Machine Learning-Based Prediction of Survival and Mitral Regurgitation Recurrence in Patients Undergoing Mitral Valve Repair

The authors used machine learning techniques on twenty-year outcome data from 436 consecutive patients who underwent mitral valve repair over an eighteen-year period. The endpoints were actuarial survival and freedom from moderate or high mitral regurgitation (MR). Five machine learning models were used, and concordance indices (C-indices) were compared. The study shows that machine learning models were able to predict overall mortality and MR recurrence after mitral valve repair. The C-indices of machine learning models were higher than those of the Cox model. Further validation will be required.