Artificial intelligence-assisted ultrasound screening for breast cancer in China: a prospective, clustered, controlled, population-based study.

Comprehensive Summary

This study investigates the effectiveness of using artificial intelligence (AI)-assisted ultrasound screening for breast cancer detection among women in China. Shen et al. conducted a prospective study within China’s “Two Cancers” screening program, comparing AI-assisted ultrasound screening in one district with routine ultrasound screening in another. The study analyzed data from over 21,790 women, 35 to 69 years old with no prior history of breast cancer, to evaluate the differences in screening sensitivity and cancer detection rates. The screening methods consisted of a clinical examination, initial breast ultrasound screening, MAM rescreen, histopathological examination, and follow-up. Results showed that the AI-assisted screening model significantly detected more breast cancers compared to routine screening (21 vs. 9 cases, P = 0.001), and had a higher sensitivity (75% vs. 42.8%). The AI group was also more effective in detecting early-stage cancers (95.2% vs. 88.9%). However, there was no significant difference in interval cancer detection between the two methods. Overall, this study demonstrates how AI-assisted ultrasounds can improve screening sensitivity and early cancer detection rates when compared to traditional methods, providing an effective alternative for populations in which mammographies are less efficient. Furthermore, the use of AI showed to enhance the identification of "true-positive" cases, increasing the accuracy and reliability of breast cancer screening.

Outcomes and Implications

This research demonstrates how AI-assisted ultrasound can enhance breast cancer detection accuracy, particularly for women with dense breast tissue, where mammography is less effective. By improving sensitivity and early-stage detection, AI-based ultrasound offers a cost-effective alternative for large-scale screening programs in China and other similar populations. Clinically, these findings suggest that AI-enhanced imaging can complement or replace conventional screening methods in specific demographic groups, ultimately improving diagnostic precision and reducing missed cancers.