08/04/2026
There are more than 30 million women aged 25 to 40 in the U.S. and screening guidelines exclude nearly all of them, and insurance follows the guidelines. So they get told no twice, once by a doctor, once by a plan.
Here’s what that means for AI. Every risk model in breast imaging was trained on women already being screened, over 40, on mammography. The tools meant to personalize screening are being built without the population that needs them most. You cannot train a model on scans nobody ordered.
BeSound is building the datasets that don’t exist: young and dense breasts, imaged the same way, over time. Triage and risk stratification for the women current models cannot see.
Today we’re excited about GE HealthCare announcement officially launching the technology we use at BeSound. Dr. Rachel Brem on our team helped bring this technology through the FDA, and BeSound is among the first sites in the world to utilize it.
We’re excited to deepen our work with GE HealthCare from here, and continue expanding access to adequate screening technologies for younger women and women with dense breasts.
