[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-combo-beats-solo-classifier-for-rop-screening-in-kenya":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},4353,"ai-combo-beats-solo-classifier-for-rop-screening-in-kenya","AI combo beats solo classifier for ROP screening in Kenya","A study on 121 Kenyan preterm infants finds that pairing image classifiers with vessel segmentation models catches more cases with fewer false alarms.","An AI screening study for a blinding eye condition common in premature infants shows that no single model does the job alone.\n\nResearchers analyzed 237 eyes from 121 preterm infants in Kenya, testing eleven different AI configurations for detecting \"plus disease\" - a severe form of retinopathy of prematurity (ROP) marked by abnormal blood vessel swelling and twisting. ROP is a leading preventable cause of childhood blindness, and diagnosing its worst stages currently requires specialists that sub-Saharan Africa largely does not have. The team tested standalone image classifiers, vessel segmentation models, and hybrid pipelines that combine both. A probability ensemble combining the two approaches hit 69.2% sensitivity and 91.4% specificity, outperforming image classifiers used alone.\n\nThe gap matters because over-referral is a real operational cost in low-resource settings - sending too many infants for specialist review that does not exist wastes the system. The finding that classifiers catch more cases but generate more false alarms, while segmentation adds precision, points toward a workflow rather than a single model as the right unit of deployment.\n\nThe dataset is small and single-site, and the authors call for prospective multi-site validation before any clinical rollout - a caveat worth taking seriously given how often AI medical tools trained in one context fail quietly in another.","[\"ai\",\"healthcare\",\"computer-vision\",\"africa\"]","2026-07-08T04:00:00.000Z","2026-07-08T06:54:46.865Z","2026-07-08T06:54:49.824Z","published",null,[],"ai",[24,26,27,28],"healthcare","computer-vision","africa",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.05825",0,{"sections":35},[36,40,45,50,55,60,65,70,75,80,85,89,94,99],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":76,"slug":77,"count":78,"latest_published_at":79},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":86,"slug":87,"count":83,"latest_published_at":88},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]