Project Grant 2516270

The National Science Foundation (NSF) awarded a $441,404 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) federal grant program to the University of California, Berkeley. This 2.5-year grant, beginning January 1, 2025, will support research aimed at improving the fairness and equity of medical decision-making, both for human clinicians and algorithmic systems. The project will analyze sources of bias in high-stakes medical decisions such as test allocation, healthcare quality assessment, and medical image interpretation. It will then develop algorithmic decision aids to help reduce human bias by highlighting medically relevant features clinicians may overlook. Additionally, the research will examine which features are appropriate to include in medical algorithms to make them more equitable. The project will work closely with clinicians to maximize the translational benefits to patients. The methods developed, which use Bayesian inference and deep learning techniques, can be applied more broadly to enhance equity in high-stakes decision-making across domains like lending and hiring.