一种基于模板匹配的复杂心音定位方法  

Location of complex heart sound based on template matching

作　　者：苗晟[1] 王威廉[1] 姚绍文[1] 

机构地区：[1]云南大学信息学院,昆明650031

出　　处：《电子测量与仪器学报》2015年第1期119-123,共5页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金(612610008);青年基金(11303094)资助项目

摘　　要：对心音的定位分割是心音处理的重要步骤之一。对于先心病患者的心音,由于受到严重的心脏杂音干扰,很难通过常规的包络或能量提取方法定位分割。根据心音特征提出一种模板匹配算法以对复杂的病理心音定位分割。该方法首先通过相关度检测和STFT变换生成第一、第二心音模板,采用该模板对模板化后的病理心音进行匹配,最后对匹配结果通过矫正算法分割定位。通过对标准心音数据库中68例实测数据和自采的57例数据分析,对于正常心音,该算法分割定位准确度达90%以上,对于先心病患者心音达到约78%,有效提高了定位准确度。Location and segmentation for heart sound is the important step for heart sound （HS） analysis. As for some ASD patient, their HS is hard to segment using envelop and energy extraction due to the influence of systolic murmur. A new complex patient HS location algorithm based on template matching is introduced. First of all, tem- plate of S1, S2 and HS are constructed through correlation and STFT method. Then, those templates are used to match the HS template. At last, the result of location is corrected using correlation algorithm. 68 cases from HS da- tabase and 57 cases from clinical data are analyzed. The results demonstrate that the effectiveness of our constructed algorithm for the positioning accuracy is 93% for normal and 81% for patient.

关 键 词：心音分割 模板匹配 杂音提取 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TN911.72[自动化与计算机技术—控制科学与工程]