Comparison of Image Analysis Methods in Skin Temperature Measurements during Physical Exercise

The analysis of thermographic images in biomedical application is usually performed by different methods of averaging the values of pixel temperature inside a Region of Interest (ROI). We used a new method, based on maximal temperatures inside a ROI (Tmax), to evaluate temperature modifications during physical exercise in trained or untrained subjects. In this paper we analysed squat exercise effects on the thighs of 6 active subjects. Data were compared with those obtained with traditional method based on the average over an area in the anatomical part under investigation. The values of temperature measured before, during and after the exercise showed the same pattern with both methods. A slightly decrease in temperature during the exercise was followed by an increase during the recovery time. Bland and Altman statistical analysis confirmed the good agreement between the new method and the one based on the average considered as the gold standard. In this study we put in evidence the features of this analysis method specifically in human application.