Diabetic Sensorimotor Polyneuropathy Severity Classification Using Adaptive Neuro Fuzzy Inference System

Diabetic sensorimotor polyneuropathy (DSPN) is an early indicator for non-healing diabetic wounds and foot ulcers, which account one of the most common complications diabetes, leading to increased healthcare cost, decreased quality life, infections, amputations, death. Early detection intelligent classification tools DSPN can allow correct diagnosis treatment painful neuropathy as well a timely intervention prevent ulceration, amputation, other complications. Hence, successfully mitigate prevalence DSPN, this study aims depict severity classifier using Adaptive Neuro Fuzzy Inference System (ANFIS). Michigan Neuropathy Screening Instrumentation (MNSI) was considered input identification stratification DSPN. Patients have been classified into four classes: Absent, Mild, Moderate, Severe. The model accuracy validated with results from different machine learning algorithms. Accuracy, sensitivity, specificity ANFIS are 91.17±1.18%, 92±2.26%, 96.72±0.93%, respectively. proposed used classify Epidemiology Diabetes Interventions Complications (EDIC) clinical trial patients observed that in first, eighth, nineteenth EDIC years 18.31%, 39.45%, 59.14% had levels This also investigates changes muscle activity during gait three lower limb muscles (vastus lateralis (VL), tibialis anterior (TA), gastrocnemius medialis (GM)) electromyography (EMG) by VL GM show increase delay activation peak decrease magnitude progression severity. Based on observation, trained extracted EMG features showed promising results. Our based both MNSI variables will help health professionals diagnose stratify signs symptoms electrophysiological due