Tuesday, November 28 | 9:40 a.m.-9:50 a.m. | T3-SSBR05-2 | Room S406B
In this talk, researchers will provide results from a multicenter study evaluating the use of AI software for assessing breast arterial calcifications.
Breast arterial calcium measurements are gaining popularity as a proxy for global atherosclerotic disease, and accurate quantitative models for automated assessment from mammograms could offer an adjunct screening tool for heart disease, according to presenter Chirag Parghi, MD, and colleagues.
After initial validation testing on 2D mammograms from 8,898 showed that the AI software an area under the curve (AUC) of 0.938 for breast arterial calcium detection, the researchers then prospectively analyzed its performance in nearly 16,000 asymptomatic women receiving screening mammography at 15 sites over a one-month period.
The algorithm spotted breast arterial calcium in 14.9% of the 15,785 women in the testing dataset, with a prevalence that increased with patient age.
“Our results suggest that AI can standardize [breast arterial calcification] detection at scale, potentially improving efficiency and reducing inter-observer variability,” Parghi et al wrote. “The age-specific prevalence of [breast arterial calcification] provided by our study can inform clinical decision-making and risk assessment with an expectation of patient volumes.”
What else did they find? Stop by this Tuesday morning talk for all of the details.
