Abstract
Monitoring the training load is crucial in sports science research, as it provides scientific evidence for assessing training effects, optimizing athletic performance, and preventing overtraining by quantifying both external and internal loads. Although traditional monitoring methods have made significant progress, infrared thermography (IRT) technology, with its non-contact, real-time, and non-invasive characteristics, is gradually emerging as an effective tool for evaluating the relationship between the training load and physiological responses. This study evaluated 31 healthy male adults (age 21.9 ± 2.7 years, weight 75 ± 8.26 kg, and training duration 240 ± 65 min/week) performing incremental exhaustive exercise on a cycle ergometer (with a 60W starting load, increasing by 20W per minute). Entropy analysis was used to quantitatively assess the surface radiation patterns of regions of interest (forehead, chest, and abdomen) obtained through thermal imaging. Compared to baseline, significant differences in the surface radiation patterns of the regions of interest were observed at the point of exhaustion (p ≤ 0.01). Correlation analysis revealed strong associations between the external load, oxygen consumption, and chest temperature entropy (r = 0.973 and 0.980). Cluster analysis of the chest entropy, external load, and oxygen consumption showed a non-linear increasing trend in their inter-relationships. Further individual analysis demonstrated positive correlations between the percentage increase in the chest entropy and both the external load (r = 0.70-0.98) and oxygen consumption (r = 0.65-0.97). Entropy analysis offers a new approach for quantitatively assessing surface radiation patterns from infrared thermography, and reveals the coupling relationship between thermoregulation and metabolic responses during exercise.