{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,18]],"date-time":"2026-11-18T20:16:54Z","timestamp":1795033014007,"version":"4.10.2"},"reference-count":7,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T00:00:00Z","timestamp":1750982400000},"content-version":"unspecified","delay-in-days":26,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","award":["PI24\/01268"],"award-info":[{"award-number":["PI24\/01268"]}]},{"DOI":"10.13039\/501100011033","name":"Agencia Estatal de Investigaci\u00f3n","doi-asserted-by":"publisher","award":["CPP2021-008340"],"award-info":[{"award-number":["CPP2021-008340"]}],"id":[{"id":"10.13039\/501100011033","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100008666","name":"Fundaci\u00f3 la Marat\u00f3 de TV3","doi-asserted-by":"publisher","award":["202118-30-31-32"],"award-info":[{"award-number":["202118-30-31-32"]}],"id":[{"id":"10.13039\/100008666","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Health Care Inform"],"accepted":{"date-parts":[[2025,6,11]]},"published-print":{"date-parts":[[2025,6]]},"abstract":"<jats:sec>\n                  <jats:title>Objectives<\/jats:title>\n                  <jats:p>The objective of this study is to evaluate whether large language models (LLMs) can achieve performance comparable to expert-developed deep neural networks in detecting flow starvation (FS) asynchronies during mechanical ventilation.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>Popular LLMs (GPT-4, Claude-3.5, Gemini-1.5, DeepSeek-R1) were tested on a dataset of 6500 airway pressure cycles from 28 patients, classifying breaths into three FS categories. They were also tasked with generating executable code for one-dimensional convolutional neural network (CNN-1D) and Long Short-Term Memory networks. Model performances were assessed using repeated holdout validation and compared with expert-developed models.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>LLMs performed poorly in direct FS classification (accuracy: GPT-4: 0.497; Claude-3.5: 0.627; Gemini-1.5: 0.544, DeepSeek-R1: 0.520). However, Claude-3.5-generated CNN-1D code achieved the highest accuracy (0.902 (0.899\u20130.906)), outperforming expert-developed models.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Discussion<\/jats:title>\n                  <jats:p>LLMs demonstrated limited capability in direct classification but excelled in generating effective neural network models with minimal human intervention. This suggests LLMs\u2019 potential in accelerating model development for clinical applications, particularly for detecting patient-ventilator asynchronies, though their clinical implementation requires further validation and consideration of ethical factors.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1136\/bmjhci-2024-101426","type":"journal-article","created":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T14:35:14Z","timestamp":1751034914000},"page":"e101426","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":6,"title":["Leveraging large language models for patient-ventilator asynchrony detection"],"prefix":"10.1136","volume":"32","clinical-trial-number":[{"clinical-trial-number":"nct02390024","registry":"10.18810\/clinical-trials-gov"},{"clinical-trial-number":"nct02714751","registry":"10.18810\/clinical-trials-gov"},{"clinical-trial-number":"nct03451461","registry":"10.18810\/clinical-trials-gov"},{"clinical-trial-number":"nct05363332","registry":"10.18810\/clinical-trials-gov"}],"author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8947-7814","authenticated-orcid":false,"given":"Francesc","family":"Su\u00f1ol","sequence":"first","affiliation":[{"name":"Institut d\u2019Investigaci\u00f3 i Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Candelaria","family":"de Haro","sequence":"additional","affiliation":[{"name":"Critical Care Department, Institut d\u2019Investigaci\u00f3 I Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"},{"name":"Centro Investigaci\u00f3n Biom\u00e9dica en Red de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2685-9377","authenticated-orcid":false,"given":"Ver\u00f3nica","family":"Santos-Pulp\u00f3n","sequence":"additional","affiliation":[{"name":"Institut d\u2019Investigaci\u00f3 i Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0267-0617","authenticated-orcid":false,"given":"Sol","family":"Fern\u00e1ndez-Gonzalo","sequence":"additional","affiliation":[{"name":"Institut d\u2019Investigaci\u00f3 i Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"},{"name":"Centro de Investigaci\u00f3n Biom\u00e9dica en Red de Salud Mental (CIBERSAM), Instituto de Salud Carlos III, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Llu\u00eds","family":"Blanch","sequence":"additional","affiliation":[{"name":"Critical Care Department, Institut d\u2019Investigaci\u00f3 I Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"},{"name":"Centro Investigaci\u00f3n Biom\u00e9dica en Red de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2359-3980","authenticated-orcid":false,"given":"Josefina","family":"L\u00f3pez-Aguilar","sequence":"additional","affiliation":[{"name":"Institut d\u2019Investigaci\u00f3 i Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"},{"name":"Centro Investigaci\u00f3n Biom\u00e9dica en Red de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0495-8422","authenticated-orcid":false,"given":"Leonardo","family":"Sarlabous","sequence":"additional","affiliation":[{"name":"Institut d\u2019Investigaci\u00f3 i Innovaci\u00f3 Parc Taul\u00ed (I3PT-CERCA), Sabadell, Spain"},{"name":"Centro Investigaci\u00f3n Biom\u00e9dica en Red de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2025,6,27]]},"reference":[{"key":"2025072913163103000_32.1.e101426.1","doi-asserted-by":"crossref","first-page":"633","DOI":"10.1007\/s00134-015-3692-6","article-title":"Asynchronies during mechanical ventilation are associated with mortality","volume":"41","author":"Blanch","year":"2015","journal-title":"Intensive Care Med"},{"key":"2025072913163103000_32.1.e101426.2","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1016\/j.ccc.2018.03.008","article-title":"Asynchrony Consequences and Management","volume":"34","author":"Pham","year":"2018","journal-title":"Crit Care Clin"},{"key":"2025072913163103000_32.1.e101426.3","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2024.40969","article-title":"Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial","volume":"7","author":"Goh","year":"2024","journal-title":"JAMA Netw Open"},{"key":"2025072913163103000_32.1.e101426.4","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2018.11100"},{"key":"2025072913163103000_32.1.e101426.5","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2023.35924","article-title":"Evaluating the Application of Large Language Models in Clinical Research Contexts","volume":"6","author":"Perlis","year":"2023","journal-title":"JAMA Netw Open"},{"key":"2025072913163103000_32.1.e101426.6","doi-asserted-by":"crossref","DOI":"10.1186\/s13054-024-04845-y","article-title":"Flow starvation during square-flow assisted ventilation detected by supervised deep learning techniques","volume":"28","author":"de Haro","year":"2024","journal-title":"Crit Care"},{"key":"2025072913163103000_32.1.e101426.7","doi-asserted-by":"crossref","first-page":"105673","DOI":"10.1016\/j.ijmedinf.2024.105673","article-title":"Evaluating the Effectiveness of advanced large language models in medical Knowledge: A Comparative study using Japanese national medical examination","volume":"193","author":"Liu","year":"2025","journal-title":"Int J Med Inform"}],"container-title":["BMJ Health &amp; Care Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjhci-2024-101426","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,29]],"date-time":"2025-07-29T20:16:41Z","timestamp":1753820201000},"score":1,"resource":{"primary":{"URL":"https:\/\/informatics.bmj.com\/lookup\/doi\/10.1136\/bmjhci-2024-101426"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6]]},"references-count":7,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,6,27]]},"published-print":{"date-parts":[[2025,6]]}},"alternative-id":["10.1136\/bmjhci-2024-101426"],"URL":"https:\/\/doi.org\/10.1136\/bmjhci-2024-101426","relation":{},"ISSN":["2632-1009"],"issn-type":[{"value":"2632-1009","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6]]}}}