{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T18:22:03Z","timestamp":1792002123100,"version":"4.4.0"},"reference-count":14,"publisher":"American Medical Association (AMA)","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JAMA Netw Open"],"abstract":"<jats:sec id=\"ab-zoi250073-4\"><jats:title>Importance<\/jats:title><jats:p>While prospective studies have investigated the accuracy of artificial intelligence (AI) for detection of diabetic retinopathy (DR) and diabetic macular edema (DME), to date, little published data exist on the clinical performance of these algorithms.<\/jats:p><\/jats:sec><jats:sec id=\"ab-zoi250073-5\"><jats:title>Objective<\/jats:title><jats:p>To evaluate the clinical performance of an automated retinal disease assessment (ARDA) algorithm in the postdeployment setting at Aravind Eye Hospital in India.<\/jats:p><\/jats:sec><jats:sec id=\"ab-zoi250073-6\"><jats:title>Design, Setting, and Participants<\/jats:title><jats:p>This cross-sectional analysis involved an approximate 1% sample of fundus photographs from patients screened using ARDA. Images were graded via adjudication by US ophthalmologists for DR and DME, and ARDA\u2019s output was compared against the adjudicated grades at 45 sites in Southern India. Patients were randomly selected between January 1, 2019, and July 31, 2023.<\/jats:p><\/jats:sec><jats:sec id=\"ab-zoi250073-7\"><jats:title>Main Outcomes and Measures<\/jats:title><jats:p>Primary analyses were the sensitivity and specificity of ARDA for severe nonproliferative DR (NPDR) or proliferative DR (PDR). Secondary analyses focused on sensitivity and specificity for sight-threatening DR (STDR) (DME or severe NPDR or PDR).<\/jats:p><\/jats:sec><jats:sec id=\"ab-zoi250073-8\"><jats:title>Results<\/jats:title><jats:p>Among the 4537 patients with 4537 images with adjudicated grades, mean (SD) age was 55.2 (11.9) years and 2272 (50.1%) were male. Among the 3941 patients with gradable photographs, 683 (17.3%) had any DR, 146 (3.7%) had severe NPDR or PDR, 109 (2.8%) had PDR, and 398 (10.1%) had STDR. ARDA\u2019s sensitivity and specificity for severe NPDR or PDR were 97.0% (95% CI, 92.6%-99.2%) and 96.4% (95% CI, 95.7%-97.0%), respectively. Positive predictive value (PPV) was 50.7% and negative predictive value (NPV) was 99.9%. The clinically important miss rate for severe NPDR or PDR was 0% (eg, some patients with severe NPDR or PDR were interpreted as having moderate DR and referred to clinic). ARDA\u2019s sensitivity for STDR was 95.9% (95% CI, 93.0%-97.4%) and specificity was 94.9% (95% CI, 94.1%-95.7%); PPV and NPV were 67.9% and 99.5%, respectively.<\/jats:p><\/jats:sec><jats:sec id=\"ab-zoi250073-9\"><jats:title>Conclusions and Relevance<\/jats:title><jats:p>In this cross-sectional study investigating the clinical performance of ARDA, sensitivity and specificity for severe NPDR and PDR exceeded 96% and caught 100% of patients with severe\u2009\u2009NPDR and PDR for ophthalmology referral. This preliminary large-scale postmarketing report of the performance of ARDA after screening 600\u202f000 patients in India underscores the importance of monitoring and publication an algorithm's clinical performance, consistent with recommendations by regulatory bodies.<\/jats:p><\/jats:sec>","DOI":"10.1001\/jamanetworkopen.2025.0984","type":"journal-article","created":{"date-parts":[[2025,3,19]],"date-time":"2025-03-19T15:17:59Z","timestamp":1742397479000},"page":"e250984","source":"Crossref","is-referenced-by-count":21,"title":["Performance of a Deep Learning Diabetic Retinopathy Algorithm in India"],"prefix":"10.1001","volume":"8","author":[{"given":"Arthur","family":"Brant","sequence":"first","affiliation":[{"name":"Oregon Vision and Pharmacy LLC, Keizer"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Preeti","family":"Singh","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiang","family":"Yin","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lu","family":"Yang","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jay","family":"Nayar","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Divleen","family":"Jeji","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yossi","family":"Matias","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Greg S.","family":"Corrado","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dale R.","family":"Webster","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sunny","family":"Virmani","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anchintha","family":"Meenu","sequence":"additional","affiliation":[{"name":"Aravind Eye Hospital, Madurai, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Naresh Babu","family":"Kannan","sequence":"additional","affiliation":[{"name":"Aravind Eye Hospital, Madurai, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan","family":"Krause","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Florence","family":"Thng","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lily","family":"Peng","sequence":"additional","affiliation":[{"name":"Verily Life Sciences LLC, South San Francisco, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yun","family":"Liu","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kasumi","family":"Widner","sequence":"additional","affiliation":[{"name":"Google