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

Area of Science:

• Biomedical Engineering
• Ophthalmology
• Neurology

Background:

• Diabetic neuropathy (DN) is a common diabetes complication affecting sympathetic nerves.
• Neuropathy impacts skin's thermal regulation and sweat duct morphology.
• Current diagnostic methods for DN can be invasive or lack sensitivity.

Purpose of the Study:

• To investigate the correlation between sweat duct characteristics and DN severity.
• To develop a predictive model for DN using optical coherence tomography (OCT) and machine learning.
• To explore OCT-assessed sweat duct features as non-invasive biomarkers for DN.

Main Methods:

• Fingerprint-guided sweat ducts were assessed using OCT.
• Principal component analysis (PCA) and back propagation neural network (BPNN) were employed for predictive modeling.
• Spatial distribution analysis using Voronoi diagrams was performed.

Main Results:

• Sweat duct number, volume, and spacing showed significant correlation with DN severity.
• Irregularities in sweat duct spatial distribution were observed in DN patients.
• The PCA-based BPNN model achieved good predictive accuracy for different DN stages.

Conclusions:

• OCT-assessed sweat duct features are correlated with DN severity.
• The developed predictive model demonstrates potential for accurate DN classification.
• OCT imaging of sweat ducts offers a promising non-invasive approach for DN biomarker discovery.