[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-how-ai-models-handle-numbers-in-medical-records":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},3471,"how-ai-models-handle-numbers-in-medical-records","How AI Models Handle Numbers in Medical Records","A new study finds that hybrid encoding strategies outperform pure discrete or continuous approaches for numeric data in clinical AI models.","Transformers processing electronic health records are bad at math — and a new study maps exactly how bad, and what to do about it.\n\nResearchers systematically compared three strategies for encoding numeric values in transformer-based models trained on EHR data: discrete (binning numbers into categories), continuous (passing raw values), and hybrid approaches that combine both. They tested these against synthetic arithmetic tasks embedded in real clinical data, plus actual clinical prediction benchmarks. The headline finding: models that explicitly capture the relationship between a numeric value and its clinical concept — say, a glucose reading tied to a \"blood sugar\" token — perform best on precision-sensitive tasks, but only when the architecture allows it. A simpler hybrid approach, which keeps raw numeric values but bins them before projection, lands nearly as well and works across more settings.\n\nThe study matters because numeric lab values sit at the core of clinical decision-making, and the way a model encodes \"creatinine: 2.4\" shapes whether it can reason about kidney function at all. The researchers found that the optimal bin count follows a power-law tied to dataset size — a practical tuning rule that teams building clinical models can actually use. Crucially, gains from lab values varied by task, suggesting that adding more numeric signal does not uniformly improve predictions.\n\nThe consistent pattern across experiments was that models achieve \"good enough\" numeric computation rather than exact arithmetic — which means deployability and robustness tend to matter more than chasing precision. That is probably the right engineering tradeoff in healthcare, where a model that generalizes beats one that overfits to lab-value arithmetic on a narrow benchmark.","[\"ai\",\"healthcare\",\"machine-learning\",\"transformers\"]","2026-07-03T04:00:00.000Z","2026-07-03T06:48:04.527Z","2026-07-03T06:48:07.421Z","published",null,[],"ai",[24,26,27,28],"healthcare","machine-learning","transformers",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.01391",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"]