基于心电动力学离散特征与SVM模型的AMI早期筛查研究

中华养生保健 ›› 2023, Vol. 41 ›› Issue (7): 4-9.

基于心电动力学离散特征与SVM模型的AMI早期筛查研究

胡伟^1,2, 刘赵阳^1,2, 罗显元^1,2,*

1.苏州大学附属常州老年病医院心内科,江苏常州,213000;
2.常州市第七人民医院心内科,江苏常州,213000

出版日期:2023-04-01 发布日期:2023-03-30
通讯作者: ^*罗显元,E-mail：azyy410@163.com。
作者简介:胡伟（1977—）,男,汉族,籍贯：吉林省延边朝鲜族自治州,本科,副主任医师,研究方向：冠心病的介入治疗。
基金资助:
苏州高新区医疗卫生科技计划项目（2018Z005）

Study on Early Screening of AMI Based on Discrete Characteristics of ECG Dynamics and SVM Model

HU Wei^1,2, LIU Zhao-yang^1,2, LUO Xian-yuan^1,2,*

1. Department of Cardiology, Changzhou Geriatric Hospital Affiliated to Soochow University, Changzhou Jiangsu, 213000, China;
2. Department of Cardiology, Changzhou No.7 People’s Hospital, Changzhou Jiangsu, 213000, China

Online:2023-04-01 Published:2023-03-30

摘要/Abstract

摘要： 目的以急性心肌梗死（AMI）患者和健康人的心电动力学（CDG）数据为基础,研究一种新早期筛查模型用以评估健康人群和AMI患者。方法训练集包括1500例AMI患者和1500名健康个体。在2000人中验证了SVM模型。测试无或有轻度症状的AMI患者的心电（ECG）信号,获取ECG信号的CDG数据,分析CDG数据的离散度特征,并基于支持向量机（SVM）建立早期筛查模型以评估健康人群和AMI患者。结果 AMI患者CDG数据的离散度特征的定量值与健康个体存在显著差异,AMI患者的CDG数据比正常人的CDG数据混乱,支持向量机模型用于AMI诊断的准确性较高。结论基于心电动力学离散特征与SVM模型的AMI早期筛查方法的应用可以有效区分AMI患者和健康个体,该方法为AMI的早期筛查提供了辅助方法。

关键词: 急性心肌梗死, 心电动力学数据, 离散特征, 支持向量机模型, 早期筛查算法

Abstract: Objective Based on the cardiodynamicsgram (CDG) data of acute myocardial infarct (AMI) patients and healthy people, a new early screening model was proposed to evaluate healthy people and AMI patients. Methods The training set included 1500 patients with AMI and 1500 healthy individuals. The SVM models were validated in 2000 individuals. ECG signal of AMI patients without or with mild symptoms was tested to obtain the CDG data of the ECG signal. The dispersion characteristics of CDG data from patients with AMI without or with mild symptoms were extracted, and then support vector machines (SVM) were used to construct an early screening model to screen and evaluate ECG signals in healthy population and patients with AMI. Results The quantitative values of the dispersion characteristics of patients with AMI were significantly different from those of healthy individuals. The CDG data of patients with AMI was more disordered than that of normal individuals. The accuracy of SVM model for AMI diagnosis was 84.05%. The average time consumed to analyze a patient by this model was 2 minutes. Conclusion The application of AMI early screening method based on the dispersion characteristics of CDG data and SVM model can effectively distinguish between patients with AMI and healthy individuals. This method provide an ancillary method of early screening of AMI.

Key words: acute myocardial infarction, cardiodynamicsgram data, dispersion characteristics, support vector machine model, early screening algorithm

中图分类号:

R446.1

胡伟, 刘赵阳, 罗显元. 基于心电动力学离散特征与SVM模型的AMI早期筛查研究[J]. 中华养生保健, 2023, 41(7): 4-9.

HU Wei, LIU Zhao-yang, LUO Xian-yuan. Study on Early Screening of AMI Based on Discrete Characteristics of ECG Dynamics and SVM Model[J]. ZHONGHUA YANGSHENG BAOJIAN, 2023, 41(7): 4-9.

