{"ok":true,"nota":{"id":"57d799e5723c2992","slug":"una-ia-lee-historias-clinicas-y-arma-la-lista-de-enfermedades-de-cada--57d799","titulo":"Una IA lee historias clínicas y arma la lista de enfermedades de cada paciente con sus fechas","resumen":"Investigadores de Berna probaron un modelo de lenguaje para extraer diagnósticos y sus fechas de notas médicas escritas en texto libre. Analizaron 2343 registros de 248 pacientes con problemas de sueño y sus otras afecciones.","fuente":"medrxiv","fuenteNombre":"medRxiv · preprints","color":"#1f7a4d","autor":"Filchenko, I., Dennstaedt, F., Zub, K., Pompizii, F. I., van der Meer, J., Adamantidis, A., Cihoric, N., Schmidt, M. H.,","url":"https://www.medrxiv.org/content/10.64898/2026.10.25.26364496v1?rss=1","imagen":"/hoy/portada/57d799e5723c2992","tema":"ciencia","subtemas":["inteligencia artificial","historia clínica","trastornos del sueño","registros médicos","comorbilidades"],"entidades":["Universidad de Berna","Bern Sleep-Wake Registry"],"formato":"dato","entretenimiento":false,"pubTs":"2026-10-26T00:00:00.000Z","indexTs":"2026-10-26T15:36:30.917Z","hash":"e0d7a44f0fed2db70055c69a23f7d6e79ed3b7115e5b15ed7c57c3ad8431d1bc","tsa":{"time":"Oct 26 15:37:02 2026 GMT","provider":"freetsa.org","sobre":"sello"},"stats":{"views":0,"likes":0,"clicks":0,"dwell":0},"tipo":"motor","registro":"hallazgo","porQueImporta":"Hasta ahora, buena parte de la información de una historia clínica quedaba escondida en notas escritas a mano o en texto libre, difícil de ordenar. Si una herramienta puede rescatar cuándo apareció cada problema, el médico ve la evolución del paciente sin revisar carpeta por carpeta. Para quien atiende en un hospital cargado, eso puede significar un diagnóstico más rápido y menos errores. El límite serio: el estudio mide qué tan bien la IA acierta con las fechas en un grupo específico de pacientes, y todavía falta comprobar que funcione igual en otros hospitales.","quienFirma":"Universidad de Berna","confianza":"preprint","consulta":{"pregunta":"¿Le confiarías tu historia clínica a una IA?"}},"sello":{"hash":"e0d7a44f0fed2db70055c69a23f7d6e79ed3b7115e5b15ed7c57c3ad8431d1bc","sello":"460944cf45b9a6e17d93c8577a648d281f8c0355a591bfdd4754940baba3e550","beacon":{"drand":{"cadena":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","ronda":6586600,"aleatorio":"b890a385cf92974374211108b695f20f26b378db0776577c156830220ff127b8","tiempo":"2026-10-26T15:37:00.000Z"},"nist":{"cadena":2,"pulso":2000493,"salida":"561AB66CD420DAD021A79CE9030A7B3512C7392399F0542651AB8883F6508DDB391A1A35AF6893D1E04E4FDDE2374CB221AE98089FAC353C5E66C30F2D8BF390","tiempo":"2026-10-26T15:36:00.000Z"},"noAntes":"2026-10-26T15:37:00.000Z"},"verificacion":{"drand":true,"nist":true},"tsa":{"time":"Oct 26 15:37:02 2026 GMT","provider":"freetsa.org","sobre":"sello"},"original":{"titulo":"Bern Sleep-Wake Registry: temporal extraction of comorbidities from unstructured clinical text using a large language model","descripcion":"Introduction Large language models (LLMs) may enable the extraction of clinically relevant information from unstructured electronic health records, but most previous work focused on identifying diagnoses without preserving their temporal context. We evaluated the ability of an LLM to extract diagnoses and associated dates from longitudinal clinical reports. Methods In an observational study using the Bern Sleep-Wake Registry, diagnostic annotations from 248 patients comprising 2343 clinical entries were analyzed. Six common comorbidity groups, subdivided into 32 hierarchical diagnostic categories, were evaluated. Using gpt-oss-20b, we extracted diagnosis presence, the number of diagnostic in","pubDate":"2026-10-26","url":"https://www.medrxiv.org/content/10.64898/2026.10.25.26364496v1?rss=1"},"indexado":"2026-10-26T15:36:30.917Z"}}