Detection of ST segment in electrocardiogram signal based on two-dimensional cloudy model theory

doi:10.3969/j.issn.2095-4344.2013.04.014

Abstract

Abstract:

BACKGROUND: There is no precise classificated mathematical model and identification standard for the morphological characteristics of abnormal ST segment, and therefore, the improvement of automatical recognization will be limited in some cardiovascular diseases diagnoses.
OBJECTIVE: To find a new method meeting the medical diagnosis logical thinking for analyzing the ST segment in electrocardiogram signal, and to provide new ideas for real-time analysis of the changes ofelectrocardiogram of ST segment.
METHODS: A new algorithm for detecting ST segment in electrocardiogram signal based on two-dimensional cloudy model theory was proposed in view of the fuzziness and randomness of electrocardiogram signal. And then, it estimated the membership grade of two-dimensional cloud generator and identified the morphology of ST segment in electrocardiogram.
RESULTS AND DISCUSSION: The algorithm was simulated with Matlab, and the algorithm accuracy was verified by standard ECG database (European Community CSE database). The statistical results showed that the algorithm had a high identification rate of ST segment, and it was shortcut and effective for processing the large amount of data which provide a new method for the accurate analysis of the ST segment.

Key words: bone and joint implants, photographs and images of bone and joint, electrocardiogram signal, ST segment, signal processing, two-dimensional cloudy model, clustering analysis, the CSE database, digital medicine, other grants-supported paper, photographs-containing paper of bone and joint implants

CLC Number:

R318

Liu Xin-xu, Su Zhi-jian, Gao Zhen-kui, Liu Yan-tao, Xia Zhen-hong, Liu Na. Detection of ST segment in electrocardiogram signal based on two-dimensional cloudy model theory[J]. Chinese Journal of Tissue Engineering Research, 2013, 17(4): 653-657.

Figures/Tables 8

2 基于二维云模型心电信号ST形态分析 2.1 聚类分析算法流程本实验将云变换和综合云思想应用于心电信号检测[11]，并实现了心电信号ST段的聚类分析。算法核心是利用云模型区分ST段，即将ST段划分为不同类别。首先考虑到任意一种的概率分布都可由若干个正态分布组合得到，因此心电信号ST段数据点的频率分布可认为是若干个正态云的叠加。同时，云模型的期望就是数据分布的重心，亦最能代表所分析的心电信号ST段特性。基于此，实验将定量的ST段数据点转化为一系列用云模型表示的定性概念。另外，由于ST段采样点电位值及其一阶导数值所对应的云滴是一次随机实现；所以将定性概念表示为ST段的类别特性，并用不同类别的相似程度表示为云滴的离散程度，可以有效地解决用云滴的离散度对定性概念(ST段特征类别)的不确定度(ST段不同形态类别)的度量问题。算法具体的流程如下：输入：心脏类疾病患者的心电信号数据。输出：具有ST段判断标准的综合云。①心电信号经预处理后，提取心电信号特征点QRS波群、J点、Ts点、等电位点[9]。②提取心电信号ST段数据，对数据进行归一化处理[10]。③统计采样点电位值及其一阶导数的分布，得到分布曲线(即频率直方图)；根据分布曲线确定基云数目，并求出各基云的数字特征，从而得出其云滴图和二维正态概率密度曲面图。④用二维逆向正态云模型对采样点电位大小及其一阶导数两个分量进行统计训练，以生成表征ST段信息定性概念的二维正态云模型的数字特征(Ex1，En1，He1，Ex2，En2，He2)。⑤采用二维正向云发生器得到ST段形态分析的各基云。 2.2 ST段形态判别在临床上，医务人员一般是根据诊断医学的逻辑思维方法来自行对ST段的形态进行分类及判别，并以此来作为ST是否具有病兆的依据。本实验首先按照ST段形态的偏移方向和曲率变化将其进行分类，为下一步实现计算机的自动判别做准备。ST段的形态改变分类，见表1。 "

References

[1] Liu X. Linchuang Xindianxue Zazhi. 2005;14(03):159-160. 刘霞.ST段改变的形态学分型[J].临床心电学杂志, 2005;14(03): 159-160.

[2] Wang ZH, Li XZ, Zheng GZ. Neimenggu Yixue Zazhi. 2001; 33(06):541-542. 王珍虎,李香枝,郑桂珍.心电图ST段改变对判断急性心肌梗死近期预后价值的探讨[J].内蒙古医学杂志,2001,33(06):541-542.

[3] Xiao DP, Xu Z, Yang H, et al. Chongqingshi Dianji Gongcheng Xuehui. 2004. 肖冬萍,徐征,杨浩,等.心电图ST段检测与分析方法研究[C].重庆市电机工程学会,2004.

[4] Weisner SJ, Tompkins WJ. A compact, microprocessor-based ECG St-segment monitor for the operating room. IEEE Trans. Biomed Eng. 1982;29:642-649.

[5] Uthamalingam S, Zheng H, Leavitt M, et al. Exercise-induced ST-segment elevation in ECG lead aVR is a useful indicator of significant left main or ostial LAD coronary artery stenosis. JACC Cardiovasc Imaging. 2011;4(2):176-186.

[6] Yang J, Wang HS, Yu MS. Beijing Shengwu Yixue Gongcheng. 2002;21(2):106-108. 杨军,王宏山,俞梦孙.心电图ST段的神经网络测量方法[J]. 北京生物医学工程,2002,21(2):106-108.

