A Novel Feature Extraction Method for Heart Sounds Classification
Abstract
The experimental results showed that the proposed method efficiently classifies heart sounds. Heart sound analysis is a basic method for heart examination, which may suggest the presence of a cardiac pathology and also provide diagnostic information. In this study, a novel feature extraction method based on Independent Component Analysis is applied to classify nine different heart sound categories. The extracted features are subjected to classification by Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) using 5 Cross Validation and by Artificial Neural Network. The experimental results showed that the proposed method efficiently classifies heart sounds.