Executive Summary : | Fatigue is defined as a typical physiological reaction to effort, lack of sleep, changes in sleep-wake patterns, stress, as well as tiredness, exhaustion, or lethargy, all of which are associated with a loss of energy. Overtime work, rapid changes in the working environment, employment of middle-aged employees, issues with working hours for part-timers, and changes in job were also discovered to be reasons. Furthermore, its repercussions have been observed to include accidents and health issues of various dimensions. Muscle fatigue is described as a failure to maintain the anticipated force, which is followed by changes in muscle electrical activity. Despite extensive research, the causes of EMG alterations in the time and frequency domain have remained a mystery until now. Many theories anticipated a linear relationship between characteristic frequencies and muscle fibre propagation velocity (MFPV), despite the reality that spectral features might decline even when MFPV does not change, or in proportion to MFPV changes. Since evidence on increased, almost unchanged, and reduced amplitude features of the EMG, M-wave, or motor unit potential (MUP) with tiredness can be found in the literature, the amplitude changes appear to be more complicated and conflicting. Furthermore, the simultaneous fall and rise in amplitude of MUP and M-wave observed with indwelling and surface electrodes were labelled as paradoxical. Despite this, EMG amplitude parameters are commonly employed for analysing fatigue causes. The ECG is generated by a weak current during the cardiac excitation process. Whenever a person is fatigued, the ECG signal's regularity decreases significantly. The heart rate (HR) index and heart rate variability (HRV) index are crucial indicators of weariness. Overwork-related problems, such as cerebrovascular/cardiovascular diseases (CCVD) and mental disorders caused by overwork, are a serious occupational and public health concern across the world, particularly in East Asian nations. We were interested in investigating the potential of implementing wearable smart electrocardiogram (ECG) sensors to detect the physical and mental exhaustion state because they are affordable, practical, popular, and widely accessible today. Fatigue among industrial workers can be detected by using EMG and ECG and various machine learning classifier like SVM, decision tree, logistic regression, K-means, Naïve bayes and neural network techniques including RNN, CNN, LSTM, RBFNN, and MPLNN. The noncontact data acquisition system was used to collect physiological signals from the biceps femoris (for EMG) and chest region (for ECG) of each subject. Fast independent component analysis (Fast ICA) and digital filter were utilized to process the original signals. Time, frequency and entropy features are extracted from the processed signal and classified. The noncontact, on-board industrial worker fatigue detection system was developed to reduce fatigue-related risks. |