[7] Fan XD, Zhu ZH, Yang SH. Shengwu Yixue Gongcheng Zazhi. 1997;14(1):51-53. 范晓东,朱泽煌,杨世豪. 动态心电图ST-T段参数测量方法[J]. 生物医学工程杂志,1997,14(1):51-53.

[8] Chahine RA, Lowery MH, Bauerlein EJ. Interpretation of the exercise-induced ST-segment elevation. Am J Cardiol. 1993; 72(1):100-102.

[9] Mao L, Zhang GM, Sun JX. Xinhao Chuli. 2009;25(09): 1360-1365. 毛玲,张国敏,孙即祥.心电图ST段形态分析方法研究[J].信号处理,2009,25(09):1360-1365.

[10] Yang CH, Li DY. Jisuanji Xuebao. 1998;21(11):961-969. 杨朝晖,李德毅.二维云模型及其在预测中的应用[J].计算机学报,1998,21(11):961-969.

[11] Miranda-Cid A, Alvarado-Serrano C. An ECG ambulatory system with mobile embedded architecture for ST-segment analysis. Conf Proc IEEE Eng Med Biol Soc. 2010;2010:78-81.

[12] Cao XW, Deng QK. Shengwu Yixue Gongchengxue Zazhi. 2001;18(3):373-377. 曹细武,邓亲恺.基于KL变换自动监测心电图ST-T段趋势的方法研究[J].生物医学工程学杂志,2001,18(3):373-377.

[13] Li DY, Du Y. Beijing:National Defence Industrial Press. 2005; 15(11):137-138. 李德毅,杜鹢.不确定性人工智能[J].国防工业出版社, 2005, 15(11): 137-138.

[1]	Zhao Dezhu, Guan Tianmin, Wu Bin, Mei Zhao. Design and implementation of an adolescent idiopathic scoliosis orthosis design expert system based on fuzzy logic [J]. Chinese Journal of Tissue Engineering Research, 2020, 24(33): 5255-5261.
[2]	Wang Haijiao, Huang Ruina, Wang Xiaojun, Zheng Bili, Chen Peina, Zeng Xiujuan. Analysis on the hotspot of the platelet-rich plasma based on VOSviewer [J]. Chinese Journal of Tissue Engineering Research, 2019, 23(18): 2947-2952.
[3]	Li Wei1, Zhang Hong, Cao Li-jun, Ren Guo-shan, Yan Zhan-ping, Zhao Chang-yi, Gao Zeng-min. Establishment of normal lumbar and osteoporosis three-dimensional finite element models [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1521-1526.
[4]	Zhang Xue-jun, Wang Chen. Digital templating alleviates leg length discrepancy after total hip arthroplasty [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1527-1534.
[5]	Li Guang-wei, Wang Hong-jun, Sun Xiao-zhi. Total knee arthroplasty for the treatment of severe knee disease [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1535-1542.
[6]	Wang Nian-hong, Yan Jun-tao, Sun Wu-quan, Hu Yong-shan, Xia Jun, Wei Li-cheng, Wu Yi, Jia Jie, Ouyang Gui-lin, He Yong, Guo Yan-ming, Xu Jie. Therapeutic effect of early Tuina rehabilitation treatment after total knee arthroplasty [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1543-1548.
[7]	Yang Peng, Bai Jian, Wang Xin-ran, Geng Yan-li. Intelligent prosthetic ankle based on the finite state machine control [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1549-1554.
[8]	Ao Jun, Wan Lei, Liao Wen-bo, Xue Zhong-lin, Yu Bo, Jin An-min. Screw stress features of the posterior asymmetric or symmetric fixation of transforaminal lumbar fusion based on a three-dimensional finite element method [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1555-1562.
[9]	Ni Wei-feng, Xu Jian-guang, Xue Feng. Three-dimensional finite element analysis of in situ and reductive interbody fusion of lumbar spondylolisthesis [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1563-1570.
[10]	Xue Jian, Jin An-min, Sun Xiao-ping, Wang Yan-bin, Xie Wei-yong. Clinical outcomes of unilateral lumbar pedicle screw combined with translamina facet screw fixation versus bilateral fixation [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1571-1578.
[11]	Zhang Qi-wei, Zhang Yao-nan, Sun Chang-tai, Xu Hong-bing. Computer-assisted versus free-hand pedicle screw implantation [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1579-1585.
[12]	Yi Guo-liang, Song Xi-zheng, Wang Wen-jun, Yao Nü-zhao. Biomechanical testing of adjacent double segment lumbar fractures treated with the spinal external fixator [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1586-1591.
[13]	Jia Ting-ting, Zhang Yun-tao, Liu Shun-zhen, Sun Xin, Hou Yu-dong. Chemokin CXC motif receptor 4 expression in bone tissue surrounding anode-oxidized implants during early implantation [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1592-1597.
[14]	Liu Yue, Zheng Shi-wei, Wu Liang, Zhang Yan, Yang Tie-yi. Femoral neck shortening after fracture fixation with cannulated screws [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1598-1604.
[15]	Xu Ke-lin, Yin Qu-dong, Gu San-jun, Sun Zhen-zhong, Shou Kui-shui. Absorbable screw internal fixation: Complications and prevention measures [J]. Chinese Journal of Tissue Engineering Research, 2013, 17(9): 1605-1610.