ML-based Sleep Disorder Identification for Athletic Performance Enhancement Using Wearables
Priyal Shah, Stuti Gulati, Lakshin Pathak, Rajesh Gupta, Sudeep Tanwar · 2025
This paper describes the application of machine learning (ML) in sleep disorder diagnosis and monitoring of athlete activities by introducing new algorithms that classify data using K – Nearest Neighbors (KNN), Logistic Regression (LR), Naive Bayes (NB), and Random Forest (RF) to increase both the diagnostic and analytic capabilities of the system. Moreover, the developed application is supported by e-healthcare systems and smart dwelling environments to watch and evaluate sleep quality parameters and other measures important for athletics. Our study show that KNN managed a remarkable 93% accuracy, which was the best of all other models used in classifying sleep disorders, and performance measurement of the athletes was also easy. The opportunity to integrate with IoT gadgets further expands this by ensuring that data can be captured continuously and introducing a new dimension of preventive medicine, personalized medicine and even better training for the athletes. It is noted that the opportunity to use such systems at the interface of ML and connected health could potentially improve health delivery outcomes for patients and athletes while offering a cost-effective system for health monitoring.