Abstract

Info

Title:: Heart sound denoising algorithm based on minimum mean-square error log-spectral amplitude estimation

Author(s):: FENG Shuai1; LIU Feifei2; WU Xinyu2; 3; ZHANG Jianwen1; LIU Ziyou4; LI Jiahao5; 1. School of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China 2. School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China 3. Shanghai PULSATE Technology Co., Ltd, Shanghai 200000, China 4. Department of Cardiovascular Surgery, First Affiliated Hospital of Gannan Medical University, Ganzhou 341000, China 5. Department of Automotive Engineering, Ganzhou Polytechnic, Ganzhou 341000, China

Keywords:: Keywords: heart sound denoising minimum mean-square error log-spectral amplitude heart sound classification electronic stethoscope

Abstract:: Abstract: The study is the first to apply a speech enhancement algorithm based on minimum mean-square error log-spectral amplitude estimation in heart sound denoising. Based on statistical modeling, the denoising is incorporated into a statistical framework. Through the optimal amplitude spectrum estimation on noisy heart sound signals, the difference between estimated and clean heart sounds is minimized to remove noise and restore the original signals. The experimental results show that the proposed algorithm can effectively improve the time-frequency characteristics of heart sound signals and achieve higher accuracy in subsequent heart sound classification. The excellent performance of the proposed method in heart sound denoising is of great significance for the development of electronic stethoscopes and automatic heart disease detection technology.