Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center
In 2018, IDx-DR was approved as a method to determine the degree of diabetic retinopathy (DR) using artificial intelligence (AI) by the FDA. We integrated IDx-DR into the consultation at a diabetology focus clinic and report the agreement between IDx-DR and fundoscopy as well as IDx-DR and ophthalmo...
Gespeichert in:
| Veröffentlicht in: | Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft 2022-07, Vol.119 (7), p.705-713 |
|---|---|
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , |
| Format: | Artikel |
| Sprache: | ger |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 713 |
|---|---|
| container_issue | 7 |
| container_start_page | 705 |
| container_title | Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft |
| container_volume | 119 |
| creator | Paul, Sebastian Tayar, Allam Morawiec-Kisiel, Ewa Bohl, Beathe Großjohann, Rico Hunfeld, Elisabeth Busch, Martin Pfeil, Johanna M Dähmcke, Merlin Brauckmann, Tara Eilts, Sonja Bründer, Marie-Christine Grundel, Milena Grundel, Bastian Tost, Frank Kuhn, Jana Reindel, Jörg Kerner, Wolfgang Stahl, Andreas |
| description | In 2018, IDx-DR was approved as a method to determine the degree of diabetic retinopathy (DR) using artificial intelligence (AI) by the FDA.
We integrated IDx-DR into the consultation at a diabetology focus clinic and report the agreement between IDx-DR and fundoscopy as well as IDx-DR and ophthalmological image assessment and the influence of different camera systems.
Adequate image quality in miosis was achieved more frequently with the Topcon camera (n = 456; NW400, Topcon Medical Systems, Oakland, NJ, USA) compared with the Zeiss camera (n = 47; Zeiss VISUCAM 500, Carl Zeiss Meditec AG, Jena, Germany). Overall, IDx-DR analysis in miosis was possible in approximately 60% of the patients. All patients in whom IDx-DR analysis in miosis was not possible could be assessed by fundoscopy with dilated pupils. Within the group of images that could be evaluated, there was agreement between IDx-DR and ophthalmic fundoscopy in approximately 55%, overestimation of severity by IDx-DR in approximately 40% and underestimation in approximately 4%. The sensitivity (specificity) for detecting severe retinopathy requiring treatment was 95.7% (89.1%) for cases with fundus images that could be evaluated and 65.2% (66.7%) when all cases were considered (including those without images in miosis which could be evaluated). The kappa coefficient of 0.334 (p |
| doi_str_mv | 10.1007/s00347-021-01556-5 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_proquest_miscellaneous_2622961701</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2622961701</sourcerecordid><originalsourceid>FETCH-LOGICAL-p141t-67e7610d683243adbd604d15ebecb34bc0f49cd386fc5a416fb4390da5039b023</originalsourceid><addsrcrecordid>eNo1kMtKA0EQRRtBTIz-gAvppZvR6udklhI0CgE3Zj30ozq2TGbG7s4if-O3-GUOGDdVFBwO9xYhNwzuGUD9kAGErCvgrAKmlK7UGZkzKUQFkosZucz5EwC0EvyCzISCJWgJc-K2GekQqEklhuii6WjsC3Zd3GHvcDpodgmxj_2OhiFRH43FEh1N0-yH0ZSPIzWFmp_vgpPEpOOJwUwdTq50Rc6D6TJen_aCbJ-f3lcv1eZt_bp63FQjk6xUusZaM_B6KbgUxluvQXqm0KKzQloHQTbOi6UOThnJdLBSNOCNAtFY4GJB7v68Yxq-DphLu4_ZTV1Mj8Mht1xz3mhWA5vQ2xN6sHv07Zjifkre_j9G_AJRlWV9</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2622961701</pqid></control><display><type>article</type><title>Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center</title><source>Springer Nature - Complete Springer Journals</source><creator>Paul, Sebastian ; Tayar, Allam ; Morawiec-Kisiel, Ewa ; Bohl, Beathe ; Großjohann, Rico ; Hunfeld, Elisabeth ; Busch, Martin ; Pfeil, Johanna M ; Dähmcke, Merlin ; Brauckmann, Tara ; Eilts, Sonja ; Bründer, Marie-Christine ; Grundel, Milena ; Grundel, Bastian ; Tost, Frank ; Kuhn, Jana ; Reindel, Jörg ; Kerner, Wolfgang ; Stahl, Andreas</creator><creatorcontrib>Paul, Sebastian ; Tayar, Allam ; Morawiec-Kisiel, Ewa ; Bohl, Beathe ; Großjohann, Rico ; Hunfeld, Elisabeth ; Busch, Martin ; Pfeil, Johanna M ; Dähmcke, Merlin ; Brauckmann, Tara ; Eilts, Sonja ; Bründer, Marie-Christine ; Grundel, Milena ; Grundel, Bastian ; Tost, Frank ; Kuhn, Jana ; Reindel, Jörg ; Kerner, Wolfgang ; Stahl, Andreas</creatorcontrib><description>In 2018, IDx-DR was approved as a method to determine the degree of diabetic retinopathy (DR) using artificial intelligence (AI) by the FDA.
