An AI model using mammogram images predicts 5-year breast cancer risk more accurately than traditional breast density assessment.
Key Details
- 1The study is presented at the RSNA annual meeting in Chicago.
- 2Clairity Breast, the first FDA-authorized image-only AI breast cancer risk model, was trained on over 421,000 mammograms.
- 3AI model performance was evaluated on more than 245,000 screening mammograms from the US and Europe.
- 4Women in the AI high-risk group had a >4x higher incidence of cancer (5.9%) than average-risk (1.3%), while density-based separation was modest (3.2% vs. 2.7%).
- 5Findings support integrating image-only AI risk scores with traditional risk factors for more personalized screening strategies.
Why It Matters
This research advances precision screening in breast imaging, offering radiologists a more nuanced, data-driven tool for risk stratification. Early identification of high-risk patients via AI could lead to tailored screening protocols and improved outcomes.
Source
EurekAlert
Related News
•EurekAlert
AI 'Tissue Clocks' Estimate Organ-Specific Biological Age from Images
Researchers developed AI models that estimate the biological age of human organs from histological images, revealing organs age at different rates.
•EurekAlert
AI-Powered 3D Intraoperative Imaging Transforms Glioma Surgery
Researchers develop ULTRA, an AI-driven platform enabling 3D intraoperative tissue imaging for detecting glioma infiltration within 30 minutes.
•EurekAlert
PolyU Launches AI Virtual Patient Simulation for Personalized Cancer Care
Hong Kong Polytechnic University develops an AI-powered digital twin system that integrates multimodal patient data, including medical imaging, for personalized cancer treatment planning.