Artificial intelligence bias in the prediction and detection of cardiovascular disease

Abstract

AI algorithms can identify those at risk of cardiovascular disease (CVD), allowing for early intervention to change the trajectory of disease. However, AI bias can arise from any step in the development, validation, and evaluation of algorithms. Biased algorithms can perform poorly in historically marginalized groups, amplifying healthcare inequities on the basis of age, sex or gender, race or ethnicity, and socioeconomic status. In this perspective, we discuss the sources and consequences of AI bias in CVD prediction or detection. We present an AI health equity framework and review bias mitigation strategies that can be adopted during the AI lifecycle.