Agreement between box- and ellipse-shaped software-derived regions of interest in the assessment of exercise-related skin vasomotor adjustments using thermography

Abstract

During the measurement of skin temperature in sports by thermography, different factors could affect the values obtained, being one of them the software-derived analysis method. Therefore, this study aimed to explore the differences between box- and ellipse-shaped software-derived regions of interest in assessing exercise-related skin vasomotor adjustments using thermography. The skin temperature (Tsk) of forty-two male experienced runners was evaluated during three different conditions (rest, after cold water immersion, and after a 5-km running time trial test) in four muscles (quadriceps, hamstrings, tibialis and triceps sural). The data obtained from the thermograms were analyzed using two software-derived analysis methods (box-vs ellipse-shaped region of interest). The potential between-methods differences were explored in the three scenarios through a mixed analysis of variance. No effect between software-derived analysis methods was found (F<0.50, p>0.50, ω_p² = 0), with excellent reliability (ICC> 0.90). As expected, differences have been shown in maximum, minimum and average Tsk by conditions with large effect size (F > 279.5, p<0.01, ω_p²>0.80). Finally, no differences were found in the interaction between variables (F<0.10, p>0.90, ω_p² = 0). In conclusion, these findings support the use of both software-derived analysis methods to measure skin temperature through thermography, independently of body muscles and type of activity. In this sense, the data obtained by box- and ellipse-shaped software-derived regions of interest can be compared due to the agreement of measurements for skin temperature.