肺音信号非高斯ARMA模型及双谱研究  

Study on Non-Gaussian ARMA Model and Bispectrum of Lung Sounds

作　　者：赵守国[1] 王素品[1] 邹原[1] 牛海军[1] 万明习[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049

出　　处：《西安交通大学学报》2001年第4期355-359,共5页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目

摘　　要：应用高阶累积量方法 ,从肺音信号辨识肺胸系统的ARMA模型参数 .分别在肺胸系统正常和异常的情况下 ,对肺泡呼吸音、哮呜音和喘呜音进行了分析 .实验结果表明 ,高阶累积量能够很好地抑制高斯噪声 ,并能提取由于肺部机能异常引起的肺胸系统声传递函数的变化 ,对肺胸系统自动分类。The ARMA model parameters of lung-thorax system from lung sound were estimated by the higher-order cumulants. Vesicular breath sounds, wheeze and stridors from the normal lung-thorax system mechanism and abnormal lung-thorax system mechanism were analyzed. Experimental results show that higher-order cummulants can reduce Gaussian noise and reflect the variation of the sound transfer function derived from abnormal lung-thorax system mechanism. It is important for automatic diagnosis of lung-thorax disease.

关 键 词：高阶累积量 非高斯ARMA模型 双谱 肺音信号 肺胸系统 诊断 肺泡呼吸音 

分 类 号：R318.04[医药卫生—生物医学工程]