心电图(Electrocardiogram,ECG)是诊断心血管疾病的重要依据。本文提出一种基于模糊最小二乘支持向量机(FLSSVM)的ECG分类诊断方法,并利用麻省理工学院(MIT-BIH)的心电图数据库中的数据进行训练和测试,通过在每个训练样本点中加入模糊隶属度,训练得到分类模型,仿真结果表明,FLSSVM分类器相比标准LSSVM和SVM,在分类正确率,分类速度以及适用的样本规模上都表现出了十足的优越性,FLSSVM分类器能够有效地处理实际分类问题。 Electrocardiogram (ECG) is an important basis for the diagnosis of cardiovascular diseases. In this paper, an ECG classification diagnosis method based on fuzzy least square support vector machine (FLSSVM) is proposed and trained and tested by using data from MIT-BIH ECG database. The fuzzy membership is added and the classification model is obtained by training. The simulation results show that FLSSVM classifier shows superiority in classification accuracy, classification speed and sample size compared with standard LSSVM and SVM. The FLSSVM classifier Can effectively deal with the actual classification problem.