Deep Wavelets for Heart Sound Classification
Deep Wavelets for Heart Sound Classification
复制标题
DOI：
10.1109/ispacs48206.2019.8986277
复制
发表时间：
2019-12
期刊：
影响因子：
--
通讯作者：
Kun Qian;Zhao Ren;Fengquan Dong;Wen-Hsing Lai;Björn Schuller;Yoshiharu Yamamoto
中科院分区：
文献类型：
--
作者：
Kun Qian;Zhao Ren;Fengquan Dong;Wen-Hsing Lai;Björn Schuller;Yoshiharu Yamamoto
心血管疾病的发病率很高，仍然是导致死亡的主要原因。在过去的二十年里，智能听诊系统的开发吸引了信号处理和机器学习领域的巨大努力。提出了一种基于小波表示和深度递归神经网络的心音识别框架，用于识别正常、轻度和重度三种心音。使用深圳心音语料库(n=170)对该方法进行了验证。实验结果表明，在严格的学科无关场景下，该方法的非加权平均召回率可以达到43.0%(概率水平：33.3%)。
Cardiovascular diseases have a high morbidity, and remain the leading cause of mortality. In the past two decades, developing an intelligent auscultation system has attracted tremendous efforts from the field of signal processing and machine learning. We propose a novel framework based on wavelet representations and deep recurrent neural networks for recognising three heart sounds, i. e., normal, mild, and severe. The Heart Sounds Shenzhen corpus (n = 170) is used to validate the proposed method. The experimental results demonstrate the efficacy of the proposed method in a rigorous subject independent scenario, which can reach an unweighted average recall at 43.0 % (chance level: 33.3%).