Abstract
Peripheral circulatory failure refers to a condition in which the blood flow through superficial capillaries is markedly reduced or completely occluded. In clinical practice, nurses strictly adhere to regular repositioning protocols to prevent peripheral circulatory failure, during which the skin condition is evaluated visually. In this study, skin colour changes resulting from pressure application were continuously captured using a camera, and supervised machine learning was employed to classify the data into two categories: before and after pressure. The evaluation of practical colour space components revealed that the h component of the JCh colour space demonstrated the highest discriminative performance (Area Under the Curve (AUC) = 0.88), followed by the a* component of the CIELAB colour space (AUC = 0.84) and the H component of the HSV colour space (AUC = 0.83). These findings demonstrate that it is feasible to quantitatively evaluate skin colour changes associated with pressure, suggesting that this approach can serve as a valuable indicator for dimensionality reduction in feature extraction for machine learning and is potentially an effective method for preventing pressure-induced skin injuries.