参考文献

[1] JUNG C, ELSÄSSER A. Update ESC-Leitlinie 2017-Akuter Myokardinfarkt (STEMI)[J]. Dtsch Med Wochenschr.2018,143(11):797-801.
[2] HEDAYATI T, YADAV N, KHANAGAVI J.Non-ST-Segment Acute Coronary Syndromes[J]. Cardiol Clin,2018,36(1):37-52.
[3] LEVINE G N, BATES E R, BITTL J A, et al.2016 ACC/AHA Guideline Focused Update on Duration of Dual Antiplatelet Therapy in Patients With Coronary Artery Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines[J]. Journal of the American College of Cardiology,2016,68(10):1082-1115.
[4] TROST J C, LANGE R A.Treatment of acute coronary syndrome: part 2: ST-segment elevation myocardial infarction[J]. Crit Care Med, 2012,40(6):1939-1945.
[5] VOGEL B, CLAESSEN B E, ARNOLD S V, et al.ST-segment elevation myocardial infarction[J]. Nat Rev Dis Primers,2019,5(1):39.
[6] JAYARAJ J C, DAVATYAN K, SUBRAMANIAN S S, et al.Epidemiology of Myocardial Infarction[M]. Intechopen,2019.
[7] HAHN J, FU Y P, BROWN M R, et al.Genetic loci associated with prevalent and incident myocardial infarction and coronary heart disease in the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium[J]. PLoS One,2020,15(11):e0230035.
[8] MYERS P D, SCIRICA B M, STULTZ C M.Machine Learning Improves Risk Stratification After Acute Coronary Syndrome. Sci Rep[J]. 2017,7(1):12692.
[9] TAKASHI T.Coherent mode decomposition using mixed Wigner functions of Hermite-Gaussian beams[J]. Optics Letters,2017,42(8):1576.
[10] WEI Z A, JIAN Y A, CY B, et al.Classification of myocardial infarction based on hybrid feature extraction and artificial intelligence tools by adopting tunable-Q wavelet transform (TQWT), variational mode decomposition (VMD) and neural networks[J]. Artificial Intelligence in Medicine,2020:106.
[11] NARULA S, SHAMEER K, SALEM OMAR A M, et al. Machine-Learning Algorithms to Automate Morphological and Functional Assessments in 2D Echocardiography. J Am Coll Cardiol[J]. 2016,68(21):2287-2295.
[12] 宿燕岗, 梁义秀.慢性心衰心脏性猝死的一级预防及其策略[J].南京医科大学学报(自然科学版),2016,36(4):402-405,410.
[13] VALEJO A, FALEIROS T, OLIVEIRA M, et al.A coarsening method for bipartite networks via weight-constrained label propagation[J]. Knowledge-Based Systems,2020,195:105678.
[14] CONG W, XUNDE D, SHANXING O U, et al.A new method for early detection of myocardial ischemia:cardiodynamicsgram(CDG)[J]. Science China(Information Sciences),2016,59(1):95-105.
[15] 文斯民. 基于确定学习的心肌缺血早期诊断系统的配套信息管理系统的开发[D].广州:华南理工大学,2016.
[16] JIANG H, ZOU B, XU C, et al.SVM-Boosting based on Markov resampling: Theory and algorithm[J]. Neural Networks,2020,131:276-290.
[17] 易力,李祥,何俊德,等.心电动力学信号数据特征提取与模式识别的研究进展[J].实用心电学杂志,2019,28(2):136-40.
[18] LUO C, JIANG Z, ZHENG Y.A novel reconstructed training-set SVM with roulette cooperative coevolution for financial time series classification[J]. Expert Systems with Application,2019,123(6):283-298.
[19] UTKIN L V.An imprecise extension of SVM-based machine learning models[J]. Neurocomputing,2018,331:18-32.
[20] XIA W Y. GA-SVM based feature selection and parameter optimization in hospitalization expense modeling[J]. Applied Soft Computing,2019,75:
323-332.
[21] MONTRIEF T, DAVIS W T, KOYFMAN A, et al.Mechanical, inflammatory, and embolic complications of myocardial infarction: An emergency medicine review[J]. Am J Emerg Med,2019,37(6):1175-1183.
[22] SUN H, LI J, WANG Y, et al.Effect of Mobile Internet on Attitude and Self-Efficacy of Patients with Coronary Heart Disease Diagnosed by 12-Lead Holter ECG[J]. J Healthc Eng,2022,13:3414178.
[23] DENG M, TANG M, WANG C, et al.Cardiodynamicsgram as a New Diagnostic Tool in Coronary Artery Disease Patients With Nondiagnostic Electrocardiograms[J]. Am J Cardiol,2017,119(5):698-704.