Type classification of heart disease

Heart disease has long been one of the causes of death worldwide, and we believe it is extremely important to extract physiological and pathological information about the heart at an early stage to determine whether specialist diagnosis is necessary. You can Heart sound classification is one of the common methods to extract cardiac information and is widely used for automatic diagnosis of heart disease. Fig.1 shows the overall image of the heart sound classification system (Convolutional Neural Network: CNN-based)*1. Heart sound classification systems generally consist of three steps: (1) segmentation, (2) feature extraction, and (3) class classification. Fig.2 is an example (second stage) when wavelet transform is used for feature extraction.

​

*1 Zhang et al., Towards Heart Sound Classification Without Segmentation Using Convolutional Neural Network, 2017