Predicting peak cardiorespiratory fitness in patients with cardiovascular disease using machine learning

Abstract

Objective This study aimed to develop machine learning (ML) models to predict peak cardiorespiratory fitness (CRF) before and after cardiac rehabilitation (CR). Methods and Results Data from 162 patients with cardiovascular disease were analyzed. Two predictive tasks were employed: Task 1 estimated peak oxygen consumption (VO2 peak) using baseline clinical and functional data and Task 2 predicted changes in VO2 peak after CR by additionally considering inter-visit exercise quantities and pre-CR cardiopulmonary exercise test (CPET) results. Four linear regression models and six ML models were trained and validated through 5-fold cross-validation technique. Both tasks demonstrated that the CatBoost and XGBoost models exhibited the highest predictive performance, effectively forecasting VO2 peak values before and after CR. Task 1 highlighted the importance of the six-minute walk distance (6MWD), Korean Activity Scale Index (KASI), and hand grip strength (HGS) in predicting the initial VO2 peak. Task 2 suggested a ceiling effect in the recovery of VO2 peak following CR and emphasized the importance of resistance exercise. Conclusion The application of ML models provides a powerful tool for predicting the peak CRF in patients with CVD undergoing CR, both at the initial assessment and after completing rehabilitation programs.