Application of Machine Learning in Predicting Exercise Adherence Among Cardiovascular Disease Patients: A Scoping Review

Cardiovascular disease (CVD) is one of the most prevalent and serious health concerns worldwide, with its high incidence and mortality rates posing a significant burden on public health systems. For patients with cardiovascular disease, maintaining regular physical activity is crucial for improving heart function, reducing complications, and enhancing overall quality of life. Exercise not only benefits cardiovascular health but also significantly lowers the risk of recurrent cardiovascular events. However, despite the well-established importance of exercise, many patients fail to adhere to post-treatment exercise recommendations, leading to poor exercise adherence. This can hinder recovery, increase the risk of complications, and place additional strain on healthcare systems. Given this issue, early identification of patients with poor exercise adherence and the timely implementation of targeted interventions are critical for improving patient outcomes. With the advancement of artificial intelligence, the application of machine learning in developing predictive models for exercise adherence has become a promising approach for the early detection of at-risk patients, enabling personalized interventions, and optimizing prognosis management. This study aims to conduct a systematic scoping review of the existing machine learning-based prediction models for exercise adherence in cardiovascular disease patients to guide future research in this field.