Cognitive Load Assessment Using Smartwatch ECG Data: Image Processing and Pan-Tompkins Algorithm for Heart Rate Variability Analysis

This study examines the impact of cognitive and physical demands on Heart Rate Variability (HRV) by utilizing smartwatch ECG data. HRV, assessed through RMSSD, acts as a physiological indicator of stress. Correlation analysis indicated a strong negative relationship between Mental Demand and RMSSD (r = -0.9932, p < 0.001), implying increased physiological stress under cognitive load. A multiple linear regression (MLR) model accounted for 50% of the variance in RMSSD (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.500), with Physical Demand (p = 0.004) and Frustration (p = 0.063) identified as significant predictors. Additionally, a mixed-effects model revealed considerable individual variability, especially regarding Temporal Demand (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.630, p < 0.0001) and Mental Demand (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.8959, p = 0.0008). Kruskal-Wallis tests affirmed significant differences in RMSSD across all workload factors (p < 0.0001). These findings underscore RMSSD as a reliable biomarker for real-time assessment of cognitive load, with important implications for workload management and stress monitoring.