Large Language Model–Based Virtual Patient Systems for History-Taking in Medical Education: Comprehensive Systematic Review
Große Sprachmodell-basierte virtuelle Patientensysteme für die Anamnese in der medizinischen Ausbildung: umfassende systematische Übersicht
JMIR Medical Informatics · 2026-01-02 · 14 · e79039
DOI: 10.2196/79039 ↗Auf einen Blick
39 Studien zu LLM-basierten virtuellen Patientensystemen zeigten meist gute Ergebnisse für die Anamneseausbildung. Multimodale Eingaben und bestimmte Techniken verbesserten Dialog und diagnostische Genauigkeit.
Fragestellung
Wie leistungsfähig sind LLM-basierte virtuelle Patientensysteme für die Anamnese in der medizinischen Ausbildung?
Methode
Systematische Übersicht nach PRISMA 2020; 9 Datenbanken von 1. Januar 2020 bis 18. August 2025 durchsucht; multidimensionale Qualitäts- und Bias-Bewertung.
Ergebnisse
39 eingeschlossene Studien. Systeme simulierten vor allem Innere Medizin und psychische Erkrankungen. Top-k accuracy: 0,45–0,98; Halluzinationsrate: 0,31 %–5 %; SUS ≥80. Kleine Stichproben, inkonsistente Metriken und wenige Kontrollen schränkten die Generalisierbarkeit ein.
Einschränkungen
Moderates Bias-Risiko, inkonsistente Metriken, kleine Kohorten und begrenzte Datensatztransparenz.
Schlussfolgerung der Autor:innen
Die eingeschlossenen Studien zeigten ein moderates Bias-Risiko, inkonsistente Metriken, kleine Kohorten und begrenzte Datensatztransparenz. LLM-basierte Systeme bilden mehrere Krankheitsbilder gut ab, aber Multimorbidität unzureichend.
Kernaussagen
- 39 Studien zu LLM-basierten virtuellen Patientensystemen
- Top-k accuracy 0,45–0,98; Halluzinationsrate 0,31 %–5 %
- Multimorbidität und seltene Erkrankungen waren selten vertreten
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
At a glance
Systematische Übersicht in JMIR Medical Informatics, published online on 2026-01-02. Written by Dongliang Li and 1 co-authors. Institution: Artificial Intelligence & Software Engineering, School of Computer Sciences, Universiti Sains Malaysia, Penang, Malaysia. Major MEDLINE topics: Education, Medical, Large Language Models, Medical History Taking. Cited 33 times so far (Crossref); 62 references. The full text is freely available in PubMed Central.
Abstract (original language)
Background
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.
Objective
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.
Methods
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–history-taking tasks were excluded. Multidimensional quality and bias assessments were conducted.
Results
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%–5%, System Usability Scale [SUS] ≥80). 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.
Conclusions
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.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Artificial Intelligence & Software Engineering, School of Computer Sciences, Universiti Sains Malaysia, Penang, Malaysia.
- Medical Informatics Department, College of Medicine and Health Sciences, Sultan Qaboos University, Al Seeb, Oman.
ORCID profiles: Dongliang Li, Syaheerah Lebai Lutfi
Cite
APA
Li, D., & Lebai Lutfi, S. (2026). Large Language Model–Based Virtual Patient Systems for History-Taking in Medical Education: Comprehensive Systematic Review. JMIR Medical Informatics, 14, e79039. https://doi.org/10.2196/79039
Vancouver
Li D, Lebai Lutfi S. Large Language Model–Based Virtual Patient Systems for History-Taking in Medical Education: Comprehensive Systematic Review. JMIR Medical Informatics. 2026;14:e79039. doi:10.2196/79039
BibTeX
@article{Li2026,
author = {Li, Dongliang and Lebai Lutfi, Syaheerah},
title = {Large Language Model–Based Virtual Patient Systems for History-Taking in Medical Education: Comprehensive Systematic Review},
journal = {JMIR Medical Informatics},
year = {2026},
volume = {14},
pages = {e79039},
doi = {10.2196/79039}
}
EUVIMED