A Gestation Diabetic Detection Technique using Muscle Energy Derived from Surface EMG

Abstract

Electromyogram (EMG) is one among the important biopotential signal reflecting the human skeletal muscle activity. EMG signals can be used for many biomedical applications pertaining to diagnosis and therapy of musculoskeletal and rheumatological problems. EMG signals are complex in nature and require advanced techniques for analysis, such as decomposition, detection, processing, and classification. Diabetes mellitus is a chronic metabolic disorder characterized by elevated levels of blood glucose. The musculoskeletal system can be affected by diabetes in a number of ways. The main aim of the paper is to identify the diabetic patient and show the classification performance of the proposed framework. In this paper EMG signal is investigated by feature extraction and are classified into normal and diabetic for comprehension of EMG signal. The primary point of this work is to recognize the diabetes utilizing different elements and to demonstrate the performance of the proposed framework. The obtained results demonstrate that the extracted feature in proposed framework displays better performance for classification the EMG signal contrasted with alternate elements. Based on the impacts of features on the EMG signal classification, different results were obtained through analysis of the SVM Classification. Experimental study shows that the proposed method's classification accuracy is 98.98%.