Artificial intelligence model boosts lung nodule diagnostic accuracy in clinical trial

A new artificial intelligence model called DeepFAN has been validated in China's first multi-center clinical trial for pulmonary nodule assessment, demonstrating high diagnostic accuracy in distinguishing malignant from benign lung nodules. Published in Nature Cancer, the study showed that when used as a clinical decision-support tool, DeepFAN substantially improved junior radiologists' diagnostic accuracy, confidence, and consistency. The findings are significant because early and accurate identification of malignant lung nodules is critical for timely lung cancer treatment, and AI-assisted diagnosis could help bridge the experience gap between junior and senior radiologists. This represents one of the most rigorously validated AI tools for lung nodule classification to date, with potential implications for large-scale lung cancer screening programs.