论文部分内容阅读
识别和描述房颤有可能自发终止还是持续,不仅可以更好地了解心律不齐发作和终止的机制,还可以更有效地治疗持续房颤。本文从非线性角度提取房颤信号特征并预测其能否自发终止。先重建心电信号的相空间,获取相空间中指定庞加莱截面上的点,然后基于形状分析提取刻画庞加莱面上相似点对相对数量的参数ρ,最后基于ρ来分类非终止房颤和可终止房颤。实验研究了一个由Holter心电信号组成的包括训练集和测试集的房颤数据库,结果表明:本文提出的特征参数ρ可正确分类90%的测试集。可见,该方法能从Holter心电信号有效地预测房颤的自发终止。
Identifying and describing the possible spontaneous termination or persistence of atrial fibrillation will not only provide a better understanding of the mechanism of arrhythmia episodes and discontinuation but may also be useful in treating persistent atrial fibrillation. This article extracts the characteristics of atrial fibrillation signals from a non-linear perspective and predicts whether they can spontaneously terminate. Firstly, the phase space of the ECG signal is reconstructed, and the points on the Poincaré section specified in the phase space are obtained. Based on the shape analysis, the parameter ρ which characterizes the relative number of similar points on the Poincaré is extracted. Finally, Flutter and termination of atrial fibrillation. Atrial fibrillation data set consisting of Holter ECG and training set including test set and test set was studied experimentally. The results show that the characteristic parameter ρ proposed in this paper can correctly classify 90% of test set. Thus, this method can effectively predict the spontaneous termination of atrial fibrillation from the Holter ECG signal.