An Ensemble Machine Learning Technique for Detection of Abnormalities in Knee Movement Sustainability

The purpose of this study was to determine electromyographically if there are significant differences in the movement associated with knee muscle, gait, leg extension from a sitting position and flexion upwards for regular abnormal sEMG data. Surface electromyography (sEMG) data were obtained lower limbs 22 people during three different exercises: sitting, standing, walking (11 11 without abnormality). Participants deformity took longer finish task than healthy subjects. signal duration patients abnormalities that patients, resulting an imbalance As result data’s bias towards majority class, developing classification model automated analysis such signals is arduous. collected denoised filtered, followed by extraction time-domain characteristics. Machine learning methods then used predicting distinct movements (sitting, walking) electrical impulses normal sets. Different anomaly detection techniques also detecting occurrences differed considerably hence enhancing performance our model. iforest technique presented work can achieve 98.5% accuracy on light gradient boosting machine algorithm, surpassing previous results which claimed maximum 92.5% 91%, improving 6–7% abnormality using learning.