Heart Sound Classification Network Based on Convolution and Transformer
Open Access
Abstract
Electronic auscultation is vital for doctors to detect symptoms and signs of cardiovascular diseases (CVDs), significantly impacting human health. Although progress has been made in heart sound classification, most existing methods require precise segmentation and feature extraction of heart sound signals before classification. To address this, we introduce an innovative approach for heart sound classification. Our method, named Convolution and Transformer Encoder Neural Network (CTENN), simplifies preprocessing, automatically extracting features using a combination of a one-dimensional convolution (1D-Conv) module and a Transformer encoder. Experimental results showcase the superiority of our proposed method in both binary and multi-class tasks, achieving remarkable accuracies of 96.4%, 99.7%, and 95.7% across three distinct datasets compared with that of similar approaches. This advancement holds promise for enhancing CVD diagnosis and treatment.Keywords
Funding Information
- 2022 Jiangsu Provincial Postgraduate Research and Innovation Program (KYCX22_0923, 210023)
- Nation-Local Joint Project Engineering Lab of RF Integration & Micropackage, Nanjing (KYCX22_0923, 210023)
This publication has 31 references indexed in Scilit:
- Classification of Heart Sound Signal Using Multiple FeaturesApplied Sciences, 2018
- Recognition of normal–abnormal phonocardiographic signals using deep convolutional neural networks and mel-frequency spectral coefficientsPhysiological Measurement, 2017
- An open access database for the evaluation of heart sound algorithmsPhysiological Measurement, 2016
- Classifying Heart Sound Recordings using Deep Convolutional Neural Networks and Mel:Frequency Cepstral CoefficientsComputing in cardiology, 2016
- Normal / Abnormal Heart Sound Recordings Classification Using Convolutional Neural NetworkComputing in cardiology, 2016
- Deep learningNature, 2015
- A new method of heart sound signal analysis based on independent function elementAIP Advances, 2014
- Selection of Dynamic Features Based on Time–Frequency Representations for Heart Murmur Detection from Phonocardiographic SignalsAnnals of Biomedical Engineering, 2009
- Support Vectors Machine-based identification of heart valve diseases using heart soundsComputer Methods and Programs in Biomedicine, 2009
- Long Short-Term MemoryNeural Computation, 1997