Application Effectiveness Analysis of Artificial Intelligence-assisted Teaching in Standardized Training for Medical Imaging Residents.

In 30 seconds

This randomized controlled trial assessed the effectiveness of an AI-assisted teaching model combined with traditional methods for enhancing diagnostic competency in radiology residents focusing on pulmonary nodules. The combined teaching group significantly outperformed traditional and AI-only groups in both theoretical (91.8 ± 3.4 vs. 82.6 ± 3.2, P < 0.001) and practical scores (93.5 ± 2.7 vs. 81.9 ± 3.0, P < 0.001).

Key findings

• The combined teaching group scored 91.8 ± 3.4 in theoretical exams, significantly higher than the traditional group's 82.6 ± 3.2.
• In practical exams, the combined group achieved 93.5 ± 2.7, outperforming the AI-assisted group's 86.8 ± 3.0.
• The combined group engaged in more deliberate analysis, indicated by longer reading times (P = 0.01).
• At 1-month follow-up, the combined group maintained higher diagnostic confidence (P = 0.002).

Why it matters

The findings highlight the potential of integrating AI into medical education, suggesting that such combined approaches can enhance diagnostic skills and confidence among residents, which is crucial for improving patient care in radiology.