AI Shows Promise in Detecting Missed Breast Cancers

A recent study highlights the potential of artificial intelligence (AI) in identifying interval breast cancers missed during initial mammogram reviews. Researchers analyzed over 200 cases of interval breast cancers—cancers diagnosed between screenings—and found that AI detected 73 out of 224 cases, representing a third of missed diagnoses.
In a separate review of 1,000 screening mammograms, the AI algorithm demonstrated high accuracy, identifying 84% of true-positive cancers, 86% of true negatives, and 73% of false positives. The AI-detected cancers were significantly larger and more likely to be lymph-node positive compared to those not detected by AI.
The study, conducted by Manisha Bahl, MD, of Massachusetts General Hospital, and colleagues, focused on digital breast tomosynthesis (DBT), the standard screening method in the U.S. Prior research on AI in breast cancer detection had primarily relied on 2D mammography. The findings suggest AI could help reduce the rate of interval breast cancers, which are often aggressive and harder to treat.
However, the study’s authors emphasize the need for further prospective, randomized controlled trials to validate AI’s clinical utility. While promising, the technology’s effectiveness in real-world settings requires rigorous testing to ensure it complements, rather than replaces, radiologists’ expertise.
The research underscores AI’s potential to improve breast cancer screening, particularly in identifying aggressive cancers that may be missed during routine reviews. Future studies will be crucial to determine how best to integrate AI into clinical practice for optimal patient outcomes.

Published: 7/30/2025