MACHINE LEARNING USING SIMILARITY ANALYSIS IMPROVES RISK STRATIFICATION BEYOND SURGICAL RISK SCORES IN PATIENTS UNDERGOING TRANSCATHETER AORTIC VALVE REPLACEMENT

Current methods to aggregate the risk factors in patients undergoing transcatheter aortic valve replacement (TAVR) remains imprecise. We used unsupervised machine learning to predict 30-day adverse events including mortality following TAVR. We included data from 228 patients who underwent TAVR at