{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T20:42:01Z","timestamp":1792096921070,"version":"4.4.0"},"reference-count":26,"publisher":"American Medical Association (AMA)","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JAMA Netw Open"],"abstract":"\n Importance<\/jats:title>\n Retinopathy of prematurity (ROP) is the leading cause of preventable childhood blindness worldwide. If detected and treated early, ROP-associated blindness is preventable; however, identifying patients who might respond to treatment requires screening over time, which is challenging in low-resource settings where access to pediatric ophthalmologists and pediatric ocular imaging cameras is limited.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n To develop and assess the performance of a machine learning algorithm that uses smartphone-collected videos to perform retinal screening for ROP in low-resource settings.<\/jats:p>\n <\/jats:sec>\n \n Design, Setting, and Participants<\/jats:title>\n This diagnostic study used smartphone-obtained videos of fundi in premature neonates with and without ROP in Mexico and Argentina between May 12, 2020, and October 31, 2023. Machine-learning (ML)\u2013driven algorithms were developed to process a video, identify the best frames within the video, and use those frames to determine whether ROP was likely or not. Eligible neonates born with gestational age less than 36 weeks or birth weight less than 1500 g were included on the study.<\/jats:p>\n <\/jats:sec>\n \n Exposures<\/jats:title>\n An ML algorithm applied to a smartphone-obtained video.<\/jats:p>\n <\/jats:sec>\n \n Main Outcomes and Measures<\/jats:title>\n The ML algorithms\u2019 ability to identify high-quality retinal images and classify those images as indicating ROP or not at the frame and patient levels, measured by accuracy, specificity, and sensitivity, compared with classifications from 3 pediatric ophthalmologists.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n A total of 524 videos were collected for 512 neonates with median gestational age of 32 weeks (range, 25-36 weeks) and median birth weight of 1610 g (range, 580-2800 g). The frame selection model identified high-quality retinal images from 397 of 456 videos (87.1%; 95% CI, 84.0%-90.1%) reserved for testing model performance. Across all test videos, 97.4% (95% CI, 96.7%-98.1%) of high-quality retinal images selected by the model contained fundus images. At the frame level, the ROP classifier model had a sensitivity of 76.7% (95% CI, 69.9%-83.5%); at the patient level, the classifier model had a sensitivity of 93.3% (95% CI, 86.4%-100%). At both levels, the model\u2019s sensitivity was higher than that for the panel of pediatric ophthalmologists (frame level: 71.4% [95% CI, 64.1%-78.7%]; patient level: 73.3% [95% CI, 61.0%-85.6%]). Specificity and accuracy were higher for ophthalmologist classification vs the ML model.<\/jats:p>\n <\/jats:sec>\n \n Conclusions and Relevance<\/jats:title>\n In this diagnostic study, a process that used smartphone-collected videos of premature neonates\u2019 fundi to determine whether high-quality retinal images were present had high sensitivity to classify such images as indicating or not indicating ROP but lower specificity and accuracy than ophthalmologist assessment. This process costs a fraction of the current process for retinal image collection and classification and could be used to expand access to ROP screening in low-resource settings, with potential to help prevent the most common cause of preventable childhood blindness.<\/jats:p>\n <\/jats:sec>","DOI":"10.1001\/jamanetworkopen.2025.7831","type":"journal-article","created":{"date-parts":[[2025,4,29]],"date-time":"2025-04-29T11:30:45Z","timestamp":1745926245000},"page":"e257831","source":"Crossref","is-referenced-by-count":23,"title":["AI-Enabled Screening for Retinopathy of Prematurity in Low-Resource Settings"],"prefix":"10.1001","volume":"8","author":[{"given":"Anthony","family":"Ortiz","sequence":"first","affiliation":[{"name":"Microsoft AI for Good Lab, Redmond, Washington"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Susana","family":"Pati\u00f1o","sequence":"additional","affiliation":[{"name":"Business Data Evolution, Mexico City, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeh\u00fa","family":"Torres","sequence":"additional","affiliation":[{"name":"Microsoft AI for Good Lab, Redmond, Washington"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan","family":"M\u00e1rmol","sequence":"additional","affiliation":[{"name":"Business