Keywords
Standardized patient
Surgical education
Clinical clerkship
Large language model
Submitted : 2026-01-06
Accepted : 2026-01-21
Published : 2026-02-05
Abstract
Objective: To address the insufficient integration of theory and practice in surgical clinical clerkship teaching, the limited availability of authentic clinical cases, and the limitations of traditional standardized patients (SPs) in terms of consistency and organizational cost, this study aimed to develop an artificial intelligence-based standardized patient (AI-SP) teaching framework grounded in structured clinical case data. Lumbar disc herniation was used as a representative condition for application and exploratory implementation. Methods: An exploratory teaching application design was adopted. Based on the learning objectives of the lumbar disc herniation chapter in standard surgery textbooks, a layered AI-SP system was developed, comprising a case data layer, a clinical rules and teaching logic layer, and an AI interaction layer. Clinical case data were derived from a single-center spine surgery practice and were de-identified and structured into teaching-oriented data units. Diagnostic and therapeutic principles from textbooks, together with expert consensus and clinical guidelines, were abstracted into rule constraints and scenario evolution logic. The interaction layer employed a large language model to support multi-turn dialogue, with standardization ensured through role restriction, rule-based control, and consistency validation. The system was embedded into three stages of clerkship teaching—pre-clerkship preparation, in-clerkship guidance, and post-clerkship consolidation—and representative interaction workflows were developed. Results: A layered AI-SP architecture and information flow model tailored for surgical clinical clerkship teaching was established. Under predefined rule constraints, the system was able to generate stable patient narratives consistent with textbook content and clinical reasoning, enabling reproducible and controllable standardized teaching scenarios. An exploratory application suggested that the AI-SP facilitated the formation of a more structured disease understanding before students entered real clinical settings, improved the consistency and focus of clerkship discussions, and supported repeated practice under conditions of limited clinical resources. Quantitative evaluation of learning outcomes was not conducted in this study. Conclusion: The AI-SP framework, developed using artificial intelligence and structured clinical case data, may serve as an auxiliary tool for surgical clinical clerkship teaching by providing a controlled and standardized interactive training environment without replacing students’ clinical judgment. Future studies should incorporate multicenter case data and employ controlled designs with quantitative outcome measures to systematically evaluate the educational effectiveness.
References
Rencic J, Trowbridge RL, Fagan M, et al., 2017, Clinical Reasoning Education at US Medical Schools: Results from a National Survey of Internal Medicine Clerkship Directors. J Gen Intern Med, 32(11): 1242–1246.
Flanagan OL, Cummings KM, 2023, Standardized Patients in Medical Education: A Review of the Literature. Cureus, 15(7): e42027.
Nestel D, Tabak D, Tierney T, et al., 2011, Key Challenges in Simulated Patient Programs: An International Comparative Case Study. BMC Med Educ, 11: 69.
Bosse HM, Nickel M, Huwendiek S, et al., 2015, Cost-Effectiveness of Peer Role Play and Standardized Patients in Undergraduate Communication Training. BMC Med Educ, 15: 183.
Kononowicz AA, Woodham LA, Edelbring S, et al., 2019, Virtual Patient Simulations in Health Professions Education: Systematic Review and Meta-Analysis by the Digital Health Education Collaboration. J Med Internet Res, 21(7): e14676.
Bateman J, Allen M, Samani D, et al., 2013, Virtual Patient Design: Exploring What Works and Why. A Grounded Theory Study. Med Educ, 47(6): 595–606.
Clever SL, Dudas RA, Solomon BS, et al., 2011, Medical Student and Faculty Perceptions of Volunteer Outpatients Versus Simulated Patients in Communication Skills Training. Acad Med, 86(11): 1437–1442.
Schmidt HG, Mamede S, 2015, How to Improve the Teaching of Clinical Reasoning: A Narrative Review and a Proposal. Med Educ, 49(10): 961–973.
Huwendiek S, Reichert F, Bosse HM, et al., 2009, Design Principles for Virtual Patients: A Focus Group Study Among Students. Med Educ, 43(6): 580–588.
Ahmed T, Aziz M, Mohammed N, 2020, De-Identification of Electronic Health Record Using Neural Network. Sci Rep, 10(1): 18600.
Boscardin CK, Gin B, Golde PB, et al., 2024, ChatGPT and Generative Artificial Intelligence for Medical Education: Potential Impact and Opportunity. Acad Med, 99(1): 22–27.
Roustan D, Bastardot F, 2025, The Clinicians’ Guide to Large Language Models: A General Perspective With a Focus on Hallucinations. Interact J Med Res, 14: e59823.
Yang Y, Jin Q, Huang F, et al., 2024, Adversarial Attacks on Large Language Models in Medicine. arXiv. https://doi.org/10.48550/arXiv.2406.12259
Graber ML, Franklin N, Gordon R, 2005, Diagnostic Error in Internal Medicine. Arch Intern Med, 165(13): 1493–1499.
Kreiner DS, Hwang SW, Easa JE, et al., 2014, An Evidence-Based Clinical Guideline for the Diagnosis and Treatment of Lumbar Disc Herniation with Radiculopathy. Spine J, 14(1): 180–191.
Briggs AM, Sumi Y, Banerjee A, 2025, The World Health Organization Guideline for Non-Surgical Management of Chronic Primary Low Back Pain in Adults: Implications for Equitable Care and Strengthening Health Systems Globally. Glob Health Res Policy, 10(1): 26.
Kung TH, Cheatham M, Medenilla A, et al., 2023, Performance of ChatGPT on USMLE: Potential for AI-Assisted Medical Education Using Large Language Models. PLOS Digit Health, 2(2): e0000198.
Berman NB, Durning SJ, Fischer MR, et al., 2016, The Role for Virtual Patients in the Future of Medical Education. Acad Med, 91(9): 1217–1222.
Wingelaar TT, Wagter JM, Arnold AE, 2012, Students’ Educational Needs for Clinical Reasoning in First Clerkships. Perspect Med Educ, 1(2): 56–66.
Norman G, 2009, Dual Processing and Diagnostic Errors. Adv Health Sci Educ Theory Pract, 14 Suppl 1: 37–49.
Adnan AI, 2022, Effectiveness of Communication Skills Training in Medical Students Using Simulated Patients or Volunteer Outpatients. Cureus, 14(7): e26717.
Cook DA, Hatala R, Brydges R, et al., 2011, Technology-Enhanced Simulation for Health Professions Education: A Systematic Review and Meta-Analysis. JAMA, 306(9): 978–988.