[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-system-spots-cavities-in-x-rays-with-97-precision":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},6647,"ai-system-spots-cavities-in-x-rays-with-97-precision","AI System Spots Cavities in X-Rays With 97% Precision","A two-stage deep learning model detects cavities in X-rays with high precision, tackling the data-labeling bottleneck limiting clinical AI adoption.","Researchers built an AI pipeline that finds cavities in dental X-rays almost as reliably as a trained dentist reading the image twice.\n\nThe system chains two neural networks together: one locates individual teeth in a panoramic X-ray, the other traces the exact outline of any cavity within them. The team trained it on 3,000 panoramic images, mixing expert-labeled data with masks generated by converting existing polygon annotations into pixel-level maps, a workaround for how scarce dentist-annotated training data actually is. The result scored 0.9013 on IoU and 0.9482 on the Dice coefficient, both measures of how closely a predicted region matches the real one, along with 94.33 percent recall and 97.74 percent precision. The researchers report this beats prior caries-detection methods and cuts false positives.\n\nThe real contribution here isn't the accuracy numbers, it's the workaround. Pixel-level dental annotations are slow and expensive to produce, since tracing a cavity's exact border takes far more expertise than just boxing a tooth. A method for turning cheaper, already-existing annotation formats into usable training masks could let other labs build similar tools without commissioning fresh labeling work from scratch.\n\nOne caveat: this is a single arXiv preprint, not a peer-reviewed clinical study, so treat talk of supporting real diagnostic decisions as a future goal, not a current one, until it's tested on X-rays from clinics outside the original dataset.","[\"ai\",\"healthcare\",\"computer-vision\",\"research\"]","2026-09-17T04:00:00.000Z","2026-09-18T04:15:12.559Z","2026-09-18T04:15:24.458Z","published",null,[],"ai",[24,26,27,28],"healthcare","computer-vision","research",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.18952",0,{"sections":35},[36,40,44,49,54,58,62,67,72,76,81,86,91,96],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",3852,"2026-09-17T08:27:09.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",648,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",338,"2026-09-11T04:00:00.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":18},"Hardware","hardware",154,{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",114,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":71},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":73,"slug":74,"count":75,"latest_published_at":18},"Dev Tools","dev-tools",73,{"name":77,"slug":78,"count":79,"latest_published_at":80},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]