Data Evolution, Mexico City, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carlos","family":"Serafin","sequence":"additional","affiliation":[{"name":"Business Data Evolution, Mexico City, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rahul","family":"Dodhia","sequence":"additional","affiliation":[{"name":"Microsoft AI for Good Lab, Redmond, Washington"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gabriela","family":"Saidman","sequence":"additional","affiliation":[{"name":"Ophthalmology Department, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vanina","family":"Schbib","sequence":"additional","affiliation":[{"name":"Ophthalmology Department, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brenda","family":"Pe\u00f1a","sequence":"additional","affiliation":[{"name":"Ophthalmology Department, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guillermo","family":"Monteoliva","sequence":"additional","affiliation":[{"name":"Ophthalmology Department, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mar\u00eda Ana","family":"Martinez-Castellanos","sequence":"additional","affiliation":[{"name":"Asociaci\u00f3n Para Evitar la Ceguera en M\u00e9xico, Mexico City, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William B.","family":"Weeks","sequence":"additional","affiliation":[{"name":"Microsoft AI for Good Lab, Redmond, Washington"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan M.","family":"Lavista Ferres","sequence":"additional","affiliation":[{"name":"Microsoft AI for Good Lab, Redmond, Washington"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10","published-online":{"date-parts":[[2025,4,29]]},"reference":[{"issue":"9902","key":"zoi250289r1","doi-asserted-by":"publisher","first-page":"1445","DOI":"10.1016\/S0140-6736(13)60178-6","article-title":"Retinopathy of prematurity.","volume":"382","author":"Hellstr\u00f6m","year":"2013","journal-title":"Lancet"},{"issue":"3","key":"zoi250289r2","first-page":"187","article-title":"Retinopathy of prematurity: a review of current screening guidelines and treatment options.","volume":"113","author":"Broxterman","year":"2016","journal-title":"Mo Med"},{"issue":"11","key":"zoi250289r3","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.32007","article-title":"A review on the incidence and related risk factors of retinopathy of prematurity across various countries.","volume":"14","author":"Nair","year":"2022","journal-title":"Cureus"},{"issue":"6","key":"zoi250289r4","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1053\/j.semperi.2019.05.002","article-title":"Epidemiology of ROP update\u2014Africa is the new frontier.","volume":"43","author":"Gilbert","year":"2019","journal-title":"Semin Perinatol"},{"key":"zoi250289r5","first-page":"1687","article-title":"Prevalence of retinopathy of prematurity in Latin America.","volume":"5","author":"Carrion","year":"2011","journal-title":"Clin Ophthalmol"},{"issue":"26","key":"zoi250289r6","doi-asserted-by":"publisher","first-page":"2515","DOI":"10.1056\/NEJMra1208129","article-title":"Mechanisms and management of retinopathy of prematurity.","volume":"367","author":"Hartnett","year":"2012","journal-title":"N Engl J Med"},{"issue":"1","key":"zoi250289r7","doi-asserted-by":"publisher","DOI":"10.1136\/bmjpo-2023-002267","article-title":"Global, regional and national burden of retinopathy of prematurity among childhood and adolescent: a spatiotemporal analysis based on the Global Burden of Disease Study 2019.","volume":"8","author":"Wang","year":"2024","journal-title":"BMJ Paediatr Open"},{"issue":"1","key":"zoi250289r8","doi-asserted-by":"publisher","first-page":"35","DOI":"10.5409\/wjcp.v5.i1.35","article-title":"Retinopathy of prematurity: past, present and future.","volume":"5","author":"Shah","year":"2016","journal-title":"World J Clin Pediatr"},{"issue":"3","key":"zoi250289r9","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1001\/jamaophthalmol.2022.6010","article-title":"Access to pediatric ophthalmological care by geographic distribution and US population demographic characteristics in 2022.","volume":"141","author":"Walsh","year":"2023","journal-title":"JAMA Ophthalmol"},{"issue":"4","key":"zoi250289r10","doi-asserted-by":"publisher","first-page":"725","DOI":"10.1016\/j.pcl.2020.04.012","article-title":"Workforce shortage for retinopathy of prematurity care and emerging role of telehealth and artificial intelligence.","volume":"67","author":"Barrero-Castillero","year":"2020","journal-title":"Pediatr