Detecting Stress from Competitive Gaming using Machine Learning-Based ECG Classification

The cardiovascular health impacts of acute stress emanating from present-day activities, including competitive gaming, have been increasingly realized. Although conventional stressors like physical exercises have been extensively studied, modern digital stressors have largely remained obscure. This investigation aimed at understanding how competitive gaming induced acute stress in the electrocardiogram (ECG) signals using advanced machine learning algorithms for more detailed analysis. The traditional approaches to assessing stress fall short because they fail to capture the subtle physiological responses to such stimuli. Electrical recordings of the heart were taken from thirty volunteers when relaxed and stressed. The signals were subjected to machine learning algorithms that showed slight changes between normal and stress-induced heart activity. In this research study, a significant variation among ECG patterns was found under different conditions of these stresses indicating insights into the effect of contemporary pressures on cardiac function. Results indicated notable changes in ECG features such as peak amplitudes and heart rate variability during stress demonstrating a 95% classification accuracy with Linear Discriminant Analysis (LDA). In this oracular accuracy, the method was demonstrated to be highly efficient in differentiating between normal and stressed states. These observations enhance the knowledge of how current stressors affect heart well-being and present a novel, precise method for analyzing ECGs that may improve clinical practices, giving more insight into heart disease risk due to stress and bettering management of pressure-related disorders.