In 2024, The American Cancer Society estimates there will be approximately 310k new cases of breast cancer. Tragically, about 42k of these cases will prove fatal. As with most cancers, early detection is critical to survival, and thanks to the research of Connie Lehman and Regina Barzilay, breast cancer may be predicted before it even presents. Through decades-long explorations, these researchers have developed AI models that predict a woman’s risk of developing breast cancer years before any signs appear, offering the potential for personalized screening and earlier interventions that could save countless lives.
Barzilay, a computer scientist at MIT, was inspired by her own battle with breast cancer. After enduring invasive treatment, she redirected her research focus, dedicating herself to exploring whether AI could diagnose tumors earlier and spare other women from the ordeal she experienced. In 1998, Lehman, a Harvard radiologist, began working on computer-aided design (CAD) to improve the detection of breast cancer. “Although studies in the lab found that CAD could make a difference, it wasn’t having the impact we were hoping for in the clinic,” she recalls.
Thanks to technological advancements, Lehman’s and Barzilay’s vision is finally becoming a reality. Krzysztof Geras, assistant professor of computer science at New York University, explains “better machinery” and the development of “deep learning, a subset of AI, that gives models the capacity to learn from large amounts of data.” In collaboration, Lehman and Barzilay have created Mirai, “an open source, state-of-the-art deep learning model that can provide a personalized risk score up to 5 years in advance just by analyzing a patient’s mammogram.” Not only is Mirai’s effectiveness impressive, but it also accurately identified cancer in 76 out of 100 mammograms, and its results were consistent across diverse populations.
With tools like Mirai, Lehman and Barzilay are not just improving detection; they are paving the way for a new standard of personalized healthcare that could transform how breast cancer is prevented and treated worldwide.
SOURCES
https://www.bcrf.org/blog/ai-breast-cancer-detection-screening/
https://medicine.wustl.edu/news/ai-assisted-breast-cancer-screening-may-reduce-unnecessary-testing/
https://www.thinkglobalhealth.org/article/ai-gains-ground-breast-cancer-diagnosis-and-prevention