An AI model accurately predicted breast cancer risk in a racially and ethnically diverse screening population using mammogram data.
Key Details
- 1Study included 206,929 women (average age 56.1) from a province-wide screening program starting at age 40.
- 2AI model used both current and up to four years of prior mammograms for risk scoring.
- 3Model achieved AUROC of 0.78 overall, 0.77 in East Asian, 0.77 in Indigenous, and 0.75 in South Asian women.
- 44,168 breast cancer cases confirmed over an average 5.3-year follow-up.
- 5Model showed consistent performance across age groups and ethnicities.
Why It Matters
Demonstrating robust, consistent AI performance across diverse populations supports its broader clinical adoption. Such risk models may enable more personalized and equitable breast cancer screening and resource allocation.
Source
AuntMinnie
Related News
•Radiology Business
Experts Urge Development of Generalist Radiology AI to Cut Costs and Improve Care
Leading scientists advocate for broader, generalist radiology AI models to overcome limitations of narrow, single-task solutions.
•AuntMinnie
General LLMs Show Promise in Detecting Critical Findings in Radiology Reports
Stanford and Mayo Clinic Arizona researchers demonstrated that LLMs like GPT-4 can categorize critical findings in radiology reports using few-shot prompting.
•AuntMinnie
Experts Outline Framework and Benefits for Generalist Radiology AI
Researchers propose key features and benefits for implementing generalist radiology AI (GRAI) frameworks over narrow AI tools.