A new mammography-based AI model predicts 10-year breast cancer risk more accurately than established clinical and AI models.
Key Details
- 1Developed by Karolinska Institute, the AI model uses digital mammogram images to assess long-term breast cancer risk.
- 2Tested in 8,696 women (KARMA and Olmsted cohorts) with 1,633 incident breast cancers.
- 3Model achieved AUC of 0.72 for invasive cancers in both KARMA and Olmsted cohorts, and 0.70 in the EMBED cohort.
- 4Outperformed the Mirai model (AUC 0.64 in Olmsted/KARMA; 0.63 in EMBED) and clinical risk models in all cohorts.
- 5In the KARMA cohort, the top 10% risk group identified by the AI model captured 33% of cancers, higher than other models.
- 6The model is being prepared for further validation in broader populations and potential integration into risk-stratified prevention programs.
Why It Matters
Improved long-term breast cancer risk prediction could enable more precise, personalized screening and prevention strategies. This model could shift mammography from a purely diagnostic tool to a critical predictor of future disease, enhancing patient care and outcomes.
Source
AuntMinnie
Related News
•AuntMinnie
Radiologists Struggle to Spot AI-Generated Radiology Images
Radiologists correctly identify AI-generated images 75% of the time, with CT and MRI images being particularly challenging to spot.
•AuntMinnie
Radiology Leads FDA AI Device Approvals Over Three Decades
Radiology accounts for 76% of all FDA-cleared AI/ML-enabled medical devices as of the end of 2025.
•Radiology Business
Automation Bias: How AI Can Compromise Radiologist Accuracy
AI decision support can induce automation bias, leading radiologists to accept incorrect interpretations and potentially reduce their diagnostic accuracy.