Academic Journal
Automated Brief Hospital Course Summarization in Cardiac Surgery Using a Lightweight Large Language Model-Based Framework: Development and Evaluation Study on the Medical Information Mart for Intensive Care-IV.
| Τίτλος: | Automated Brief Hospital Course Summarization in Cardiac Surgery Using a Lightweight Large Language Model-Based Framework: Development and Evaluation Study on the Medical Information Mart for Intensive Care-IV. |
|---|---|
| Συγγραφείς: | Gao X; Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China.; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.; Information Center, First Hospital of Tsinghua University, Beijing, China., Wang Y; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Wang Z; Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China.; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Yuan J; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Huang S; Department of Cardiac Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, Beijing, China., Zhang XY; Institute of Automation of Chinese Academy of Sciences, Beijing, China.; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China., Sun Z; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Xing Y; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Liu Y; Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China.; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Wu X; Department of Information Center, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China., Hu Z; Department of Cardiac Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, Beijing, China., Zhao W; Statistical Information Center of the National Health Commission, Beijing, China. |
| Πηγή: | Journal of medical Internet research [J Med Internet Res] 2026 Sep 24; Vol. 28, pp. e90870. Date of Electronic Publication: 2026 Sep 24. |
| Τύπος έκδοσης: | Journal Article; Evaluation Study |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: JMIR Publications Country of Publication: Canada NLM ID: 100959882 Publication Model: Electronic Cited Medium: Internet ISSN: 1438-8871 (Electronic) Linking ISSN: 14388871 NLM ISO Abbreviation: J Med Internet Res Subsets: MEDLINE |
| Imprint Name(s): | Publication: <2011- > : Toronto : JMIR Publications Original Publication: [Pittsburgh, PA? : s.n., 1999- |
| Ιατρικοί όροι (MeSH): | Cardiac Surgical Procedures*/education , Natural Language Processing*, Large Language Models ; Humans |
| Περίληψη: | Background: Physician documentation requirements are a known contributor to clinician burnout, with the manual creation of brief hospital course (BHC) summaries being particularly time-consuming. Automating BHC summarization may mitigate this workload and reduce documentation errors. However, current natural language processing (NLP) methods are often limited to single-document inputs, and large language models (LLMs) face privacy and deployment challenges. Furthermore, existing methods often require manual information extraction and struggle to maintain temporal accuracy. Objective: We developed and evaluated LiteMedDoc, a lightweight, locally deployable LLM-based framework to automatically generate BHC summaries without model fine-tuning. Our objective was to determine whether clinically useful clinical summaries could be generated under strict privacy and resource constraints, making automated summarization feasible in real-world hospital environments. Methods: LiteMedDoc is a modular pipeline built upon an 8-billion-parameter open-source LLM (Llama 3.1). It features 3 specialized modules: a static dynamic information hierarchy module to condense multisource inputs and structure clinical events chronologically; a similar document retrieval-augmented generation module that retrieves contextually relevant prior case summaries; and a self-adaptive feedback optimization module used offline for prompt optimization. All processing was performed locally without any model fine-tuning. We evaluated the framework on a retrospective cohort of 4538 coronary artery bypass grafting (CABG) surgery cases from the Medical Information Mart for Intensive Care (MIMIC)-IV database. A held-out test set of 403 cases was used to generate BHC summaries. The model-generated summaries were compared to reference BHCs using 8 standard NLP metrics covering lexical overlap (BLEU-4 [Bilingual Evaluation Understudy-4 gram] and ROUGE [Recall-Oriented Understudy for Gisting Evaluation]), semantic similarity (BERTScore and METEOR [Metric for Evaluation of Translation With Explicit Ordering]), and clinical relevance (AlignScore [Alignment Score] and MEDCON [Medical Concept Overlap]). Additionally, 15 cardiac surgeons conducted a clinical evaluation of a sample of model-generated summaries, rating them on completeness, correctness, readability, conciseness, and global quality using a 5-point Likert scale. Results: Without any model training, LiteMedDoc achieved strong performance across individual automated metrics, nearly matching a fine-tuned model and exceeding a 70-billion-parameter model on all metrics. Additionally, in a within-database cross-domain evaluation on lobectomy cases, the framework maintained encouraging performance after prompt adaptation and outperformed both the base model and the CABG-fine-tuned model. Surgeons rated the AI-generated summaries above the prespecified acceptability threshold across all domains (mean scores ≥3.0), specifically praising their structure and conciseness. Conclusions: By integrating 3 specialized modules, the proposed framework offers a practical, locally deployable solution for clinician-in-the-loop BHC draft generation under privacy and resource constraints, and holds promise for improving documentation efficiency and enhancing information continuity during care transitions. (©Xiaoyuan Gao, Yang Wang, Zixing Wang, Jing Yuan, Shengkang Huang, Xu-Yao Zhang, Zhaohong Sun, Yun Xing, Yiyang Liu, Xintong Wu, Zhan Hu, Wei Zhao. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 24.09.2026.) |
| References: | J Biomed Inform. 2023 May;141:104358. (PMID: 37023846) Fed Pract. 2025 May;42(5):188-194. (PMID: 40831649) J Med Internet Res. 2024 Nov 19;26:e63445. (PMID: 39561354) JMIR Med Inform. 2024 Nov 18;12:e49724. (PMID: 39556827) Int J Nurs Stud. 2023 Jan;137:104385. (PMID: 36423423) JAMA Netw Open. 2025 Aug 01;8(8):e2526339. (PMID: 40802185) J Med Internet Res. 2025 May 15;27:e68998. (PMID: 40371947) Sci Data. 2023 Sep 6;10(1):586. (PMID: 37673893) JAMA Netw Open. 2025 Aug 01;8(8):e2526350. (PMID: 40802190) Res Social Adm Pharm. 2025 Aug;21(8):580-588. (PMID: 40133138) J Med Internet Res. 2024 Jul 24;26:e57721. (PMID: 39047282) J Gen Intern Med. 2026 Jan;41(2):323-329. (PMID: 40711633) J Med Internet Res. 2023 Jun 28;25:e48568. (PMID: 37379067) PLOS Digit Health. 2022 Dec 12;1(12):e0000158. (PMID: 36812600) Sci Rep. 2025 May 12;15(1):16466. (PMID: 40355506) JAMA Netw Open. 2024 Dec 2;7(12):e2448723. (PMID: 39625719) Patient Educ Couns. 2013 Aug;92(2):153-9. (PMID: 23746770) NEJM AI. 2024 Feb;1(2):. (PMID: 38343631) JAMA Netw Open. 2024 Dec 2;7(12):e2448729. (PMID: 39625728) Sci Data. 2023 Jan 3;10(1):1. (PMID: 36596836) J Med Internet Res. 2025 Oct 21;27:e80557. (PMID: 41118646) JMIR Med Inform. 2024 Apr 8;12:e55318. (PMID: 38587879) Int J Med Inform. 2024 Dec;192:105654. (PMID: 39437512) J Med Internet Res. 2020 Oct 23;22(10):e19810. (PMID: 33095174) NPJ Digit Med. 2024 Apr 3;7(1):84. (PMID: 38570554) Ann Intern Med. 2016 Dec 06;165(11):753-760. (PMID: 27595430) Lancet Reg Health West Pac. 2024 Sep 28;51:101211. (PMID: 39881943) JAMA. 2017 Jul 18;318(3):270-278. (PMID: 28719692) Ann Emerg Med. 2024 Aug;84(2):128-138. (PMID: 38483426) J Med Internet Res. 2024 Sep 12;26:e55247. (PMID: 39264712) JAMA. 2007 Feb 28;297(8):831-41. (PMID: 17327525) Semin Ophthalmol. 2023 Jul;38(5):503-507. (PMID: 37133418) Lancet Digit Health. 2023 Mar;5(3):e107-e108. (PMID: 36754724) JAMA Intern Med. 2025 Jul 01;185(7):818-825. (PMID: 40323616) Nature. 2023 Aug;620(7972):172-180. (PMID: 37438534) Mayo Clin Proc. 2016 Jul;91(7):836-48. (PMID: 27313121) J Am Med Inform Assoc. 2015 Sep;22(5):938-47. (PMID: 25882031) J Affect Disord. 2023 Mar 15;325:633-639. (PMID: 36640809) |
| Contributed Indexing: | Keywords: automated summarization; clinician burnout; large language models; lightweight; natural language processing; patient discharge summaries; privacy-preserving |
| Entry Date(s): | Date Created: 20260924 Date Completed: 20260924 Latest Revision: 20261011 |
| Update Code: | 20261011 |
| PubMed Central ID: | PMC13650456 |
| DOI: | 10.2196/90870 |
| PMID: | 42785733 |
| Βάση Δεδομένων: | MEDLINE |
καταχωρήστε σχόλιο πρώτοι!