</creatorcontrib><creatorcontrib>Corrado, Greg S</creatorcontrib><creatorcontrib>Peng, Lily</creatorcontrib><creatorcontrib>Webster, Dale R</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Raumviboonsuk, Paisan</au><au>Krause, Jonathan</au><au>Chotcomwongse, Peranut</au><au>Sayres, Rory</au><au>Raman, Rajiv</au><au>Widner, Kasumi</au><au>Campana, Bilson J L</au><au>Phene, Sonia</au><au>Hemarat, Kornwipa</au><au>Tadarati, Mongkol</au><au>Silpa-Acha, Sukhum</au><au>Limwattanayingyong, Jirawut</au><au>Rao, Chetan</au><au>Kuruvilla, Oscar</au><au>Jung, Jesse</au><au>Tan, Jeffrey</au><au>Orprayoon, Surapong</au><au>Kangwanwongpaisan, Chawawat</au><au>Sukulmalpaiboon, Ramase</au><au>Luengchaichawang, Chainarong</au><au>Fuangkaew, Jitumporn</au><au>Kongsap, Pipat</au><au>Chualinpha, Lamyong</au><au>Saree, Sarawuth</au><au>Kawinpanitan, Srirat</au><au>Mitvongsa, Korntip</au><au>Lawanasakol, Siriporn</au><au>Thepchatri, Chaiyasit</au><au>Wongpichedchai, Lalita</au><au>Corrado, Greg S</au><au>Peng, Lily</au><au>Webster, Dale R</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program</atitle><date>2018-10-18</date><risdate>2018</risdate><abstract>Deep learning algorithms have been used to detect diabetic retinopathy (DR)
with specialist-level accuracy. This study aims to validate one such algorithm
on a large-scale clinical population, and compare the algorithm performance
with that of human graders. 25,326 gradable retinal images of patients with
diabetes from the community-based, nation-wide screening program of DR in
Thailand were analyzed for DR severity and referable diabetic macular edema
(DME). Grades adjudicated by a panel of international retinal specialists
served as the reference standard. Across different severity levels of DR for
determining referable disease, deep learning significantly reduced the false
negative rate (by 23%) at the cost of slightly higher false positive rates
(2%). Deep learning algorithms may serve as a valuable tool for DR screening.</abstract><doi>10.48550/arxiv.1810.08290</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.1810.08290 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_1810_08290 |
| source | arXiv.org |
| subjects | Computer Science - Computer Vision and Pattern Recognition |
| title | Deep Learning vs. Human Graders for Classifying Severity Levels of Diabetic Retinopathy in a Real-World Nationwide Screening Program |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-27T01%3A49%3A34IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Deep%20Learning%20vs.%20Human%20Graders%20for%20Classifying%20Severity%20Levels%20of%20Diabetic%20Retinopathy%20in%20a%20Real-World%20Nationwide%20Screening%20Program&rft.au=Raumviboonsuk,%20Paisan&rft.date=2018-10-18&rft_id=info:doi/10.48550/arxiv.1810.08290&rft_dat=%3Carxiv_GOX%3E1810_08290%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |