A Comparative Study on the Accuracy of EEG-Based and sEMG Based Fatigue Level Measurement

Abstract

Manual handling relates to transporting of an item from one place to another by means of either carrying, pushing, or pulling. Muscle fatigue results from a muscle constantly generating or applying force by doing manual labor or vigorous exercise. In line with this, measuring the fatigue level of a person is of great importance. This paper presents a comparative study between two systems. The first system uses electroencephalogram (EEG) while the second system uses surface electromyography (sEMG) in measuring muscle fatigue. A fatigue estimation methodology based on EEG signal processing is proposed for the first system in this paper to measure the muscle fatigue. Another fatigue estimation methodology based on sEMG signal processing is proposed for the second system to measure the muscle fatigue. In both cases, Artificial Neural Network (ANN) was used as the learning algorithm. The study volunteers were composed of 20 male devoid of gym workout for at least 2 years and had a complete rest before conducting the test. The volunteers were asked to perform elbow flexion using their dominant arm while lifting a 5 lb. dumbbell for 90 seconds. The accuracy of the two systems was compared and the results using sEMG in measuring muscle fatigue of a person is more accurate, having a testing accuracy of 94.6%, than using EEG having only 92.9% testing accuracy. It is believed that this paper will be beneficial for researchers in the field of bioengineering. This paper will help ease the difficulty in muscle fatigue detection through the use of Deep Learning algorithms.