We integrated IDx-DR into the consultation at a diabetology focus clinic and report the agreement between IDx-DR and fundoscopy as well as IDx-DR and ophthalmological image assessment and the influence of different camera systems.
Adequate image quality in miosis was achieved more frequently with the Topcon camera (n = 456; NW400, Topcon Medical Systems, Oakland, NJ, USA) compared with the Zeiss camera (n = 47; Zeiss VISUCAM 500, Carl Zeiss Meditec AG, Jena, Germany). Overall, IDx-DR analysis in miosis was possible in approximately 60% of the patients. All patients in whom IDx-DR analysis in miosis was not possible could be assessed by fundoscopy with dilated pupils. Within the group of images that could be evaluated, there was agreement between IDx-DR and ophthalmic fundoscopy in approximately 55%, overestimation of severity by IDx-DR in approximately 40% and underestimation in approximately 4%. The sensitivity (specificity) for detecting severe retinopathy requiring treatment was 95.7% (89.1%) for cases with fundus images that could be evaluated and 65.2% (66.7%) when all cases were considered (including those without images in miosis which could be evaluated). The kappa coefficient of 0.334 (p < 0.001) shows sufficient agreement between IDx-DR and physician's image analysis based on the fundus photograph, considering all patients with IDx-DR analysis that could be evaluated. The comparison between IDx-DR and the physician's funduscopy under the same conditions shows a low agreement with a kappa value of 0.168 (p < 0.001).
The present study shows the possibilities and limitations of AI-assisted DR screening. A major limitation is that sufficient images cannot be obtained in miosis in approximately 40% of patients. When sufficient images were available the IDx-DR and ophthalmological diagnosis matched in more than 50% of cases. Underestimation of severity by IDx-DR occurred only rarely. For integration into an ophthalmologist's practice, this system seems suitable. Without access to an ophthalmologist the high rate of insufficient images in miosis represents an important limitation.</description><identifier>EISSN: 1433-0423</identifier><identifier>EISSN: 2731-7218</identifier><identifier>DOI: 10.1007/s00347-021-01556-5</identifier><identifier>PMID: 35080640</identifier><language>ger</language><publisher>Germany</publisher><ispartof>Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft, 2022-07, Vol.119 (7), p.705-713</ispartof><rights>2022. The Author(s).</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,27901,27902</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/35080640$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Paul, Sebastian</creatorcontrib><creatorcontrib>Tayar, Allam</creatorcontrib><creatorcontrib>Morawiec-Kisiel, Ewa</creatorcontrib><creatorcontrib>Bohl, Beathe</creatorcontrib><creatorcontrib>Großjohann, Rico</creatorcontrib><creatorcontrib>Hunfeld, Elisabeth</creatorcontrib><creatorcontrib>Busch, Martin</creatorcontrib><creatorcontrib>Pfeil, Johanna M</creatorcontrib><creatorcontrib>Dähmcke, Merlin</creatorcontrib><creatorcontrib>Brauckmann, Tara</creatorcontrib><creatorcontrib>Eilts, Sonja</creatorcontrib><creatorcontrib>Bründer, Marie-Christine</creatorcontrib><creatorcontrib>Grundel, Milena</creatorcontrib><creatorcontrib>Grundel, Bastian</creatorcontrib><creatorcontrib>Tost, Frank</creatorcontrib><creatorcontrib>Kuhn, Jana</creatorcontrib><creatorcontrib>Reindel, Jörg</creatorcontrib><creatorcontrib>Kerner, Wolfgang</creatorcontrib><creatorcontrib>Stahl, Andreas</creatorcontrib><title>Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center</title><title>Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft</title><addtitle>Ophthalmologe</addtitle><description>In 2018, IDx-DR was approved as a method to determine the degree of diabetic retinopathy (DR) using artificial intelligence (AI) by the FDA.
We integrated IDx-DR into the consultation at a diabetology focus clinic and report the agreement between IDx-DR and fundoscopy as well as IDx-DR and ophthalmological image assessment and the influence of different camera systems.
