Speaker
Description
This paper explores the use of infrared thermography to analyze body heat distribution patterns in cyclists, to evaluate its potential contribution to professional bike fitting. Seven participants completed a standardized 15-minute cycling exercise, during and after which thermographic images were captured. The goal was to detect temperature asymmetries or irregularities that may indicate discomfort, poor posture, or injuries. The thermal data revealed visible patterns, particularly in the knees, upper back, and hands, that may reflect uneven load distribution, muscular imbalance or injuries. Although promising, the study also faced limitations. Specifically, attempts to automate injury detection using the MoveNet pose estimation model were unsuccessful due to its poor performance on infrared images, as it was originally trained on RGB datasets. Given the small sample size and preliminary nature of the findings, further research with a more robust methodology is needed. Nonetheless, the study demonstrates that thermography holds potential as a non-invasive supplementary tool in cycling ergonomics and injury prevention.
