Non-Invasive Physiological Signal Analysis for Stress and Fatigue Detection in Wearable Technology

This study analyses physiological signals collected from wearable devices during structured stress induction and exercise sessions. Data from 94 participants were used to compare physiological responses across gender and activity conditions (aerobic, anaerobic, and stress). The study focused on features such as accelerometer (ACC), blood volume pulse (BVP), electrodermal activity (EDA), heart rate (HR), inter-beat interval (IBI), and skin temperature (TEMP). Results revealed significant physiological differences between genders and activity conditions, with an ANOVA test confirming significant variations in HR across conditions (p < 0.0001). These findings contribute to improving stress and exercise classification models using wearable sensor data.