EMG based Muscle Fatigue and Muscle Growth Detection

Abstract

The study presents a novel approach for long term muscle development analysis as well as short term muscle fatigue analysis using surface EMG signals. Muscle based weight training is an exercise driven routine that is meant to grow muscle and improve its functioning. These exercises are used by athletes, and by individuals who suffer from motor deficiencies, so that their ability to perform motor activities is increased. Rehabilitation of individuals with motor deficiencies can be especially difficult since users often suffer from poor muscle control and are more susceptible to fatigue. Thus, muscle fatigue is a central consequence that affects the muscle recovery period during rehabilitation and can be an important metric to determine the progress of an individual. It is desired then that a metric can be used to directly quantify fatigue during exercise, and how it can change over time. With a long term goal to solve the above mentioned challenges by devising a wearable and portable device, this thesis presents a proof of concept study that analyzes and classifies short term muscle fatigue and long term muscle growth. In this study, it has been found that, with fatigue there is an increase in amplitude and an increase in power of the EMG signal. In addition, with muscle growth, the power of the EMG signal is found to decrease over time, specifically in this study power decreased by 9.67 dB over a period of four months during repetitive training. This project is a step toward the development of a wearable exercise optimization device for post-stroke recovery and athletes.