Indoor Heat Strain Early Detection Using Heart Rate Variability

The increasing frequency and severity of heatwaves, exacerbated by climate change, pose significant health risks, particularly in underserved communities where high indoor temperatures are a growing concern. This study evaluates the applicability of root mean square of successive differences (RMSSD) in heart rate variability (HRV) as an indicator of heat strain, while highlighting inter-individual differences in autonomic responses. Eight healthy adults were exposed to controlled heat conditions ranging from 24°C to 40°C with relative humidity between 18% to 32% in a climate chamber. Non-invasive sensors recorded core temperature and HRV. K-Nearest Neighbors and Random Forest regressors were employed to predict RMSSD using lagged environmental and physiological features. Model performance varied among individuals, showing moderate to high performance, with adjusted R² values ranging from 0.67 to 0.89. These findings highlight the potential of RMSSD for enabling proactive building thermal management strategies to enhance occupant comfort and health during extreme heat events.