基于ESN的单导联动态心电信号的QRST消除

发布时间：2018-02-28 02:47

本文关键词： 动态心电图心电监护房颤 QRST消除 ESN　出处：《河北大学》2017年硕士论文　论文类型：学位论文

【摘要】：房颤作为最常见的老龄化疾病之一,发病率逐年增长,目前由于难以有效地跟踪房颤的状态变化,致使房颤的诊断准确率较低,治疗的效率和效果不理想。而随着远程心电监护系统的发展,医疗设备的小型化、便捷化,单导联的心电监护系统实现了更多应用,而利用远程心电监护系统监测到的心电图属于动态心电图,因此,研究适用于单导联动态心电信号的QRST波(由心室电活动产生)的消除方法有着非常重要的意义。本文的主要研究内容如下:(1)将ESN应用于单导联动态心电信号的QRST消除。在基于ESN的两导联心电信号的QRST消除方法(ESNTL)的基础上进行改进,将其用两导联心电信号消除QRST的方法改成用单导联心电信号消除QRST,并给ESN增加一路标记了R波位置的输入信号,使网络能够更准确的找到QRST的位置,提高了网络预测QRST的能力,增强了QRST消除的效果。(2)针对QRST消除问题,用微分进化算法和二进制粒子群算法对ESN的参数进行优化。采用微分进化算法对ESN的储备池参数进行优化,找到了一组比较合适的储备池参数,然后用二进制粒子群算法对ESN的输出层连接进行优化,简化了网络,提高了网络的性能,最后用优化的网络进行QRST预测,用房颤信号减去QRST,完成QRST消除。(3)将本文提出的方法应用于本实验室自主研发的心电监护服务平台中。通过对实际采集的心电信号进行QRST消除,验证本文提出方法的可行性。分别采用MIT-BIH数据库中的数据和本实验室采集的数据对本文方法进行可行性验证,实验结果表明本文提出的方法具有良好的QRST消除能力。
[Abstract]:As one of the most common aging diseases, the incidence of atrial fibrillation (AF) is increasing year by year. At present, it is difficult to track the state of AF effectively, which leads to the low diagnostic accuracy of AF. With the development of remote ECG monitoring system, the miniaturization of medical equipment and the convenience of treatment, the single-lead ECG monitoring system has realized more applications. The electrocardiogram monitored by the remote ECG monitoring system is a dynamic electrocardiogram, so, It is of great significance to study the elimination method of QRST wave (generated by ventricular electrical activity) suitable for single lead dynamic ECG signal. The main contents of this paper are as follows: 1) ESN is applied to QRST of single lead dynamic ECG signal. Based on the QRST elimination method of two-lead ECG signal based on ESN, The method of eliminating QRST by two-lead ECG signal is changed to that of single-lead ECG signal, and the input signal of R-wave position is added to ESN, so that the network can find the position of QRST more accurately and improve the ability of network to predict QRST. In order to solve the problem of QRST elimination, the differential evolution algorithm and binary particle swarm optimization algorithm are used to optimize the parameters of ESN. The differential evolution algorithm is used to optimize the parameters of ESN reserve pool. A set of suitable reserve pool parameters is found, and then the output layer connection of ESN is optimized by using binary particle swarm optimization algorithm, which simplifies the network and improves the performance of the network. Finally, the optimized network is used to predict the QRST. Using atrial fibrillation signal subtract QRST to complete QRST elimination. 3) the method proposed in this paper is applied to the ECG monitoring service platform developed independently by our laboratory. The QRST elimination of the actual ECG signal is carried out. The feasibility of the proposed method is verified by using the data in the MIT-BIH database and the data collected in our laboratory. The experimental results show that the proposed method has a good capability of QRST elimination.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R541.75;TN911.7

【相似文献】