摘要
心音能有效地反应心脏尤其是瓣膜活动状况,研究基于心音的心脏病决策系统具有重大意义。心音分段是建立心音决策系统的基础和前提,其目的是定位心音的主要成份,为特征提取与模式识别提供定位基准。本文通过使用双门限、迭代等方法,改进了基于信号能量的分段算法,并首次引入短时过零率以更准确地定位分段边界。实验结果表明,该算法对正常心音及常见异常心音分段效果良好。
Heart sound reflects the behavior of hearts, especially their valves, so it is of great significance to develop a support system. The segmentation of heart sound signal is the first step of analysis and the most important procedure in the automatic diagnosis of heart sounds. In this article, double threshold, iteration is applied to improve segmentation algorithm based on signal energy, and Short-time Zero-crossing Rate is proposed for the first time to determine the exact boundaries between different components of heart sound. Experimental results indicte that the algorithm is validated to perform well for normal and typical abnormal heart sounds.
出处
《北京生物医学工程》
2007年第1期48-51,共4页
Beijing Biomedical Engineering
关键词
心音分段算法
香农能量
短时过零率
heart sound segmentation
shannon energy
short-time zero-crossing rate ( ST - ZCR)