融合散射变换与心率变异性分析的房颤检测
发布时间:2019-06-21 16:25
【摘要】:由于我国人口众多,加上经济与生活水平的提高,人口老龄化问题加剧,具有较高的发病率和死亡率的房颤也越来越受到医学界的关注。为了使房颤的无创电生理治疗成为可能,就必须保证房颤检测和房颤成分提取的精确度。到目前为止,已有的房颤算法研究在检测准确率上存在不足。本论文将散射变换与心率变异性分析融合来对房颤进行检测,主要研究工作如下:1、介绍了一种基于心率变异性分析的房颤检测,其中包括三个统计学方法,分别是香农熵、相邻RR间期差值均方根以及转折点比,并对房颤判断准则进行介绍;2、散射变换可以获得有效的信号特征,又可恢复损失的高频信息,本文基于散射变换对房颤进行检测;3、在房颤检测时选择性融合散射变换与心率变异性分析,首先基于心率变异性分析对房颤进行检测,估计其检测不好的样本用散射变换的方法进行检测,将两者进行融合得到最终的检测结果。通过MIT-BIH房颤数据库来评价本方法房颤检测准确率,并且与参考方法进行对比,验证了本方法的可行性。
[Abstract]:Due to the large population of our country, coupled with the improvement of economy and living standards, the problem of population aging is aggravated, and atrial fibrillation with high incidence and mortality has been paid more and more attention by the medical profession. In order to make the noninvasive electrophysiological treatment of atrial fibrillation possible, it is necessary to ensure the accuracy of atrial fibrillation detection and atrial fibrillation component extraction. Up to now, there are some shortcomings in the detection accuracy of the existing Atrial Fibrillation algorithms. In this paper, the fusion of scattering transformation and heart rate variability analysis is used to detect atrial fibrillation. The main research work is as follows: 1. A detection of atrial fibrillation based on heart rate variability analysis is introduced, including three statistical methods, namely, Shennong entropy, mean square root mean square difference of adjacent RR interval and turning point ratio, and the judgment criteria of atrial fibrillation are introduced. 2. Scattering transform can obtain effective signal characteristics and recover the high frequency information of loss. In this paper, AF is detected based on scattering transform. 3, selective fusion scattering transform and heart rate variability analysis are used in AF detection. Firstly, AF is detected based on HRV analysis, and the samples whose detection is not good are detected by scattering transformation method, and the final detection results are obtained by fusion of the two. The accuracy of AF detection was evaluated by MIT-BIH Atrial Fibrillation Database, and the feasibility of this method was verified by comparing it with the reference method.
【学位授予单位】:河北大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:R541.75;TN911.7
本文编号:2504211
[Abstract]:Due to the large population of our country, coupled with the improvement of economy and living standards, the problem of population aging is aggravated, and atrial fibrillation with high incidence and mortality has been paid more and more attention by the medical profession. In order to make the noninvasive electrophysiological treatment of atrial fibrillation possible, it is necessary to ensure the accuracy of atrial fibrillation detection and atrial fibrillation component extraction. Up to now, there are some shortcomings in the detection accuracy of the existing Atrial Fibrillation algorithms. In this paper, the fusion of scattering transformation and heart rate variability analysis is used to detect atrial fibrillation. The main research work is as follows: 1. A detection of atrial fibrillation based on heart rate variability analysis is introduced, including three statistical methods, namely, Shennong entropy, mean square root mean square difference of adjacent RR interval and turning point ratio, and the judgment criteria of atrial fibrillation are introduced. 2. Scattering transform can obtain effective signal characteristics and recover the high frequency information of loss. In this paper, AF is detected based on scattering transform. 3, selective fusion scattering transform and heart rate variability analysis are used in AF detection. Firstly, AF is detected based on HRV analysis, and the samples whose detection is not good are detected by scattering transformation method, and the final detection results are obtained by fusion of the two. The accuracy of AF detection was evaluated by MIT-BIH Atrial Fibrillation Database, and the feasibility of this method was verified by comparing it with the reference method.
【学位授予单位】:河北大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:R541.75;TN911.7
【参考文献】
相关期刊论文 前6条
1 季虎;孙即祥;王春光;;基于小波变换的自适应QRS-T对消P波检测算法[J];电子与信息学报;2007年08期
2 陈丽丽;杨水祥;;心房颤动发生机制和治疗方案的研究[J];国际心血管病杂志;2010年03期
3 刘学胜;;基于PCA和SVM算法的人脸识别[J];计算机与数字工程;2011年07期
4 陈煜;辛怡;王振宇;李勤;;基于多尺度小波熵的阵发性房颤识别方法[J];航天医学与医学工程;2013年05期
5 陈灏珠;;心房颤动诊断与治疗的进展和展望[J];中国实用内科杂志;2006年02期
6 毛玲;张国敏;孙即祥;;基于位置估计与识别后处理的心电信号P波检测[J];信号处理;2009年06期
相关硕士学位论文 前1条
1 季加阳;面向人体多生理参数的非线性动力特征分析[D];哈尔滨工业大学;2013年
,本文编号:2504211
本文链接:https://www.wllwen.com/yixuelunwen/xxg/2504211.html
最近更新
教材专著