Artificial intelligence is opening a new window into breast cancer screening, with researchers testing whether years of mammograms can reveal risk long before a diagnosis. The approach is designed to help doctors identify women at higher risk before cancer develops.

In a study published in the American Journal of Roentgenology, researchers at NYU Langone Health and its Perlmutter Cancer Center evaluated an AI model. It was built to estimate a woman's likelihood of developing breast cancer within five years.

Rather than examining only the latest image, the tool reviews several years of a woman's 3D mammograms. By following breast tissue changes over time, it searches for clues that may not be captured by an assessment focused on a single scan.

The model was developed using more than 313,000 mammograms from more than 161,000 women, according to the study. That extensive image collection allowed researchers to evaluate how tissue patterns changed across multiple years.

Connie Lehman, a Massachusetts breast imaging specialist and professor of radiology at Harvard Medical School, was not involved in the study. She told Fox News Digital that AI may identify increased risk that would otherwise escape notice.

"For decades, as a doctor and breast imaging specialist, I have cared for women who were shocked by a breast cancer diagnosis," Lehman said. "Again and again, I heard: 'How can this be? No one in my family has had breast cancer.'"

Traditional breast cancer risk assessments can consider factors such as genetic mutations and family history. AI may add another source of information by examining clues embedded within the breast tissue visible on mammograms.

"A mammogram can tell us more than whether cancer is visible today — it can also provide information about a woman's risk of developing cancer in the years ahead," Lehman said.

That information could eventually help doctors adjust screening according to each woman's individual level of risk. Women found to have increased risk could speak with their health care providers about additional screening, including MRI.

Women considered to have average risk may continue receiving routine mammograms, according to Lehman. The AI analysis could therefore support more personalized conversations about screening without functioning as a diagnosis by itself.

The technology could also examine mammograms that a woman has already received. Because it uses existing images, obtaining the additional risk information would not require another imaging examination or expose the woman to additional radiation, Lehman noted.

Lehman cautioned that an AI risk score cannot determine with certainty whether a woman will develop breast cancer. "A higher score can prompt a conversation with your health care provider about additional screening and risk-reduction options," she said.

A result suggesting lower risk also has important limits and should not lead women to abandon routine screening. "A lower score does not eliminate risk or replace regular mammography," Lehman said.

The NYU model remains a research tool and has not received FDA authorization for clinical use, according to Lehman. Researchers also acknowledged that additional testing will be necessary before the technology can move beyond its current research setting.

The study relied on mammograms collected at NYU Langone hospitals and used imaging equipment from one manufacturer. That limitation means the model still needs evaluation with images collected in other clinical environments and on different equipment.

The research team plans to test the model with data from other medical centers and different 3D mammography systems, according to NYU Langone. Fox News Digital also contacted the researchers for comment.

Dr. Elisa Port, Mount Sinai's Breast Surgery Director and author of “The Breast Advice,” has highlighted how AI can uncover hidden tumors and reduce false positive results, particularly among women with dense breasts. The technology also aims to improve screening adherence among the 30% to 40% of women who do not receive regular mammograms.

For now, the research presents AI as a possible addition to breast cancer risk assessment rather than a replacement for mammography or medical guidance. Its potential rests on finding meaningful patterns across years of existing breast images while preserving regular screening and discussions with health care providers.