AI Detects Breast Cancer Risk Before Humans, But Hospital Adoption Faces Hurdles
Researchers at the Massachusetts Institute of Technology have developed an artificial intelligence system capable of detecting subtle signs in mammograms that indicate a high risk of developing breast cancer, often before human radiologists can identify them. The algorithm, named Mirai, analyzes medical images to calculate the probability of cancer development within five years, acting as a predictive warning system. This technology aims to facilitate earlier interventions for patients who lack traditional risk factors such as genetic mutations or family history. Despite its potential to save lives by identifying precancerous conditions early, widespread adoption in hospitals remains slow due to regulatory hurdles, insurance coverage issues, and implementation costs. The article highlights the personal story of Ellen Costello, whose precancerous growth was detected by the AI, and features insights from MIT professor Regina Barzilay, who created the tool following her own breast cancer diagnosis. While studies involving over two million mammograms demonstrate the model's effectiveness across diverse demographics, it currently remains primarily in the research phase, with gradual integration into clinical practice expected as guidelines evolve.
Wire timeline
AI Detects Breast Cancer Risk Before Humans, But Hospital Adoption Faces Hurdles
Researchers at the Massachusetts Institute of Technology have developed an artificial intelligence system capable of detecting subtle signs in mammograms that indicate a high risk of developing breast cancer, often before human radiologists can identify them. The algorithm, named Mirai, analyzes medical images to calculate the probability of cancer development within five years, acting as a predictive warning system. This technology aims to facilitate earlier interventions for patients who lack traditional risk factors such as genetic mutations or family history. Despite its potential to save lives by identifying precancerous conditions early, widespread adoption in hospitals remains slow due to regulatory hurdles, insurance coverage issues, and implementation costs. The article highlights the personal story of Ellen Costello, whose precancerous growth was detected by the AI, and features insights from MIT professor Regina Barzilay, who created the tool following her own breast cancer diagnosis. While studies involving over two million mammograms demonstrate the model's effectiveness across diverse demographics, it currently remains primarily in the research phase, with gradual integration into clinical practice expected as guidelines evolve.
WBUR News