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

Abstract

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

CLC Number:

R446.1

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.

References

[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.