New FDA-cleared AI tool could help predict if breast cancer will return
Originally published: October 15, 2026
By Ava Tsapatsaris and Dr. Tara Howard
For many women diagnosed with breast cancer, one of the biggest worries is whether the cancer will come back after treatment. A newly FDA-cleared artificial intelligence tool may help doctors answer that question earlier, and unlike many current tests, it can provide an estimate before surgery.
▶ Watch the video: "AI tool could help predict breast cancer recurrence" (ABC News)
What is TumorSight Risk?
TumorSight Risk, developed by SimBioSys and powered by NVIDIA technology, estimates the chance that a patient's breast cancer will return. It uses NVIDIA's graphics processing units (GPUs) to analyze complex MRI data together with large amounts of medical history that would be difficult for people to synthesize on their own. The company says it is working with Medicare, Medicaid and private insurers to get the test covered.
What did researchers find?
A new study found the tool may help predict recurrence risk in women with early-stage, hormone receptor-positive, HER2-negative breast cancer, a type that makes up about 74% of U.S. breast cancer diagnoses, according to the National Cancer Institute.
Researchers analyzed previously collected data from 2,129 women.
95.5% of women classified as low risk remained cancer-free after five years, compared with 89.6% of those classified as high risk.
Women classified as high risk had more than three times the risk of recurrence compared with those classified as low risk.
Each 10-point increase in the AI-generated score was associated with a 22% increase in relative recurrence risk.
How does the tool work?
TumorSight Risk doesn't require any extra scans. It uses the breast MRI a patient already had during her initial diagnosis. The AI examines the tumor and surrounding tissue on the MRI, then combines those findings with pathology results and clinical information to generate a personalized estimate of how likely the cancer is to return over the next five to ten years. The result is a score from 0 to 100 that places a patient in a low, intermediate or high risk group.
According to SimBioSys president and CEO Stacey Stevens, because the estimate can be available before surgery, it gives women and their doctors more information sooner as they discuss treatment options. The company previously developed TumorSight Viz, which turns a breast MRI into a 3D model of the tumor to help surgeons visualize it.
Could this change breast cancer treatment?
Doctors would consider the TumorSight Risk result alongside other tests and clinical findings. Dr. Poornima Saha, a breast medical oncologist at Endeavor Health, says the tool may help build a more complete picture, so doctors and patients can weigh each woman's risk, potential side effects and personal priorities together.
Treatment to lower the chance of recurrence can include surgery, radiation, chemotherapy or hormone-blocking medications, some of which come with difficult side effects. For hormone receptor-positive breast cancer, these conversations can continue well after initial treatment, including whether to keep taking hormone therapy beyond five years. A clearer picture of recurrence risk could help doctors identify which patients are most likely to benefit.
The technology is starting to be used in hospitals and cancer centers across the country. Patients who are curious can ask their cancer care team whether it's available at their treatment center and whether it's right for them.
This post is for general information and is not medical advice.
Source: Adapted from reporting by Ava Tsapatsaris and Dr. Tara Howard for ABC News, "New FDA-cleared AI tool could help predict if breast cancer will return," October 15, 2026.





Comments