Heart sound classification based on log Mel-frequency spectral coefficients features and convolutional neural networks. (August 2021)

• Record Type: Journal Article
• Authors: Kui, Haoran; Pan, Jiahua; Zong, Rong; Yang, Hongbo; Wang, Weilian
• Is Part Of: Biomedical signal processing and control. Volume 69(2021)
• Publication Date: 2021-08
• Subjects: Heart sounds classification -- MFSC features based on dynamic frame length -- Convolution neural network -- Heart sound segmentation -- Gaussian mixture distribution; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.102893
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The MFSC feature based on dynamic frame length was proposed. The modeling of the heart sound state by the Gaussian mixture model improves the segmentation accuracy. The sample size was expanded by dividing the heart sound into the cardiac cycle. The performance of the classifier was improved by the majority voting algorithm. The MFSC features have achieved excellent performance in two-classification and multi-classification problems. Abstract: In view of the important role of heart sound signals in diagnosing and preventing congenital heart disease, a novel method about feature extraction and classification of heart sound signals was put forward in this study. Firstly, the heart sound signals were de-noised by using the wavelet algorithm. Subsequently, the improved duration-dependent hidden Markov model (DHMM) was used to segment the heart sound signal according to the heart cycle. Then, the dynamic frame length method was used to extract log Mel-frequency spectral coefficients (MFSC) features from the heart sound signal based on the heart cycle. Afterward, the convolution neural network (CNN) was used to classify the MFSC features. Finally, the majority voting algorithm was used to get the optimal classification results. In this paper, two-classification and multi-classification models were built. An accuracy of 93.89% for two-classification and an accuracy of 86.25% for multi-classification were achieved using the novel method.