Facial erythema detects diabetic neuropathy using the fusion of machine learning, random matrix theory and self organized criticality

Knud B Yderstraede,Esmaeil S Nadimi,Victoria Blanes-Vidal,Tomas Majtner

doi:10.1038/s41598-020-73744-3

Knud B Yderstraede, Esmaeil S Nadimi + Show 2 more

Open Access

https://doi.org/10.1038/s41598-020-73744-3

Copy DOI

Abstract

Rubeosis faciei diabeticorum, caused by microangiopathy and characterized by a chronic facial erythema, is associated with diabetic neuropathy. In clinical practice, facial erythema of patients with diabetes is evaluated based on subjective observations of visible redness, which often goes unnoticed leading to microangiopathic complications. To address this major shortcoming, we designed a contactless, non-invasive diagnostic point-of-care-device (POCD) consisting of a digital camera and a screen. Our solution relies on (1) recording videos of subject’s face (2) applying Eulerian video magnification to videos to reveal important subtle color changes in subject’s skin that fall outside human visual limits (3) obtaining spatio-temporal tensor expression profile of these variations (4) studying empirical spectral density (ESD) function of the largest eigenvalues of the tensors using random matrix theory (5) quantifying ESD functions by modeling the tails and decay rates using power law in systems exhibiting self-organized-criticality and (6) designing an optimal ensemble of learners to classify subjects into those with diabetic neuropathy and those of a control group. By analyzing a short video, we obtained a sensitivity of 100% in detecting subjects diagnosed with diabetic neuropathy. Our POCD paves the way towards the development of an inexpensive home-based solution for early detection of diabetic neuropathy and its associated complications.

Highlights

Rubeosis faciei diabeticorum, caused by microangiopathy and characterized by a chronic facial erythema, is associated with diabetic neuropathy
Diabetic neuropathy (DN), a type of nerve damage caused by long-term elevated glucose levels, is the most common complication of both type 1 and 2 diabetes and occurs in more than half of affected individuals
Our method relies on the fusion of information obtained from (1) applying Eulerian Video Magnification (EVM) to the videos, (2) studying empirical spectral density (ESD) function of the largest eigenvalues of our tensors obtained from EVM using theories of random matrices, (3) quantifying these ESD functions and modeling their tails in terms of Generalized Pareto distribution and power law exponents f (x) = Cx−γ in systems exhibiting SOC, and (4) designing an ensemble learning-based classifier to split the subspace corresponding to group DM from that of group C

Summary

Introduction

Rubeosis faciei diabeticorum, caused by microangiopathy and characterized by a chronic facial erythema, is associated with diabetic neuropathy. Facial erythema of patients with diabetes is evaluated based on subjective observations of visible redness, which often goes unnoticed leading to microangiopathic complications To address this major shortcoming, we designed a contactless, non-invasive diagnostic point-of-care-device (POCD) consisting of a digital camera and a screen. A substantially increased mortality among individuals diagnosed with diabetes peripheral neuropathy (DPN) who have undergone a major amputation, with 5-year mortality ranging from 44 to 68% has been observed. This calls for urgent action towards addressing this growing global health p roblem[3]. A recent novel non-invasive technique based on corneal confocal microscopy to quantify small fibre pathology in peripheral neuropathies and to provide in-vivo images of corneal nerve fibres was introduced b y10

Objectives

Methods

Results

Conclusion