{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T14:57:39Z","timestamp":1792249059411,"version":"4.4.0"},"reference-count":62,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs), such as GPT-3.5 and GPT-4 (OpenAI), have been transforming virtual patient systems in medical education by providing scalable and cost-effective alternatives to standardized patients. However, systematic evaluations of their performance, particularly for multimorbidity scenarios involving multiple coexisting diseases, are still limited.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This systematic review aimed to evaluate LLM-based virtual patient systems for medical history-taking, addressing four research questions: (1) simulated patient types and disease scope, (2) performance-enhancing techniques, (3) experimental designs and evaluation metrics, and (4) dataset characteristics and availability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020, 9 databases were searched (January 1, 2020, to August 18, 2025). Nontransformer LLMs and non\u2013history-taking tasks were excluded. Multidimensional quality and bias assessments were conducted.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>A total of 39 studies were included, screened by one computer science researcher under supervision. LLM-based virtual patient systems mainly simulated internal medicine and mental health disorders, with many addressing distinct single disease types but few covering multimorbidity or rare conditions. Techniques like role-based prompts, few-shot learning, multiagent frameworks, knowledge graph (KG) integration (top-k accuracy 16.02%), and fine-tuning enhanced dialogue and diagnostic accuracy. Multimodal inputs (eg, speech and imaging) improved immersion and realism. Evaluations, typically involving 10-50 students and 3-10 experts, demonstrated strong performance (top-k accuracy: 0.45-0.98, hallucination rate: 0.31%\u20135%, System Usability Scale [SUS] \u226580). However, small samples, inconsistent metrics, and limited controls restricted generalizability. Common datasets such as MIMIC-III (Medical Information Mart for Intensive Care-III) exhibited intensive care unit (ICU) bias and lacked diversity, affecting reproducibility and external validity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Included studies showed moderate risk of bias, inconsistent metrics, small cohorts, and limited dataset transparency. LLM-based virtual patient systems excel in simulating multiple disease types but lack multimorbidity patient representation. KGs improve top-k accuracy and support structured disease representation and reasoning. Future research should prioritize hybrid KG-chain-of-thought architectures integrated with open-source KGs (eg, UMLS [Unified Medical Language System] and SNOMED-CT [Systematized Nomenclature of Medicine - Clinical Terms]), parameter-efficient fine-tuning, dialogue compression, multimodal LLMs, standardized metrics, larger cohorts, and open-access multimodal datasets to further enhance realism, diagnostic accuracy, fairness, and educational utility.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/79039","type":"journal-article","created":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T17:05:06Z","timestamp":1761239106000},"page":"e79039","source":"Crossref","is-referenced-by-count":30,"title":["Large Language Model\u2013Based Virtual Patient Systems for History-Taking in Medical Education: Comprehensive Systematic Review"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-8675-3702","authenticated-orcid":false,"given":"Dongliang","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7349-0061","authenticated-orcid":false,"given":"Syaheerah","family":"Lebai Lutfi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,1,2]]},"reference":[{"key":"ref1","unstructured":"BrownTBMannBRyderNSubbiahMKaplanJDhariwalPNeelakantanAShyamPSastryGAskellALanguage models are few-shot learners2020NIPS'20: Proceedings of the 34th International Conference on Neural Information Processing Systems2020 December 06Red Hook, NY, United States18771901"},{"key":"ref2","author":"Achiam, J","year":"2024","journal-title":"ArXiv. Preprint posted online on March 4,"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1097\/SIH.0000000000000526"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1080\/0142159X.2024.2376879"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/tmm.2022.3205456"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1016\/j.rcsop.2024.100419"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.3390\/biomedinformatics4010047"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.2196\/59213"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1186\/s13104-025-07157-8"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2409.18924"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1111\/medu.15402"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare12222241"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1097\/MD.0000000000039250"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1186\/s13643-021-01626-4"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.3390\/app15094862"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.3390\/s25144305"},{"key":"ref17","author":"Liu, Z","year":"2025","journal-title":"ArXiv. Preprint posted online on March 11,"},{"key":"ref18","author":"Du, Z","year":"2025","journal-title":"ArXiv. Preprint posted online on June 7,"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2505.04653"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.2196\/58753"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-08866-7"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.2196\/53961"},{"key":"ref23","author":"Li, Y","year":"2024","journal-title":"ArXiv. Preprint posted online on April 25,"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2025.05.025"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2502.19175"},{"key":"ref26","unstructured":"KumarRGattaniDSinghKEnhancing medical history collection using llms2024ACSW '24: Proceedings of the 2024 Australasian Computer Science Week2024 May 13NSW, Sydney, Australia140143"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.3233\/SHTI251254"},{"key":"ref28","doi-asserted-by":"crossref","unstructured":"TakataTYamadaRRen\u00e9ADevelopment of a virtual patient model for Kampo medical interview: new approach for enhancing empathy and understanding of Kampo medicine pathological concepts20242024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems (SCIS&ISIS)2024 November 01Himeji, Japan15","DOI":"10.1109\/SCISISIS61014.2024.10759962"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1016\/j.teln.2024.02.005"},{"key":"ref30","author":"Rashidian, S","year":"2025","journal-title":"ArXiv. Preprint posted online on June 4,"},{"key":"ref31","first-page":"177","volume":"22","author":"Kenny, PG","year":"2024","journal-title":"Annual Review of Cybertherapy And Telemedicine"},{"key":"ref32","author":"Wang, R","year":"2024","journal-title":"ArXiv. Preprint posted online on October 3,"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.13614"},{"key":"ref34","author":"Staples, P","year":"2025","journal-title":"PsyArXiv"},{"issue":"4","key":"ref35","first-page":"413","volume":"35","author":"Ryu, JW","year":"2024","journal-title":"Journal of Oriental Neuropsychiatry"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-025-06756-0"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2503.22250"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.jhsa.2024.06.002"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-024-06399-7"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.2196\/63312"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3610978.3640592"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01841-6"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.2139\/ssrn.4890649"},{"key":"ref44","unstructured":"JohriSJeongJTranBAet alCRAFT-MD: A conversational evaluation framework for comprehensive assessment of clinical LLMs2024AAAI 2024 Spring Symposium on Clinical Foundation ModelsMarch 25-27, 2024Stanford University, Stanford, California"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1056\/aioa2300138"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-92970-0_17"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2024\/1267"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2506.11221"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-36272-9_71"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.2196\/68486"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/s40670-025-02473-x"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35"},{"key":"ref53","author":"Fansi, TA","year":"2022","journal-title":"ArXiv. Preprint posted online on October 13,"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2405.02957"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1145\/3637528.3671576"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2003.03875"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocv080"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.5860\/choice.189890"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1145\/3706598.3714014"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkh061"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1197\/jamia.M1767"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00638"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e79039","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:11:54Z","timestamp":1782843114000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e79039"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,2]]},"references-count":62,"URL":"https:\/\/doi.org\/10.2196\/79039","relation":{"has-preprint":[{"id-type":"doi","id":"10.2196\/preprints.79039","asserted-by":"object"}]},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,2]]}}}