AI can flag breast cancer years earlier, but access still depends on your zip code
A George Washington University radiologist says AI reads can spot tumors years sooner, yet only a fraction of U.S. practices offer them, often for a fee.
In brief
- Dr. Rachel Brem says research shows AI-assisted mammograms can find cancer up to five years earlier.
- George Washington University has run AI on every mammogram for over four years at no extra cost.
- Brem advises yearly screening from 40, or from 35 with a family history of breast cancer.
- Half of women have dense breasts and may need ultrasound, MRI or contrast-enhanced mammograms.
Artificial intelligence layered on top of a standard mammogram can flag warning signs a radiologist reading the image alone might miss, according to Dr. Rachel Brem, a board-certified diagnostic radiologist who directs Breast Imaging and Intervention at George Washington University. She also argues that many women should begin screening sooner than national guidelines suggest, a view she laid out in an interview reported by CBS News on October 14, 2026.
A radiologist shaped by her mother’s diagnosis
Brem’s career began with a family crisis. She was 12 when her mother learned she had breast cancer, and she resolved then to cure the disease. Doctors had given her mother about six months, but because the tumor was caught early, she lived far beyond that estimate. Brem later founded the Brem Foundation, which over two decades has pushed for innovative screening, education and awareness, and fair access to care.
Today she works with ScreenPoint Medical on Transpara, a software tool that analyzes mammogram images. According to the CBS report, such software can also estimate a woman’s future risk of breast cancer even when her current scan shows nothing. Brem adds that if women started AI-assisted readings at 35, the system would keep refining its detection with every image it processed. That age sits below the starting point several major medical organizations recommend, and no medical body has endorsed it for AI screening.
“AI is going to change everything in medicine, and certainly breast cancer detection,”
Dr. Rachel Brem, George Washington University
Who can get an AI reading today
At George Washington University, every mammogram has gone through AI analysis for more than four years, and patients pay nothing extra for it. Elsewhere the picture is thinner.
Where the technology is offered, it often carries an added charge, which can put it out of reach. Brem also points out that women with fewer resources have less access to mammograms in the first place. She hopes AI could narrow those gaps, though that outcome has not been shown.
Another tool is still in development. Brem is helping build a breath test that would pick up a gas released by diseased breast tissue. It is not yet available; the idea is to offer a noninvasive way to decide who needs further imaging.
Screening start ages compared
| Source | Starting age | Frequency |
|---|---|---|
| American College of Physicians (2026) | Over 50 | Every other year |
| Dr. Rachel Brem, family history | 35 | Every year |
Brem’s position rests on her own research and differs from both official recommendations. The CBS report also notes that cancer rates are rising among younger women, and Brem says 99% of women whose breast cancer is found early survive and thrive.
What to raise with your doctor
- Ask about extra imaging. Brem urges women to insist that supplemental exams be ordered, since some doctors do not routinely include them.
- Discuss family history. It is the basis for her advice to begin at 35 instead of 40.
- Ask whether AI is used. Availability and price vary widely between practices.
- Address fear directly. Anxiety is a common reason women put off or skip screening.
This advice reflects one radiologist’s view; women whose doctors follow the ACP or task force guidance should weigh their personal risk in that conversation. For tools beyond the clinic, see our look at FemTech for women’s health.
On anxiety, Brem is blunt: “We don’t get rid of the disease by not knowing we have it.”



