基于二维图谱转换的心电信号分类方法

ECG signal classification method based on two⁃dimensionalatlas conversion

目的提出一种基于二维图谱转换的心电信号分类方法,以期为临床应用提供参考。方法首先通过格拉姆角场将一维心电信号转换成二维图像,其次对图像提取灰度共生矩阵和颜色矩特征,并运用支持向量机(support vector machines,SVM)对心律失常信号分类,最后使用MIT-BIH公共数据集对此分类器进行训练和测试。结果该方法分类的总准确率为99.9%,可实现对心律失常的有效分类。结论与传统波形形态分类算法相比,本文提出的将信号转换成二维图谱的分类方法有效解决了模型抗干扰能力差的问题,从而提升了分类器的准确率,具有潜在的临床应用可行性。

Objective To propose an ECG signal classification method based on two⁃dimensional atlas conversion,and to provide reference for clinical application.Methods Firstly,the one⁃dimensional ECG signal was converted into a two⁃dimensional image through the Gram angle field,and then the gray level co⁃occurrence matrix and color moment features were extracted from the image,and the arrhythmia signal was classified by support vector machines(SVM).Finally,this classifier was trained and tested using the MIT⁃BIH public dataset.Results The overall classification accuracy of the method was 99.9%,which could achieve effective classification of arrhythmias.Conclusions Compared with the traditional waveform shape classification algorithm,the classification method converting the signal to a two⁃dimensional map proposed in this paper effectively solves the problem of poor anti⁃interference ability of the model,thereby improving the accuracy of the classifier,and has potential clinical application feasibility.