Related Experiment Video
Updated: Jul 17, 2026
05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
Application Effectiveness Analysis of Artificial Intelligence-assisted Teaching in Standardized Training for Medical
Lu Zhang1, Qiuying Chen1, Bin Zhang1
1Department of Radiology, The First Affiliated Hospital of Jinan University, Guangzhou, PR China (L.Z., Q.C., B.Z., J.F., X.M., S.Z.).
Academic Radiology
|July 11, 2026
Summary
A combined AI and traditional teaching model significantly improved radiology residents' diagnostic skills for pulmonary nodules compared to traditional or AI-only methods. This integrated approach enhances learning outcomes and clinical reasoning.
Area of Science:
- Radiology Education
- Artificial Intelligence in Medicine
- Medical Diagnostics
Background:
- Pulmonary nodules are a common finding on chest CT scans, requiring accurate diagnosis.
- Traditional teaching methods for radiology residents may not fully optimize diagnostic competency.
- The integration of artificial intelligence (AI) presents new opportunities for medical education.
Purpose of the Study:
- To evaluate the efficacy of a combined AI-assisted and traditional teaching model for improving radiology residents' diagnostic competency in identifying pulmonary nodules on chest CT.
- To compare the combined model against traditional teaching (TT) and AI-assisted teaching (AI-AT) alone.
Main Methods:
- A randomized controlled trial involving 36 residents allocated to three groups: TT, AI-AT, and Combined Teaching (CT).
- Standardized theory instruction followed by group-specific practical training (TT: mentored film-reading; AI-AT: AI platform training; CT: AI training + expert review).
- Assessments included theoretical/practical exams, diagnostic process metrics, and a 1-month follow-up survey on confidence and AI perceptions.
Main Results:
- The CT group achieved significantly higher post-test theoretical and practical scores than both TT and AI-AT groups (P < 0.001).
- CT group demonstrated more deliberate analysis (longer reading time, P = 0.01) and superior completeness in describing imaging features (P < 0.001).
- CT group maintained higher diagnostic confidence (P = 0.002) and showed a more balanced perception of AI as a collaborative tool at 1-month follow-up.
Conclusions:
- The combined AI-assisted and traditional teaching model is significantly more effective than either method alone for enhancing diagnostic performance and clinical reasoning in radiology residents for pulmonary nodules.
- This integrated approach fosters deeper cognitive engagement and promotes sustainable learning outcomes.
- The study provides a framework for competency-based medical education in the era of AI.