Novel artificial intelligence applications in heart transplantation

Abstract

The advent of Artificial intelligence (AI) in medicine has allowed for transformative advances across clinical domains. In the field of heart transplantation (HT), existing paradigms for patient identification, transplant outcomes optimization, post-transplant surveillance and more are amenable to improved performance by Machine Learning (ML) and other models. Studies published thus far have largely relied on retrospective designs and focused on clinical outcomes; nevertheless, emerging evidence is tackling organ allocation considerations, optimal donor-recipient matching schemes, predictors of primary graft dysfunction (PGD) and reduced survival. This review will survey current and future applications of AI in HT, and identify focus areas for clinicians and researchers.