Development of Preventing Myofascial Pain Syndrome Automation with Ultrasonic-based and Machine Learning

Abstract

Physical readiness is one factor that promotes student learning achievement. However, sitting for long periods can lead to myofascial pain syndrome, affecting undergraduate students’ learning outcomes who took lecturebased and computer-based sessions online for long periods. This study aims to develop a set of economic reminders using the Internet of Things and prediction modeling by the Machine Learning technique. The developed preventing myofascial pain syndrome automation system reminds the student to change sitting postures or get up and prevent myofascial pain syndrome. This system applies consecutively to students for two academic semesters. Further, this system applied the prediction models by four Machine Learning techniques. The evaluation results of model efficiency revealed that the model developed with Multi-Layer Perceptron Neural Network has the highest accuracy of 93.98%. The model with the second highest accuracy performance was the Support Vector Machine, k-Nearest Neighbor, and Decision Tree techniques were modeled with accuracy values of 91.77%, 91.31%, and 90.56%, respectively. Furthermore, the results showed that the preventing myofascial pain syndrome automation system promoted higher student learning outcomes than the group without the preventing myofascial pain syndrome automation system at a significance level of 0.05. The developed system with the prediction model also effectively prevents and reduces the number of students from myofascial pains. Thus, the developed system has shown that educational management focusing on the learners’ health will enhance learning effectiveness.