Machine Learning Application in Genomic, Exercise, and Vital Datasets

Abstract

PURPOSE: Machine learning (ML) refers to newly developed computer algorithms that are improved through iterative experiences. ML applications are expected to assist humans in analyzing large amounts of data. This review has outlined the application of ML in analyzing variable vital data such as walking steps, exercise intensity, heart rate, sleeping hours, sleep quality, resting heart rate, blood pressure, and calorie consumption in a day. Vital data consist of different variables that are closely related to genomic or exercise data. The prediction of healthy traits from a vital dataset has become a necessity in personalized medicine.METHODS: Considerations and repeated tasks in supervised, semi-supervised, and unsupervised ML methods are presented. ML methods such as artificial neural networks, Bayesian networks, support vector machines, and decision trees have been widely used in biomedical studies to develop predictive models. Through vital data, these models can help in effective and accurate decision-making for a healthier life.<br/>PURPOSE: Models based on genomic, exercise, and vital datasets provide a healthy lifestyle through regular exercise. We have provided guidelines to help in the selection of these ML methods and their practical application for variable vital data analysis.CONCLUSIONS: Our guidelines could serve as a foundation for implementing both participatory medicine and data-driven exercise science.