摘要
目的:提出一种新的基于波形特征和SVM的心电信号自动分类实现方法。方法:定义并提取了基于时域特征、小波域特征和高阶统计量特征等三大类心电特征参数,将一次性直接求解多类模式的SVM方法应用于心电信号分类。结果:通过对心电数据库典型心律失常信号的分类测试,验证了所提出心电信号分类方法的有效性。结论:本方法的实现可以有效提高了分类识别精度和速度。
Objective： This paper put forward for classifying cardiac arrhythmia signals based on ECG wavcform features and support vector machine （SVM）. Methods： The time-domain feather,wavelet transform domain characteristics and higher-order cumulants of ECG signals arc defined and extracted as three types of ECG features .In this paper the SVM multi-class classifi- cation method is used in ECG for the first time. Results： The classification method is validated with the cardiac arrhythmia signals obtained from thc ECG databasc. Conclusions： SVM based ECG classification is also carried out respectively with thrcc different types of feature set.this method improvc the efficiency and prccision of classificr.
出处
《中国医学物理学杂志》
CSCD
2010年第4期2043-2046,共4页
Chinese Journal of Medical Physics
基金
山东省高等学校科技计划项目(项目编号:J09LG25)
关键词
波形特征
支持向量机(SVM)
自动分类
wavcform feature
support vector machine（SVM）
automatic classifying
