Multi-class Support Vector Machine (SVM) Classifiers -- An Application in Hypothyroid Detection and Classification
In this paper, a new approach is presented for the analysis and the identification of the surface electromyography (EMG) signals of biceps and triceps muscles. The objective of this study is the accurate classification of elbow flexion and extension movements. We propose a cropping method based on the agreement of the movement changes and the EMG signal using the upper limb kinematic. Then, we perform the extraction and selection of several well known features in time and frequency domain. The selected features are used as inputs for our support vector machine classifier that is designed using an optimal weight vector criterion. Afterward, the training and test steps are performed in the proposed scheme. Finally, numerical simulation assesses the accuracy of the classification, as well as the robustness of the proposed approach considering noisy measurements.