{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T11:41:40Z","timestamp":1762256500864,"version":"build-2065373602"},"reference-count":0,"publisher":"SEMES (Sociedad Espanola de Medicina de Urgencias y Emergencias)","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Emergencias"],"published-print":{"date-parts":[[2025,11,4]]},"abstract":"<jats:p>Introducci\u00f3n. Evaluar el rendimiento diagn\u00f3stico para la detecci\u00f3n de anomal\u00edas de dos sistemas comerciales de inteligencia artificial (IA), ChestView para radiograf\u00edas de t\u00f3rax (RxT) y BoneView para radiograf\u00edas \u00f3seas (RxO), en un servicio de urgencias hospitalario (SUH), y comparar su validez con la de observadores de diferente perfil profesional y\nexperiencia: urgenci\u00f3logos, radi\u00f3logos en formaci\u00f3n y radi\u00f3logos expertos.\nM\u00e9todo. Estudio de evaluaci\u00f3n de pruebas diagn\u00f3sticas en una selecci\u00f3n aleatoria de 346 RxT y 261 RxO solicitadas\nen urgencias. Las exploraciones fueron analizadas de forma independiente por los sistemas de IA y los diferentes observadores. El diagn\u00f3stico de referencia (gold standard) fue establecido mediante consenso por tres radi\u00f3logos, recurriendo a otras pruebas de imagen disponibles o informaci\u00f3n cl\u00ednica cuando era necesario. Se calcularon y compararon la sensibilidad, especificidad, y los valores predictivos positivo y negativo (VPN).\nResultados. Para la RxT, la IA (ChestView) mostr\u00f3 una sensibilidad global (64,4%) significativamente superior a la de\nlos m\u00e9dicos de urgencias (49,2%; p = 0,018), aunque inferior a la del radi\u00f3logo experto (83,9%; p &lt; 0,001). El rendimiento fue notable para la detecci\u00f3n de n\u00f3dulos\/masas (sensibilidad 80,0%) y neumot\u00f3rax (VPN 99,7%), pero inferior para consolidaciones (sensibilidad 40,4%). Para RxO, la IA (BoneView) alcanz\u00f3 una sensibilidad para la detecci\u00f3n\nde fracturas (87,5%) superior a la del radi\u00f3logo experto (77,1%), con un VPN del 96,9%. Sin embargo, su rendimiento fue menor para la detecci\u00f3n de luxaciones (sensibilidad 60,0%) y derrames articulares (sensibilidad 25,0%).\nConclusi\u00f3n. Los sistemas de IA evaluados demuestran un rendimiento cl\u00ednicamente relevante en el entorno de urgencias, y mejorar significativamente la capacidad diagn\u00f3stica de los urgenci\u00f3logos. Su elevada sensibilidad para la detecci\u00f3n de fracturas y su elevado VPN para los n\u00f3dulos pulmonares, neumot\u00f3rax y fracturas hacen que se consolide como un sistema de seguridad de alto impacto<\/jats:p>","DOI":"10.55633\/s3me\/093.2025","type":"journal-article","created":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T11:34:04Z","timestamp":1762256044000},"source":"Crossref","is-referenced-by-count":0,"title":["Reliability and validity of an artificial intelligence-assisted system for the detection of abnormalities in chest and bone radiographs in an emergency department"],"prefix":"10.55633","author":[{"given":"Raissa de F\u00e1tima","family":"Silva Afons","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pilar","family":"Gallardo-Rodr\u00edguez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bego\u00f1a","family":"Espinosa","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alejandro","family":"Bautista","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Javier","family":"Serrano","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"M\u00f3nica","family":"Veguillas","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mar\u00eda","family":"Corell","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ra\u00fal","family":"Garrido Chamorro","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Juan","family":"Arenas Jim\u00e9nez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Celia","family":"Astor Rodr\u00edguez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"\u00c1lvaro","family":"Abell\u00f3n Fern\u00e1ndez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"\u00c1lvaro","family":"Palaz\u00f3n Ru\u00edz de Tremi\u00f1o","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mar\u00eda Javiera","family":"Garfias Baladr\u00f3n","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"V\u00edctor","family":"Marquina Arribas","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pablo","family":"Chico-S\u00e1nchez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paula","family":"Gras Valenti","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Miguel","family":"Cabrer Gonz\u00e1lez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Carlos","family":"Mart\u00ednez Riera","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Moliner Mateu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jos\u00e9 Mar\u00eda","family":"Salinas Serrano","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Emilio","family":"Vivancos Rubio","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bernardo","family":"Valdivieso Mart\u00ednez","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Luis","family":"Concepci\u00f3n-Aramendia","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jos\u00e9","family":"S\u00e1nchez-Pay\u00e1","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pere","family":"Llorens","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"33356","published-online":{"date-parts":[[2025,11,4]]},"container-title":["Emergencias"],"original-title":["Fiabilidad y validez de un sistema asistido por inteligencia artificial para la detecci\u00f3n de anomal\u00edas en las radiograf\u00edas de t\u00f3rax y \u00f3seas en un servicio de urgencias hospitalario"],"deposited":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T11:34:05Z","timestamp":1762256045000},"score":1,"resource":{"primary":{"URL":"https:\/\/revistaemergencias.org\/articulo\/fiabilidad-y-validez-de-un-sistema-asistido-por-inteligencia-artificial-para-la-deteccion-de-anomalias-en-las-radiografias-de-torax-y-oseas-en-un-servicio-de-urgencias-hospitalario\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,4]]},"references-count":0,"URL":"https:\/\/doi.org\/10.55633\/s3me\/093.2025","relation":{},"ISSN":["1137-6821","2386-5857"],"issn-type":[{"value":"1137-6821","type":"print"},{"value":"2386-5857","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,4]]}}}