OPTIMIZING HEALTH MONITORING SYSTEMS IN SPORTS FACILITIES USING GREY WOLF OPTIMIZATION ALGORITHM AND TIME SERIES ANALYSIS

Abstract

In sports medicine, the rapid and accurate detection of physiological abnormalities is crucial, especially during high-intensity training or competition. This research introduces a novel approach to monitoring athletes' health by leveraging Grey Wolf Optimization (GWO) and time series analysis for fault detection in health monitoring systems used in sports settings. Our methodology employs advanced machine learning technologies and signal processing to detect, categorize, and localize physiological abnormalities from the data collected via wearable sensors. The use of the Continuous Wavelet Transform (CWT) enables the extraction of informative features from physiological data, enhancing the detection capabilities. The GWO algorithm optimizes the classification algorithm, improving the accuracy of fault localization in real-time health monitoring. To validate the efficacy of our model, we implemented it using Python 3.11 on a high-performance computing platform and compared its performance with traditional methods using metrics such as precision, recall, F1-score, Mean Absolute Percentage Error (MAPE), and Mean Squared Error (MSE). Preliminary results suggest that this optimized fault detection and localization approach significantly enhances the reliability of health monitoring systems in sports, potentially reducing the risk of injury by providing timely interventions. This method not only supports the resilience and optimization of athlete health management but also aligns with the growing need for sophisticated health monitoring solutions in sports.