Teacher-student-machine interaction autonomous learning: a structured LLM-integrated framework for developing independent clinical reasoning in residency training.
8 affiliations
- 1Department of Surgery and Field Surgery, Changzheng Hospital, Navy Military Medical University, Shanghai, 200003, China.
- 2Department of Orthorpedic Oncology, Changzheng Hospital, Navy Military Medical University, 415th Fengyang Road, Shanghai, 200003, China.
- 3Department of Emergency Medicine, Naval Medical Center, 338th West Huaihai Road, Shanghai, 200050, China.
- 4Department of Cardiology, the 961st Hospital of Joint Logistic Support Force of PLA, 189th Jianhua Middle Road, Qiqihar City, 161000, Heilongjiang Province, China.
- 5Department of Cardiology, Shanghai Changzheng Hospital, Naval Medical University, No. 415, Fengyang Road, Huangpu District, Shanghai, China.
- 6Department of Orthorpedic Oncology, Changzheng Hospital, Navy Military Medical University, 415th Fengyang Road, Shanghai, 200003, China. [email protected].
- 7Medical Administration Division, Chinese PLA General Hospital, 28th Fuxing Road, Beijing, 100853, China. [email protected].
- 8Department of Orthorpedic Oncology, Changzheng Hospital, Navy Military Medical University, 415th Fengyang Road, Shanghai, 200003, China. [email protected].
Abstract
Background
This study evaluated whether a structured pedagogical framework integrating large language models (LLMs) into residency training could develop clinical reasoning competencies that transfer to independent performance.Methods
In this prospective randomized controlled trial, residents were assigned to Teacher-Student-Machine Interaction Autonomous Learning (TSM-AL) group or traditional teaching group. The TSM-AL framework positioned LLMs as supervised cognitive auxiliaries within a five-step case analysis process featuring controlled information release, predefined reasoning tasks, and structured utilization guidelines. Six formative assessments were conducted over 16 weeks, followed by summative examinations under both LLM-assisted and unassisted conditions. Outcomes included clinical reasoning scores, entrustable professional activity (EPA) levels, global faculty ratings, expert-aligned diagnostic accuracy, and standardized patient examination performance.Results
Significant between-group differences in diagnostic pathway completeness emerged at week 14 (3.53 ± 1.36 vs. 2.68 ± 1.49, P = 0.018) and week 16 (3.35 ± 1.43 vs. 2.37 ± 1.51, P = 0.009). In the unassisted final examination, the TSM-AL group demonstrated significantly higher clinical reasoning scores (7.32 ± 2.30 vs. 5.12 ± 1.66, P < 0.001). EPA distributions differed significantly (P = 0.034), with fewer TSM-AL residents requiring complete supervision (11.76% vs. 29.41%). Global faculty ratings also differed significantly (P = 0.019), with 23.53% of TSM-AL residents achieving Level 5 compared to none in the traditional group. Expert alignment for core problem identification (67.65% vs. 44.12%, P = 0.042) and differential diagnosis matching (73.53% vs. 50.00%, P = 0.046) favored the TSM-AL group. Standardized patient examination scores were significantly higher in the TSM-AL group (80.88 ± 3.82 vs. 78.71 ± 3.90, P = 0.025). Notably, TSM-AL residents performed better without LLM assistance than with assistance.Conclusions
The TSM-AL framework significantly enhanced clinical reasoning competencies that transferred to independent performance, demonstrating that structured LLM integration develops autonomous reasoning rather than fostering technological dependency.
Related
- A Reproducible Decomposed Large Language Model Pipeline for Structured Abstraction of Multimodal Neuroimaging Reports: Development and Independent Validation in 1,669 Emergency Department Encounters for Dizziness and Vertigo
- Integrated malaria control and drug resistance in an age- and sex-structured model with seasonal malaria chemoprevention, mass drug administration, and routine case management
- Adolescent idiopathic scoliosis and comorbid anxiety and depression: advances in psychophysiological interaction mechanisms and integrated interventions
- Integrated pharmacological and structural profiling reveals CB1 residue interaction patterns associated with synthetic cannabinoid receptor agonist efficacy
- Extracorporeal Membrane Oxygenation (ECMO) Prediction of Mortality and Integrated Complication with Infection Risk Assessment by Machine Learning: Structured Narrative Review.
BibTeX
@article{wang2026teacherstudentmachine,
title = {{Teacher-student-machine interaction autonomous learning: a structured LLM-integrated framework for developing independent clinical reasoning in residency training.}},
author = {Pengru Wang and He Li and Dingyuan Tu and Huilong Qu and Mengli Chang and Qiying Zhang and Gan Xu and Bo Li and Yingtian Wang and Wei Xu},
journal = {BMC medical education},
year = {2026},
month = aug,
doi = {10.1186/s12909-026-09650-5},
url = {https://europepmc.org/article/MED/42968068},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Teacher-student-machine interaction autonomous learning: a structured LLM-integrated framework for developing independent clinical reasoning in residency training.
AU - Pengru Wang
AU - He Li
AU - Dingyuan Tu
AU - Huilong Qu
AU - Mengli Chang
AU - Qiying Zhang
AU - Gan Xu
AU - Bo Li
AU - Yingtian Wang
AU - Wei Xu
PY - 2026
DA - 2026/08/04
JO - BMC medical education
DO - 10.1186/s12909-026-09650-5
UR - https://europepmc.org/article/MED/42968068
AB - Background This study evaluated whether a structured pedagogical framework integrating large language models (LLMs) into residency training could develop clinical reasoning competencies that transfer to independent performance.Methods In this prospective randomized controlled trial, residents were assigned to Teacher-Student-Machine Interaction Autonomous Learning (TSM-AL) group or traditional teaching group. The TSM-AL framework positioned LLMs as supervised cognitive auxiliaries within a five-step case analysis process featuring controlled information release, predefined reasoning tasks, and structured utilization guidelines. Six formative assessments were conducted over 16 weeks, followed by summative examinations under both LLM-assisted and unassisted conditions. Outcomes included clinical reasoning scores, entrustable professional activity (EPA) levels, global faculty ratings, expert-aligned diagnostic accuracy, and standardized patient examination performance.Results Significant between-group differences in diagnostic pathway completeness emerged at week 14 (3.53 ± 1.36 vs. 2.68 ± 1.49, P = 0.018) and week 16 (3.35 ± 1.43 vs. 2.37 ± 1.51, P = 0.009). In the unassisted final examination, the TSM-AL group demonstrated significantly higher clinical reasoning scores (7.32 ± 2.30 vs. 5.12 ± 1.66, P < 0.001). EPA distributions differed significantly (P = 0.034), with fewer TSM-AL residents requiring complete supervision (11.76% vs. 29.41%). Global faculty ratings also differed significantly (P = 0.019), with 23.53% of TSM-AL residents achieving Level 5 compared to none in the traditional group. Expert alignment for core problem identification (67.65% vs. 44.12%, P = 0.042) and differential diagnosis matching (73.53% vs. 50.00%, P = 0.046) favored the TSM-AL group. Standardized patient examination scores were significantly higher in the TSM-AL group (80.88 ± 3.82 vs. 78.71 ± 3.90, P = 0.025). Notably, TSM-AL residents performed better without LLM assistance than with assistance.Conclusions The TSM-AL framework significantly enhanced clinical reasoning competencies that transferred to independent performance, demonstrating that structured LLM integration develops autonomous reasoning rather than fostering technological dependency.
SN - 1472-6920
ER -