K-Nearest Neighbour-based classification of skin barrier damage using Raman imaging: A non-invasive assessment method...

The stratum corneum is essential for maintaining the skin barrier and preventing transepidermal water loss. Damage to this layer increases skin sensitivity, potentially leading to inflammation and related complications.

This study aimed to establish reproducible skin damage models and to develop a non-invasive classification method for assessing the severity of skin barrier damage.

Two controlled models of injury, tape stripping (TS) and friction-induced injury, were used to simulate barrier impairment. Raman imaging was employed to analyze biomarkers, including stratum corneum thickness, lipid content, and keratin levels. These parameters were combined with the K-nearest neighbour (KNN) algorithm to categorize skin damage severity.

The integrated KNN-based model successfully classified skin damage severity ranging from normal to severe. It also enabled the prediction of severity in unknown samples, providing a robust diagnostic framework.

This non-invasive model demonstrates strong potential for diagnostics and monitoring in dermatology.

KNN-based classification combined with Raman imaging offers a reliable method for evaluating skin barrier damage severity.