基于主成分分析的心音特征降维处理研究
Principal Component Analysis-Based Heart Sound Features Reduction Research
DOI: 10.12677/JISP.2018.74024, PDF,   
作者: 张俊杰, 张弼强, 王丁彬, 胡飞燕：南阳理工学院电子与电气工程学院，河南 南阳
关键词: 主成分分析；小波分解；功率谱；Principal Component Analysis (PCA)； Wavelet Decomposition； Power Spectrum Density
摘要: 本研究提出一种基于主成分分析的心音特征降维处理方法，实现以最优维度描述不同种类心脏病的心音特征分布。本文分三阶段进行论述：第一阶段，基于美国3M公司3200型电子听诊器的心音信号采集及基于小波变换的心音预处理；第二阶段，利用功率谱对心音信号进行频域分析，采用阈值法定义心音频域特征；第三阶段，基于主成分选取准则并结合散点图分布结果，确定以表征七维特征96.1%信息量的两维特征作为最终心音量。典型心脏病例的研究结果表明，异类心脏病心音特征分布呈现出明显区分。
Abstract: In this research, a method based on principal component analysis (PCA) is proposed for using the optimal dimensional features to describe the distribution of heart sound characteristics in different kinds of heart diseases. This study is described in three stages. In stage 1, heart sound signal is collected via 3M-3200 electronic stethoscope and preprocessed based on wavelet transform. In stage 2, the power spectrum density combined with threshold method is proposed to define the cardiac sound 7-dimensional feature. In stage 3, based on principal component selection criteria combined with scatter plot distribution results, the final heart sound is determined to be a 2-dimensional feature representing 96.1% information of 7-dimensional feature. The results of the research on the typical heart diseases indicate that there are obviously differences in the dis-tribution of the heart sound features among different kinds of heart diseases.
文章引用：张俊杰, 张弼强, 王丁彬, 胡飞燕. 基于主成分分析的心音特征降维处理研究[J]. 图像与信号处理, 2018, 7(4): 213-219. https://doi.org/10.12677/JISP.2018.74024

参考文献

[1] 《中国心血管病报告2017》概要[J]. 中国循环杂志, 2018(1): 1-8.
[2] 罗玉光, 心脏听诊的基础知识——怎样听心音[J]. 人民军医, 1964(7): 41-44.
[3] https://www.littmann.com
[4] 陈新华, 成谢锋. 一种改进型综合去噪算法在心音信号预处理上的研究[J]. 南京邮电大学学报(自然科学版), 2010, 30(6): 96-100.
[5] 单正娅. 基于人工神经网络及小波分析的心音诊断系统的研究[D]: [硕士学位论文]. 无锡: 江苏大学, 2006.
[6] Sun, S.P., et al. (2014) Segmentation-Based Heart Sound Feature Extraction Combined with Classifier Models for a VSD Diagnosis System. Expert Systems with Applications, 41, 1769-1780.
[Google Scholar] [CrossRef] 
[7] 刘立汉, 王海滨, 王燕, 陶婷, 魏秀波. 基于改进的希尔伯特–黄变换的心音信号特征分析[J]. 西华大学学报(自然科学版), 2010, 29(6): 14-18.
[8] 胡玉良, 王海滨, 陈健, 等. 心音在时频两域中解析方法的研究[J]. 西华大学学报(自然科学版), 2009(5): 5-8, 26.
[9] Joy, R., Acharya, U.R., Mandana, K.M., Ray, A.K. and Chakraborty, C. (2012) Expert Systems with Applications Application of Principal Component Analysis to ECG Signals for Automated Diagnosis of Cardiac Health. Expert Systems with Applications, 39, 11792-11800.
[Google Scholar] [CrossRef] 
[10] Giri, D., Acharya, U.R., Martis, R.J., Sree, S.V., Lim, T.-C., Ahamed, T. and Suri, J.S. (2013) Automated Diagnosis of Coronary Artery Disease Affected Patients Using LDA, PCA, ICA and Discrete Wavelet Transform. Knowledge-Based Systems, 37, 274-282.
[Google Scholar] [CrossRef] 
[11] Lee, J. and Jun, C.-H. (2013) PCA-Based High-Dimensional Noisy Data Clustering via Control of Decision Errors. Knowledge-Based Systems, 37, 338-345.
[Google Scholar] [CrossRef] 
[12] Shilaskar, S. and Ghatol, A. (2013) Feature Selection for Medical Diagnosis: Evaluation for Cardiovascular Diseases. Expert Systems with Applications, 40, 4146-4153.
[Google Scholar] [CrossRef] 
[13] Zhu, B., Ding, Y. and Hao, K. (2014) Multiclass Maximum Margin Clustering via Immune Evolutionary Algorithm for Automatic Diagnosis of Electrocardiogram Arrhythmias. Applied Mathematics and Compu-tation, 227, 428-436.
[Google Scholar] [CrossRef]