CAJ | 학술논문

运用计算机技术辅助激光治疗心律失常疾病，通过非线性动力学分析心电信号，采用模式识别工具实现心律失常疾病的初步诊断。首先提出一种基于经验模式分解（EMD）的时间尺度滤波法，从而有效的抑制心电信号噪声。然后提取预处理后数据的排列熵（Permutatior Entropy，PE）特征值。最后利用支持向量机（SVM）分类器验证特征值的分类。实验结果表明该算法能有效的实现正常和异常心电信号的自动分类，从而简便快速的初步诊断心律失常疾病。
운용계산궤기술보조격광치료심률실상질병，통과비선성동역학분석심전신호，채용모식식별공구실현심률실상질병적초보진단。수선제출일충기우경험모식분해（EMD）적시간척도려파법，종이유효적억제심전신호조성。연후제취예처리후수거적배렬적（Permutatior Entropy，PE）특정치。최후이용지지향량궤（SVM）분류기험증특정치적분류。실험결과표명해산법능유효적실현정상화이상심전신호적자동분류，종이간편쾌속적초보진단심률실상질병。
In the paper,in order to complete computer- aided technology in laser therapy for cardiac arrhythmia the Nonlinear dynamics was used to analyze the ECG signals and Pattern recognition was used to realize the preliminary diagnosis of the cardiac an-hythmia. Firstly, in order to effectively restrain noise of the signals, a filtering of time scales was implied, then the permutatior entropy of ECG signals can be obtained. Finally, support vector machine {SVM） is adopted to achieve the optimal classification. Experimental results prove the reliability and veracity of the method.