Summary of work. Classification of EMG signals using CSVM for the diagnosis of muscle fatigue Josue Martinez Hernandez IPD410 - Metodos Matematicos en control Automatico Universidad Tecnica Federico Santa Maria I Semestre 2020 Mayo 15-2020 In this work we will the use of the support vector machine classifier (GSVM), applying a Gaussian kernel, to classify EMG signals and to diagnose muscle fatigue. The original article [1] focuses on the use of a BFA battery algorithm to optimize the GSVM. For this, an isometric experiment is performed with healthy patients in which EMG signals are extracted from the arm, in order to determine muscle fatigue, carrying out different series. They then extract features using the EEMD technique, which uses the Hilbert transform to decompose the signal and then classify it. For the purpose of this course, the focus will be on the subject of a support vector machine, where the geometric concept of the hyperplane will be studied in depth, which is defined as a flat and related subspace of dimensions p, and on the use of a Gaussian kernel. The SVM generates the hyperplane to be able to separate the classes and then apply a kernel depending on the behavior of the data. The reference [2] will be used to investigate the mathematical concept of the SVM It is expected that the basic mathematical concept of SVM will be investigated to be used as a tool for my Thesis since I intend to classify EMG signals from the larynx muscles and determine the level of activation in phonation. Such information is expected to be used in the mathematical models of the vocal cords of my tutor Matias Zañartu. Referencias [1] Wu, Qi, Chen Xi, Lu Ding, Chuanfeng Wei, He Ren, Rob Law, Honghui Dong y Xiao Li Li: Classification of EMG Signals by BFA-Optimized GSVCM for Diagnosis of Fatigue Status. IEEE Transactions on Automation Science and Engineering, 14(2):915–930, Abril 2017, ISSN 1558-3783. Conference Name: IEEE Transactions on Automation Science and Engineering. [2] Hastie, Trevor, Robert Tibshirani y Jerome Friedman: The Elements of Statistical Learning – Data Mining, Inference, and Prediction. 1