完全性束支传导阻滞优化判别的研究  

Optimization Study on Complete Bundle Branch Block Discrimination

作　　者：王明松[1] 张跃[1] 张集墨[1] 

机构地区：[1]清华大学深圳研究生院嵌入式系统与技术实验室,广东深圳518055

出　　处：《计算机仿真》2013年第5期395-398,共4页Computer Simulation

基　　金：湖南省科技重大专项基金资助项目(2007FJ1004)

摘　　要：在电心图优化诊断系统的研究中,研究完全性束支传导阻滞判别问题。目前完全性束支传导阻滞的判别方法中存在的效率低、准确性差等问题,提出了一种模板匹配优化判别算法。根据心电波形RR间期构造可变宽模板,采用数据压缩的方式将模板对齐,计算相关系数进行判别,并同时采用模板队列自动更新的方式,提高了判别效率。在实验中采用了MIT-BIH心电数据库的数据进行对比测试,得到了97%以上的平均准确识别率,实验表明,模板匹配优化判别算法可以准确高效地判别完全性束支传导阻滞。In this paper, the complete bundle branch block discrimination problem was studied. The tradition al- gorithms of complete bundle branch block discrimination have the defects of low accuracy and slow determination speed. In order to improve the performance and efficiency of complete bundle branch block discrimination, an optimi- zation algorithm based on template matching was presented. The correlation coefficient was used to detect complete bundle branch block beats from normal beats to improve the accuracy of complete bundle branch block discrimination, with a compressed method specially designed to adjust the width of the tested heartbeat or the template heartbeat, u- sing the RR interval to construct the adjustable width templates. Template queue was automatically updated to im- prove the discrimination efficiency. The performance of this algorithm was evaluated on the MIT-BIH arrhythmia da- tabase, and the result shows that the achieved average accuracy for complete bundle branch block is more than 97%. The experiments show that this optimization template matching algorithm can discriminate complete bundle branch block beats from normal beats accurately and effectively.

关 键 词：模板匹配 相关系数 完全性束支传导阻滞 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]