Correlation Between Fingerprint-Guided Sweat Ducts Features From OCT and Diabetic Neuropathy Using Voronoi Diagram.

Abstract

Diabetic neuropathy (DN) is a prevalent chronic complication of diabetes. Sweat glands are directly controlled by the sympathetic nervous system, whose neuropathy affects the thermal regulation of the skin and results in morphological changes in sweat ducts. This study aims to investigate the correlation between the characteristics of fingerprint-guided sweat ducts assessed by optical coherence tomography and DN based on a predictive model using a back propagation neural network (BPNN) and principal component analysis (PCA). The results demonstrate that the number, volume, and spacing of sweat ducts are correlated with the severity of DN. The Voronoi diagram of the sweat duct distribution demonstrates irregularities in the spatial distribution among patients with DN. Furthermore, the PCA-based BPNN model has good predictive accuracy between patients with non-neuropathic, neuropathic, and severe neuropathic diabetes. These findings suggest that OCT-assessed sweat duct features may serve as non-invasive biomarkers for DN in patients with diabetes.