New guidelines recommend AI-based breast cancer risk assessments
Artificial intelligence has officially made its way onto the National Comprehensive Cancer Network’s (NCCN) Clinical Practice Guidelines.
The 2026 update now recommend the use of image-based artificial intelligence risk assessment as a primary tool for identifying individuals at increased risk of developing breast cancer. Specifically, NCCN suggests that five-year future breast cancer risk assessments based on routine mammograms should be integrated into standard practices.
The group also offered additional specifics relative to thresholds, noting that women with an AI-based five-year risk of developing breast cancer that is equal to or greater than 1.7%, as determined by AI analysis, should be considered for additional imaging.
For those determined to be above the threshold, NCCN recommends:
Discussions with their provider regarding supplemental imaging, such as MRI or ultrasound.
Lifestyle modifications based on individual risk levels.
Additional recommendations include:
Risk assessments starting at age 35, not 40.
Periodic re-assessments of risk over time, due to the potential for longitudinal changes in breast tissue.
The guidelines pointed specifically to the use of tools such as the Clairity Breast platform, which is the first FDA-cleared AI tool that predicts breast cancer risk based on routine mammograms. Clinical trial data suggest that the software is over three times more accurate than radiologist-reported breast density at predicting breast cancer risk.
“For decades, we’ve known that the mammogram contains critical information—not just about the presence of cancer, but about a woman’s future risk,” Connie Lehman, MD, PhD, founder and CEO of Clairity, said in a statement. “Advances in AI now allow us to extract that information in a clinically meaningful way."
“We have long relied on family history, genetic testing and breast density to assess breast cancer risk, but these approaches fail to identify many women at higher risk” added Robert Smith, PhD, director of the American Cancer Society Center for Early Cancer Detection. “Most women diagnosed with breast cancer have no known genetic mutation or strong family history, and although breast density is common, is too broad a measure to meaningfully stratify individual risk. This situation highlights a fundamental gap in our current approach, where AI-based analysis of mammograms represents an important new direction to overcome these limitations, and hopefully move toward more precise and individualized risk assessment.”
Read more here.
The NCCN Guidelines can be found here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.