AI misses 14% of invasive breast cancers on mammography, with luminal cancers and dense tissue posing significant challenges.
Key Details
- 1Study analyzed 1,097 breast cancers from 1,082 women using Lunit Insight MMG AI between 2014–2020.
- 2AI missed 154 cancers (14%), often in younger women with smaller (<2 cm), lower-grade tumors and dense breasts.
- 3Missed cancers were more likely luminal subtype (false-negative rate 17.2%), compared to HER2-enriched (9%) and triple-negative (14.5%).
- 4Major reasons for AI misses included dense breast tissue (n=56), nonmammary zone location (n=22), architectural distortion (n=12), and amorphous calcifications (n=5).
- 561.7% of AI-missed cancers were deemed 'actionable' on further review by radiologists.
Why It Matters
Understanding which cancer features are often overlooked by AI tools enables radiologists to better interpret AI results, reducing missed diagnoses. These insights highlight the continued importance of human expertise in mammography, especially for high-risk groups like women with dense breast tissue.
Source
AuntMinnie
Related News
•AuntMinnie
Study: Computer Vision Models Best LLMs in Chest CT Breast Abnormality Detection
Computer vision models (CVMs) surpass large language models (LLMs) in accurately labeling incidental breast abnormalities on chest CT scans.
•AuntMinnie
Deep Learning Models Rival Radiologists for Pancreatic Cancer Detection on CT
Deep-learning models achieved comparable or superior accuracy to experienced radiologists in detecting pancreatic cancer on CT scans, especially for small tumors.
•Radiology Business
Radiology AI Devices at Elevated Risk for FDA Recalls, Study Finds
Radiology AI devices are more likely to face FDA recalls, largely due to deviations from intended use and incomplete clinical data.