LLC, Mountain View, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kim","family":"Ramasamy","sequence":"additional","affiliation":[{"name":"Aravind Eye Hospital, Madurai, India"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10","published-online":{"date-parts":[[2025,3,19]]},"reference":[{"issue":"12","key":"zoi250073r1","doi-asserted-by":"publisher","first-page":"e1764","DOI":"10.1016\/S2214-109X(22)00411-9","article-title":"Prevalence of diabetic retinopathy in India stratified by known and undiagnosed diabetes, urban-rural locations, and socioeconomic indices: results from the SMART India population-based cross-sectional screening study.","volume":"10","author":"Raman","year":"2022","journal-title":"Lancet Glob Health"},{"issue":"22","key":"zoi250073r2","doi-asserted-by":"publisher","first-page":"2402","DOI":"10.1001\/jama.2016.17216","article-title":"Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.","volume":"316","author":"Gulshan","year":"2016","journal-title":"JAMA"},{"issue":"104930","key":"zoi250073r4","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2022.104930","article-title":"Assessing the effects of data drift on the performance of machine learning models used in clinical sepsis prediction.","volume":"173","author":"Rahmani","year":"2023","journal-title":"Int J Med Inform"},{"issue":"3","key":"zoi250073r5","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1056\/NEJMc2104626","article-title":"The clinician and dataset shift in artificial intelligence.","volume":"385","author":"Finlayson","year":"2021","journal-title":"N Engl J Med"},{"issue":"20","key":"zoi250073r6","doi-asserted-by":"publisher","first-page":"2191","DOI":"10.1001\/jama.2013.281053","article-title":"World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects.","volume":"310","author":"World Medical Association","year":"2013","journal-title":"JAMA"},{"issue":"9","key":"zoi250073r7","doi-asserted-by":"publisher","first-page":"1677","DOI":"10.1016\/S0161-6420(03)00475-5","article-title":"Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales.","volume":"110","author":"Wilkinson","year":"2003","journal-title":"Ophthalmology"},{"issue":"4","key":"zoi250073r8","doi-asserted-by":"publisher","first-page":"e235","DOI":"10.1016\/S2589-7500(22)00017-6","article-title":"Real-time diabetic retinopathy screening by deep learning in a multisite national screening programme: a prospective interventional cohort study.","volume":"4","author":"Ruamviboonsuk","year":"2022","journal-title":"Lancet Digit Health"},{"key":"zoi250073r9","doi-asserted-by":"publisher","DOI":"10.1155\/2020\/8839376","article-title":"Longitudinal screening for diabetic retinopathy in a nationwide screening program: comparing deep learning and human graders.","volume":"2020","author":"Limwattanayingyong","year":"2020","journal-title":"J Diabetes Res"},{"issue":"1","key":"zoi250073r10","doi-asserted-by":"publisher","first-page":"68","DOI":"10.1038\/s41746-019-0146-5","article-title":"Author correction: deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program.","volume":"2","author":"Ruamviboonsuk","year":"2019","journal-title":"NPJ Digit Med"},{"issue":"22","key":"zoi250073r11","doi-asserted-by":"publisher","first-page":"2402","DOI":"10.1001\/jama.2016.17216","article-title":"Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.","volume":"316","author":"Gulshan","year":"2016","journal-title":"JAMA"},{"issue":"9","key":"zoi250073r12","doi-asserted-by":"publisher","first-page":"987","DOI":"10.1001\/jamaophthalmol.2019.2004","article-title":"Performance of a deep-learning algorithm vs manual grading for detecting diabetic retinopathy in India.","volume":"137","author":"Gulshan","year":"2019","journal-title":"JAMA Ophthalmol"},{"issue":"12","key":"zoi250073r14","doi-asserted-by":"publisher","first-page":"1171","DOI":"10.1001\/jamaophthalmol.2024.4493","article-title":"Use of artificial intelligence-based detection of diabetic retinopathy in the US.","volume":"142","author":"Shah","year":"2024","journal-title":"JAMA Ophthalmol"},{"key":"zoi250073r13","volume-title":"The Landmark Trials: Diabetic Retinopathy Study and Early Treatment Diabetic Retinopathy Study","author":"Royle","year":"2015"},{"key":"zoi250073r3","unstructured":"U.S. Food and Drug Administration. Good machine learning practice for medical device development: guiding principles. October 20, 2023. Accessed November 29, 2024. https:\/\/www.fda.gov\/medical-devices\/software-medical-device-samd\/good-machine-learning-practice-medical-device-development-guiding-principles"}],"container-title":["JAMA Network Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/articlepdf\/2831702\/brant_2025_oi_250073_1744295100.15536.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,14]],"date-time":"2025-04-14T15:00:55Z","timestamp":1744642855000},"score":1,"resource":{"primary":{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/fullarticle\/2831702"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,19]]},"references-count":14,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,3,3]]}},"URL":"https:\/\/doi.org\/10.1001\/jamanetworkopen.2025.0984","relation":{},"ISSN":["2574-3805"],"issn-type":[{"value":"2574-3805","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,19]]}}}