Adequate image quality in miosis was achieved more frequently with the Topcon camera (n = 456; NW400, Topcon Medical Systems, Oakland, NJ, USA) compared with the Zeiss camera (n = 47; Zeiss VISUCAM 500, Carl Zeiss Meditec AG, Jena, Germany). Overall, IDx-DR analysis in miosis was possible in approximately 60% of the patients. All patients in whom IDx-DR analysis in miosis was not possible could be assessed by fundoscopy with dilated pupils. Within the group of images that could be evaluated, there was agreement between IDx-DR and ophthalmic fundoscopy in approximately 55%, overestimation of severity by IDx-DR in approximately 40% and underestimation in approximately 4%. The sensitivity (specificity) for detecting severe retinopathy requiring treatment was 95.7% (89.1%) for cases with fundus images that could be evaluated and 65.2% (66.7%) when all cases were considered (including those without images in miosis which could be evaluated). The kappa coefficient of 0.334 (p < 0.001) shows sufficient agreement between IDx-DR and physician's image analysis based on the fundus photograph, considering all patients with IDx-DR analysis that could be evaluated. The comparison between IDx-DR and the physician's funduscopy under the same conditions shows a low agreement with a kappa value of 0.168 (p < 0.001).
The present study shows the possibilities and limitations of AI-assisted DR screening. A major limitation is that sufficient images cannot be obtained in miosis in approximately 40% of patients. When sufficient images were available the IDx-DR and ophthalmological diagnosis matched in more than 50% of cases. Underestimation of severity by IDx-DR occurred only rarely. For integration into an ophthalmologist's practice, this system seems suitable. Without access to an ophthalmologist the high rate of insufficient images in miosis represents an important limitation.</description><issn>1433-0423</issn><issn>2731-7218</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><recordid>eNo1kMtKA0EQRRtBTIz-gAvppZvR6udklhI0CgE3Zj30ozq2TGbG7s4if-O3-GUOGDdVFBwO9xYhNwzuGUD9kAGErCvgrAKmlK7UGZkzKUQFkosZucz5EwC0EvyCzISCJWgJc-K2GekQqEklhuii6WjsC3Zd3GHvcDpodgmxj_2OhiFRH43FEh1N0-yH0ZSPIzWFmp_vgpPEpOOJwUwdTq50Rc6D6TJen_aCbJ-f3lcv1eZt_bp63FQjk6xUusZaM_B6KbgUxluvQXqm0KKzQloHQTbOi6UOThnJdLBSNOCNAtFY4GJB7v68Yxq-DphLu4_ZTV1Mj8Mht1xz3mhWA5vQ2xN6sHv07Zjifkre_j9G_AJRlWV9</recordid><startdate>20220701</startdate><enddate>20220701</enddate><creator>Paul, Sebastian</creator><creator>Tayar, Allam</creator><creator>Morawiec-Kisiel, Ewa</creator><creator>Bohl, Beathe</creator><creator>Großjohann, Rico</creator><creator>Hunfeld, Elisabeth</creator><creator>Busch, Martin</creator><creator>Pfeil, Johanna M</creator><creator>Dähmcke, Merlin</creator><creator>Brauckmann, Tara</creator><creator>Eilts, Sonja</creator><creator>Bründer, Marie-Christine</creator><creator>Grundel, Milena</creator><creator>Grundel, Bastian</creator><creator>Tost, Frank</creator><creator>Kuhn, Jana</creator><creator>Reindel, Jörg</creator><creator>Kerner, Wolfgang</creator><creator>Stahl, Andreas</creator><scope>NPM</scope><scope>7X8</scope></search><sort><creationdate>20220701</creationdate><title>Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center</title><author>Paul, Sebastian ; Tayar, Allam ; Morawiec-Kisiel, Ewa ; Bohl, Beathe ; Großjohann, Rico ; Hunfeld, Elisabeth ; Busch, Martin ; Pfeil, Johanna M ; Dähmcke, Merlin ; Brauckmann, Tara ; Eilts, Sonja ; Bründer, Marie-Christine ; Grundel, Milena ; Grundel, Bastian ; Tost, Frank ; Kuhn, Jana ; Reindel, Jörg ; Kerner, Wolfgang ; Stahl, Andreas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p141t-67e7610d683243adbd604d15ebecb34bc0f49cd386fc5a416fb4390da5039b023</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>ger</language><creationdate>2022</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Paul, Sebastian</creatorcontrib><creatorcontrib>Tayar, Allam</creatorcontrib><creatorcontrib>Morawiec-Kisiel, Ewa</creatorcontrib><creatorcontrib>Bohl, Beathe</creatorcontrib><creatorcontrib>Großjohann, Rico</creatorcontrib><creatorcontrib>Hunfeld, Elisabeth</creatorcontrib><creatorcontrib>Busch, Martin</creatorcontrib><creatorcontrib>Pfeil, Johanna M</creatorcontrib><creatorcontrib>Dähmcke, Merlin</creatorcontrib><creatorcontrib>Brauckmann, Tara</creatorcontrib><creatorcontrib>Eilts, Sonja</creatorcontrib><creatorcontrib>Bründer, Marie-Christine</creatorcontrib><creatorcontrib>Grundel, Milena</creatorcontrib><creatorcontrib>Grundel, Bastian</creatorcontrib><creatorcontrib>Tost, Frank</creatorcontrib><creatorcontrib>Kuhn, Jana</creatorcontrib><creatorcontrib>Reindel, Jörg</creatorcontrib><creatorcontrib>Kerner, Wolfgang</creatorcontrib><creatorcontrib>Stahl, Andreas</creatorcontrib><collection>PubMed</collection><collection>MEDLINE - Academic</collection><jtitle>Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Paul, Sebastian</au><au>Tayar, Allam</au><au>Morawiec-Kisiel, Ewa</au><au>Bohl, Beathe</au><au>Großjohann, Rico</au><au>Hunfeld, Elisabeth</au><au>Busch, Martin</au><au>Pfeil, Johanna M</au><au>Dähmcke, Merlin</au><au>Brauckmann, Tara</au><au>Eilts, Sonja</au><au>Bründer, Marie-Christine</au><au>Grundel, Milena</au><au>Grundel, Bastian</au><au>Tost, Frank</au><au>Kuhn, Jana</au><au>Reindel, Jörg</au><au>Kerner, Wolfgang</au><au>Stahl, Andreas</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center</atitle><jtitle>Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft</jtitle><addtitle>Ophthalmologe</addtitle><date>2022-07-01</date><risdate>2022</risdate><volume>119</volume><issue>7</issue><spage>705</spage><epage>713</epage><pages>705-713</pages><eissn>1433-0423</eissn><eissn>2731-7218</eissn><abstract>In 2018, IDx-DR was approved as a method to determine the degree of diabetic retinopathy (DR) using artificial intelligence (AI) by the FDA.
We integrated IDx-DR into the consultation at a diabetology focus clinic and report the agreement between IDx-DR and fundoscopy as well as IDx-DR and ophthalmological image assessment and the influence of different camera systems.
Adequate image quality in miosis was achieved more frequently with the Topcon camera (n = 456; NW400, Topcon Medical Systems, Oakland, NJ, USA) compared with the Zeiss camera (n = 47; Zeiss VISUCAM 500, Carl Zeiss Meditec AG, Jena, Germany). Overall, IDx-DR analysis in miosis was possible in approximately 60% of the patients. All patients in whom IDx-DR analysis in miosis was not possible could be assessed by fundoscopy with dilated pupils. Within the group of images that could be evaluated, there was agreement between IDx-DR and ophthalmic fundoscopy in approximately 55%, overestimation of severity by IDx-DR in approximately 40% and underestimation in approximately 4%. The sensitivity (specificity) for detecting severe retinopathy requiring treatment was 95.7% (89.1%) for cases with fundus images that could be evaluated and 65.2% (66.7%) when all cases were considered (including those without images in miosis which could be evaluated). The kappa coefficient of 0.334 (p < 0.001) shows sufficient agreement between IDx-DR and physician's image analysis based on the fundus photograph, considering all patients with IDx-DR analysis that could be evaluated. The comparison between IDx-DR and the physician's funduscopy under the same conditions shows a low agreement with a kappa value of 0.168 (p < 0.001).
The present study shows the possibilities and limitations of AI-assisted DR screening. A major limitation is that sufficient images cannot be obtained in miosis in approximately 40% of patients. When sufficient images were available the IDx-DR and ophthalmological diagnosis matched in more than 50% of cases. Underestimation of severity by IDx-DR occurred only rarely. For integration into an ophthalmologist's practice, this system seems suitable. Without access to an ophthalmologist the high rate of insufficient images in miosis represents an important limitation.</abstract><cop>Germany</cop><pmid>35080640</pmid><doi>10.1007/s00347-021-01556-5</doi><tpages>9</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | EISSN: 1433-0423 |
| ispartof | Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft, 2022-07, Vol.119 (7), p.705-713 |
| issn | 1433-0423 2731-7218 |
| language | ger |
| recordid | cdi_proquest_miscellaneous_2622961701 |
| source | Springer Nature - Complete Springer Journals |
| title | Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-15T13%3A27%3A46IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Use%20of%20artificial%20intelligence%20in%20screening%20for%20diabetic%20retinopathy%20at%20a%C2%A0tertiary%20diabetes%20center&rft.jtitle=Der%20Ophthalmologe%20:%20Zeitschrift%20der%20Deutschen%20Ophthalmologischen%20Gesellschaft&rft.au=Paul,%20Sebastian&rft.date=2022-07-01&rft.volume=119&rft.issue=7&rft.spage=705&rft.epage=713&rft.pages=705-713&rft.eissn=1433-0423&rft_id=info:doi/10.1007/s00347-021-01556-5&rft_dat=%3Cproquest_pubme%3E2622961701%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2622961701&rft_id=info:pmid/35080640&rfr_iscdi=true |