Comparative study of morphological ECG features classificators: An application on athletes undergone to acute physical stress

摘要: Several methods for automatic heartbeat classification have been developed, but few efforts devoted to the recognition of small ECG changes occurring in healthy people as a response stimuli. Herein, we describe procedure extraction, selection and features summarizing morphological changes. The proposed is composed by following stages: 1) extraction set features; 2) subset 3) subject normalization 4) classification. enabled us summarize with only three non redundant features. In addition performed comparison between four classificators: k-nearest-neighbors (k-NN), artificial neural networks (ANN), support vector machines (SVM) naive Bayes classifiers (nB). order cope possible linear separation problem, evaluated two strategies: factor on feature space usage kernel functions classifiers. results recommended normalization, irrespectively from classificator: or without had best performance linear-SVM ANN.