Thermopixelgraphy: a new method for analyzing the skin temperature of the lower limbs through infrared thermography

Abstract

This study proposes a new method called “thermopixelgraphy” for analyzing thermal images (TI) through quantifying pixels classified as cold, neutral and hot. For this, we analyzed 30 professional male football players (27.2 ± 4.4 yrs., 69.8 ± 6.6 kg; of 9.8 ± 3.0% of body fat). Thermographic images were obtained in 13 games played in the Brazilian league. Images of the lower limbs (anterior and posterior) were obtained at three moments: M1 (24 h before); M2 (24 h after); M3 (48 h after the match). The total number of pixels in each image were quantified and classified as: (a) cold 28.0–31.0 °C; (b) neutral 31.1–33.0 °C and; (c) Hot 33.1–36.0 °C. The main results showed a higher frequency of pixels in the cold (mainly between 30 and 31 °C; n = 1551.9) and neutral zones (mainly between 32 and 33 °C; n = 2253.9) for the anterior view of the right lower limb at moment M1 and a higher frequency of pixels in the neutral (mainly between 32 and 33 °C; n = 2622.8) and hot zones (mainly between 33 and 34 °C; n = 2579.5) at moment M2 for the posterior view (M1: cold zone 30–31 °C; n = 1319.1; neutral zone 31–32 °C; n = 2835.8; M2: neutral zone 32–33 °C; n = 2939.5; and hot 33–34 °C; n = 3066.5). Similar results were observed for the left leg. In conclusion, “thermopixelgraphy” quantifies the number of pixels from TI and the image analysis process is faster, more accurate, encompasses all areas of the thigh and leg muscles, and does not require extensive prior experience from the evaluator.