The National Cancer Institute, part of the National Institutes of Health (NIH), has awarded Aimilia Gastounioti a five-year $3 million grant to improve breast cancer risk assessments for Black women. Gastounioti is an assistant professor of radiology in the Computational Imaging Research Center in for Mallinckrodt Institute of Radiology (MIR) at Washington University School of Medicine in St. Louis. This is her first R01 grant.
The project aims to address an important need for a patient population that is not well served by current risk assessment tools. Black women are slightly less likely to develop breast cancer than white women are, but their mortality rates for the disease are strikingly higher — they are 40% more likely to die from the disease than white women. The risk models that are used to identify and plan treatments for patients who may develop the disease are not well suited, and historically have been less accurate, for Black women.
New tools that provide personalized risk assessments using artificial intelligence have shown enormous promise, but so far have largely been trained on data from digital mammography of white patients. Gastounioti, a principal investigator with MIR’s Computational Imaging Research Center, and her team will develop deep learning and medical imaging informatics tools on a database of more than 95,000 digital breast tomosynthesis (DBT) exams from Black women. The database is a collaboration of WashU, Emory University in Georgia, University of Pennsylvania and Columbia University Irving Medical Center in New York. Unlike standard digital mammography, DBT captures images of breast tissue from multiple angles, resulting in more-detailed tissue representations.
The goal is to develop and distribute an accurate screening tool specific to the patient population, which has the potential to drastically improve early breast cancer detection and prevention for Black women.
