Researchers at UMass Chan Medical School are exploring the use of artificial intelligence to improve breast cancer detection, potentially identifying women at higher risk and detecting cancers that standard mammograms may miss. The function centers on an AI-driven risk assessment model developed in collaboration with investigators at the Massachusetts Institute of Technology.
Breast cancer remains a leading cancer among women in the United States, and early detection is crucial for improving outcomes, according to the American Cancer Society. While mammography is the current standard screening method, it has limitations, particularly for women with dense breast tissue, where tumors can be more difficult to visualize.
The AI tool analyzes routine screening mammograms and assigns a risk score estimating a woman’s likelihood of developing breast cancer within the next few years. Mohammed Salman Shazeeb, PhD, associate professor of radiology, and Gopal Vijayaraghavan, MD, MPH, professor of radiology, are leading the research. The project is supported by grants from state agencies and the Breast Cancer Research Foundation.
In a study of the first 145 participants, researchers found that MRI screenings – conducted on women who scored above a predetermined risk threshold – identified four additional cancers that were not detected by initial mammograms. This represents a significantly higher yield than typically observed with mammography alone, according to researchers.
“Among the roughly 6 to 7 percent of women who scored above our risk threshold, we invited them for contrast-enhanced breast MRI,” said Dr. Shazeeb. “What’s striking is that all had normal screening mammograms, yet MRI found cancers in some of them that we would otherwise have missed.”
MRI is currently considered the gold standard for detecting many breast cancers, but its widespread use is limited by its cost and time requirements. Dr. Vijayaraghavan explained that the AI tool aims to focus these resources on women at the highest risk, potentially making early detection more efficient and personalized.
The AI models function by detecting subtle imaging features that may be imperceptible to the human eye, leveraging patterns learned from extensive datasets. “The AI can process many more features on an image than a radiologist can visually,” Vijayaraghavan said. “But it doesn’t feel the way a physician does. The tool is trained for performance, not understanding. That’s why these tools are designed to augment, not replace, clinical judgment.”
The Breast Oncology Fellowship at UMass Chan Medical School-Baystate, accredited by the Society of Surgical Oncologists and the American Society of Breast Surgeons, offers a one-year multidisciplinary training program centered at the Baystate Breast & Wellness Center, the largest dedicated breast surgery practice in western Massachusetts. The fellowship emphasizes a patient-centered approach, shared decision-making, and provides training in all aspects of breast disease, women’s health, and survivorship.
Researchers acknowledge that several hurdles remain before AI-guided risk assessment can be routinely implemented in clinical practice. These include obtaining FDA approval, establishing appropriate reimbursement policies, and ensuring equitable access for all populations. “Before this can be widely applied, we necessitate larger-scale validation and real-world implementation data,” Shazeeb said. “We also have to ensure that it works fairly across diverse populations and different mammography systems.”
A UMass Chan Medical School research team led by Arthur Mercurio, a professor of molecular, cell, and cancer biology, has also been awarded $2.6 million from the National Institutes of Health to develop a therapeutic treatment addressing radiation resistance in triple-negative breast cancer.
Patient engagement is also a key consideration. Sara Schiller, senior research program manager in the Department of Radiology, noted that many patients are receptive to the use of AI in imaging, particularly those with a family history of breast cancer who are eager to contribute to research.
“Many women I speak with just want to help further research and are not hesitant about AI per se. Many have family histories of breast cancer and are eager to contribute,” Schiller said.
Experts emphasize that AI is intended to be a tool to personalize screening, not to replace existing methods. “The goal isn’t to replace mammograms,” said Vijayaraghavan. “It’s to add another layer of insight, a decision support tool, that helps us uncover cancers earlier, when treatment is more effective and less invasive.”
