DANVILLE — Geisinger has begun using a new artificial intelligence (AI)-enabled breast cancer screening program to identify high-risk patients and refer them for appropriate care.

The new AI model evaluates all female patients with Geisinger primary care providers who are overdue for a mammogram on a monthly basis.

Of those patients, the model flags 50 patients at the highest risk for breast cancer based on the information in their electronic health record, such as family history or other health conditions. Geisinger’s Care Gaps team then contacts these patients to inform them about their risk and assist them in scheduling a mammogram.

The team also works with primary care providers to refer these patients to Geisinger’s high-risk breast clinic for ongoing monitoring. Research shows that early detection of breast cancer can lead to a 99% 5-year survival rate.

“This high-risk breast cancer model reflects Geisinger’s strengths in interdisciplinary care, quality and population health, and our investments in new technologies — in this case artificial intelligence — to identify women who are at high risk of breast cancer so they can be screened and treated,” said Keith Boell, D.O., Geisinger’s chief quality officer. “This work will literally save lives and is an example of how we at Geisinger continue to work to be the best.”

“Geisinger has developed a cutting-edge AI model for predicting breast cancer risk using readily available electronic health record data so that it is broadly applicable to a larger population. This program increases efficiency of outreach and improves screening rates for high-risk patients, while effectively integrating with current workflows,” said Biplab Bhattacharya, Ph.D., program director for data science at Geisinger, who led the technical development of the model. “This program is a critical step toward improving early detection and outcomes.”

For more information about comprehensive breast cancer care at Geisinger, visit geisinger.org/breastcare.