TY - JOUR ID - TI - Pattern Recognition of Composite Motions based on EMG Signal via Machine Learning AU - Thamir R. Saeed AU - Sameer K. Salihc, AU - Mahmuod H. Al-Muifrajeb, AU - Noof T. Mahmooda*, PY - 2021 VL - 39 IS - 2 part (A) Engineering SP - 295 EP - 305 JO - Engineering and Technology Journal مجلة الهندسة والتكنولوجيا SN - 16816900 24120758 AB - In the past few years, physical therapy plays a crucial role duringrehabilitation. Numerous efforts are made to demonstrate the effectivenessof medical/ clinical and human-machine interface (HMI) applications.One of the most common control methods is using electromyography(EMG) signals generated by muscle contractions to implement theprosthetic human body parts. This paper presents an EMG signalclassification system based on the EMG signal. The data is collected frombiceps and triceps muscles for six different motions, i.e., bowing, clapping,handshaking, hugging, jumping, and running using a Myo armband witheight electromyography sensors. The Root Mean Square, DifferenceAbsolute Standard Deviation Value, and Principle Component Analysisare used to extract the raw signal data and enhance classificationaccuracy. The machine learning method is applied, i.e., Support VectorMachine and K-Nearest Neighbors are used for classification; the resultsshow that the K-Nearest Neighbors method achieves a higher accuracypercentage than the SVM. Making high training accuracy for differentphysical actions helps implement human prosthetic parts to help thepeople who suffer from an amputee

ER -