Machine Learning-Based Intelligent Smart Embedded Sensors for Automatic Detection and Classification of Neuromuscular Disorders using EMG Signals

Abstract

Abstract The objective of this work is to create a novel computer-aided health monitoring system for diagnosing neuromuscular disorders (NMDs). Additionally, we will propose the use of embedded sensor networks to facilitate proactive patient care and remote health monitoring. The proposed method combines the discrete wavelet transform (DWT) with two supervised machine learning algorithms: the multi-class support vector machine (SVM) and the k-nearest neighbors (k-NN) classifiers. The dataset includes ten normal subjects, aged between 21 and 37 years. Out of these subjects, six are males and four are females. The results were presented on a graphical user interface (GUI) based on LabVIEW and implemented using a real embedded CompactRIO-9035 real-time controller. Additionally, the proposed embedded system has the capability to serve as a portable diagnostic device for the automatic detection of NMDs.