Classification of muscle fatigue during prolonged driving

• Main Authors: Ab Ghafar, Noor Azlyn, Azmi, Nur Liyana, Md Nor, Khairul Affendy, Nordin, Nor Hidayati Diyana
• Format: Article
• Language: English
• Published: Penerbit UTM Press, Universiti Teknologi Malaysia 2022
• Subjects: TA164 Bioengineering
• Online Access: http://irep.iium.edu.my/104250/2/104250_Classification%20of%20muscle%20fatigue.pdf; http://irep.iium.edu.my/104250/; https://elektrika.utm.my/index.php/ELEKTRIKA_Journal/article/view/376; https://doi.org/10.11113/elektrika.v21n3.376

Driving has become essential in transporting people from one place to another. However, prolonged driving could cause muscle fatigue, leading to drowsiness and microsleep. Electromyography (EMG) is an important type of electropsychological signal that is used to measure electrical activity in muscles. The current study attempted to use machine learning algorithms to classify EMG signals recorded from the trapezius muscle of 10 healthy subjects in non-fatigue and fatigue conditions. The EMG signals and the time when muscle fatigue was experienced by the subjects were recorded. The mean frequency (MNF) and median frequency (MDF) of the EMG signals were extracted as dataset features. Six machine learning models were used for the classification: Logistic Regression, Support Vector Machine, Naïve Bayes, k-nearest Neighbour, Decision Tree and Random Forest. The results show that both the MNF and MDF are lower when fatigue conditions exist. In term of the classification performance, the Random Forest, Decision Tree and k-nearest Neighbour classifiers produced the accuracy levels of 85.00%, 83.75% and 81.25% respectively. Thus, the highest accuracy for classifying muscle fatigue due to prolonged driving was obtained by the Random Forest classifier, using both the MNF and MDF of the EMG signals. This method of using the MNF and MDF will be useful in classifying driver’s non-fatigue and fatigue conditions during prolonged driving.