Clin North Am"},{"issue":"6","key":"zoi250289r11","first-page":"322","article-title":"A multi-country, cross-sectional observational study of retinopathy of prematurity in Latin America and the Caribbean.","volume":"39","author":"Arnesen","year":"2016","journal-title":"Rev Panam Salud Publica"},{"issue":"7","key":"zoi250289r12","doi-asserted-by":"publisher","first-page":"e69","DOI":"10.1016\/j.ophtha.2022.02.008","article-title":"Artificial intelligence for retinopathy of prematurity: validation of a vascular severity scale against international expert diagnosis.","volume":"129","author":"Campbell","year":"2022","journal-title":"Ophthalmology"},{"issue":"8","key":"zoi250289r13","doi-asserted-by":"publisher","first-page":"791","DOI":"10.1001\/jamaophthalmol.2022.2135","article-title":"External validation of a retinopathy of prematurity screening model using artificial intelligence in 3 low- and middle-income populations.","volume":"140","author":"Coyner","year":"2022","journal-title":"JAMA Ophthalmol"},{"issue":"12","key":"zoi250289r14","doi-asserted-by":"publisher","first-page":"2518","DOI":"10.1038\/s41433-022-02366-y","article-title":"Artificial intelligence for the diagnosis of retinopathy of prematurity: a systematic review of current algorithms.","volume":"37","author":"Ramanathan","year":"2023","journal-title":"Eye (Lond)"},{"key":"zoi250289r16","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1016\/j.ebiom.2018.08.033","article-title":"Automated retinopathy of prematurity screening using deep neural networks.","volume":"35","author":"Wang","year":"2018","journal-title":"EBioMedicine"},{"key":"zoi250289r17","first-page":"1224","article-title":"Deep learning for image quality assessment of fundus images in retinopathy of prematurity.","volume":"2018","author":"Coyner","year":"2018","journal-title":"AMIA Annu Symp Proc"},{"issue":"4","key":"zoi250289r19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jaapos.2018.07.215","article-title":"ROP telemedicine m-health network with smartphone holder helmet device: hands-free digital indirect ophthalmoscopy technique.","volume":"22","author":"Monteoliva","year":"2018","journal-title":"J AAPOS"},{"issue":"4","key":"zoi250289r20","doi-asserted-by":"publisher","first-page":"945","DOI":"10.3390\/diagnostics12040945","article-title":"Comparison of RetCam and smartphone-based photography for retinopathy of prematurity screening.","volume":"12","author":"Lin","year":"2022","journal-title":"Diagnostics (Basel)"},{"issue":"5","key":"zoi250289r21","doi-asserted-by":"publisher","first-page":"690","DOI":"10.1109\/TPAMI.1987.4767964","article-title":"The adaptive Hough transform.","volume":"9","author":"Illingworth","year":"1987","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"1","key":"zoi250289r24","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on image data augmentation for deep learning.","volume":"6","author":"Shorten","year":"2019","journal-title":"J Big Data"},{"key":"zoi250289r15","author":"Worrall"},{"key":"zoi250289r22","author":"He","year":"2016"},{"key":"zoi250289r23","author":"Kingma"},{"key":"zoi250289r25","author":"Guo","year":"2017"},{"key":"zoi250289r18","unstructured":"Open-label, randomized, 2-arm, controlled study to assess the efficacy, safety, and tolerability of intravitreal (IVT) aflibercept compared to laser photocoagulation in patients with retinopathy of prematurity (ROP). ClinicalTrials.gov identifier: NCT04004208. June 23, 2020. Accessed March 29, 2024. https:\/\/classic.clinicaltrials.gov\/ProvidedDocs\/08\/NCT04004208\/Prot_000.pdf"},{"key":"zoi250289r26","unstructured":"Raschka? S. Model evaluation, model selection, and algorithm selection in machine learning.? arXiv. Preprint posted online November 13, 2018. doi:10.48550\/arXiv.1811.12808"}],"container-title":["JAMA Network Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/articlepdf\/2833218\/ortiz_2025_oi_250289_1744987114.75592.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,29]],"date-time":"2025-04-29T11:30:49Z","timestamp":1745926249000},"score":1,"resource":{"primary":{"URL":"https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/fullarticle\/2833218"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,29]]},"references-count":26,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,4,1]]}},"URL":"https:\/\/doi.org\/10.1001\/jamanetworkopen.2025.7831","relation":{},"ISSN":["2574-3805"],"issn-type":[{"value":"2574-3805","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,29]]}}}