Application of Artifical Neural Network and Logistic Regression Analysis in Classifying Intramuscular Electromyography

Authors: Phạm Mạnh Hùng*, Vũ Duy Hải, Nguyễn Văn Khang


Quantitative electromyography (QEMG) parameters could be a used for classification of muscular disease into neuropathy or myopathy. The paper describes a new classification method based on the Artifical Neural Network (ANN) and the logistic regression analysis using some of our QEMG parameters. By using the EMG signals in EMGlab database , we obtained 90% accuracy for classifying the EMG signals in two classes (neuropathy or other, and myopathy or other). Similarly, we obtained > 80% accuracy for classifying the EMG signals in three classes (neuropathy or myopathy or normal). These methods have higher accuracy than other researched methods for classification, and could offer advantages if applied in clinical practice


Classification of EMG signals, Quantitative electromyography, Electromyography
Pages : 69-74

Related Articles:

Authors : Phan Thị Lệ Thuyền*, Võ Trung Hùng
Authors : Trương Ngọc Minh*, Lã Minh Khánh, Nguyễn Hoàng Việt