Identification of Patients with Myocardial Infarction^[*]

Summary

Background: The largest morbidity and mortality group worldwide continues to be that suffering Myocardial Infarction (MI). The use of vectorcardiography (VCG) and electrocardiography (ECG) has improved the diagnosis and characterization of this cardiac condition.

Objectives: Herein, we applied a novel ECGVCG combination technique to identifying 95 patients with MI and to differentiating them from 52 healthy reference subjects. Subsequently, and with a similar method, the location of the infarcted area permitted patient classification.

Methods: We analyzed five depolarization and four repolarization indexes, say: a) volume; b) planar area; c) QRS loop perimeter; d) QRS vector difference; e – g) Area under the QRS complex, ST segment and T-wave in the (X, Y, Z) leads; h) ST-T Vector Magnitude Difference; i) T-wave Vector Magnitude Difference; and j) the spatial angle between the QRS complex and the T-wave.

For classification, patients were divided into two groups according to the infarcted area, that is, anterior or inferior sectors (MI-ant and MI-inf, respectively).

Results: Our results indicate that several ECG and VCG parameters show significant differences (p-value<0.05) between Healthy and MI subjects, and between MI-ant and MI-inf. Moreover, combining five parameters, it was possible to classify the MI and healthy subjects with a sensitivity = 95.8%, a specificity = 94.2%, and an accuracy = 95.2%, after applying a linear discriminant classifier method. Similarly, combining eight indexes, we could separate out the MI patients in MI-ant vs MI-inf with a sensitivity = 89.8%, 84.8%, respectively, and an accuracy = 89.8%.

Conclusions: The new multivariable MI patient identification and localization technique, based on ECG and VCG combination indexes, offered excellent performance to differentiating populations with MI from healthy subjects. Furthermore, this technique might be applicable to estimating the infarcted area localization. In addition, the proposed method would be an alternative diagnostic technique in the emergency room.

Keywords

Myocardial infarction - electrocardiography - vectorcardiography - discriminant analysis

^* Supplementary material published on our web-site http://dx.doi.org/10.3414/ME15-01-0101

^§ These authors contributed equally to this work.

• References

• 1 Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD. et al. Third Universal Definition of Myocardial Infarction. Journal of the American College of Cardiology 2012; 60 (Suppl. 16) 1581-1598.
• 2 Manocha AK, Singh M. An overview of ischemia detection techniques. International Journal of Scientific & Engineering Research 2011; 2 (Suppl. 11) 1-6.
• 3 Dilger J, Pietsch-Breitfeld B, Stein W, Overkamp D, Ickrath O, Renn W. et al. Simple computer-assisted diagnosis of acute myocardial infarction in patients with acute thoracic pain. Methods Inf Med 1992; 31: 263-267.
• 4 Correa R, Arini PD, Valentinuzzi ME, Laciar E. Novel set of vectorcardiographic parameters for the identification of ischemic patients. Medical Engineering & Physics 2013; 35 (Suppl. 01) 16-22.
• 5 Correa R, Arini PD, Correa L, Valentinuzzi ME, Laciar E. Acute myocardial ischemia monitoring before and during angioplasty by a novel vectorcardiographic parameter set. Journal of electrocardiology 2013; 46 (Suppl. 06) 635-643.
• 6 Correa R, Arini PD, Correa LS, Valentinuzzi M, Laciar E. Novel technique for ST-T interval characterization in patients with acute myocardial ischemia. Computers in biology and medicine 2014; 50: 49-55.
• 7 Bakul G, Tiwary U. Automated risk identification of myocardial infarction using relative frequency band coefficient (RFBC) features from ECG. The open biomedical engineering journal 2010; 4: 217.
• 8 Keshtkar A, Seyedarabi H, Sheikhzadeh P. et al. Discriminant analysis between myocardial infarction patients and healthy subjects using Wavelet Transformed signal averaged electrocardiogram and probabilistic neural network. Journal of Medical Signals and Sensors 2013; 3 (Suppl. 04) 225-230.
• 9 Arif M, Malagore IA, Afsar FA. Detection and localization of myocardial infarction using K-nearest neighbor classifier. Journal of medical systems 2012; 36 (Suppl. 01) 279-289.
• 10 Safdarian N, Dabanloo NJ, Attarodi G. A New Pattern Recognition Method for Detection and Localization of Myocardial Infarction Using T-Wave Integral and Total Integral as Extracted Features from One Cycle of ECG Signal. Journal of Biomedical Science and Engineering. Scientific Research Publishing. 2014
• 11 Bousseljot R, Kreiseler D, Schnabel A. Nutzung der EKG-Signaldatenbank CARDIODAT der PTB über das Internet. Biomedizinische Technik, Band 40, Ergänzungsband 1995; 1: 317.
• 12 Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PC, Mark RG. et al. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. Circulation 2000; 101 (Suppl. 23) e215-e220.
• 13 Gutierrez A, Lara M, Hernandez PR. A QRS Detector Based on Haar Wavelet, Evaluation with MIT-BIH Arrhythmia and European ST-T Databases. Computacion y Sistemas 2005; 8: 293-302.
• 14 Martinez JP, Almeida R, Olmos S, Rocha AP, Laguna P. A wavelet-based ECG delineator: evaluation on standard databases. IEEE Transactions on Biomedical Engineering 2004; 51 (Suppl. 04) 570-581.
• 15 Laciar E, Jane R, Brooks DH. Improved alignment method for noisy high-resolution ECG and Holter records using multiscale cross-correlation. IEEE Transactions on Biomedical Engineering 2003; 50 (Suppl. 03) 344-353.
• 16 Kardys I, Kors JA, van der Meer IM, Hofman A, van der Kuip DA, Witteman JCM. Spatial QRS-T angle predicts cardiac death in a general population. European Heart Journal. The Oxford University Press 2003; 24 (Suppl. 14) 1357-1364.
• 17 Rubulis A, Bergfeldt L, Rydén L, Jensen J. Prediction of cardiovascular death and myocardial infarction by the QRS-T angle and T vector loop morphology after angioplasty in stable angina pectoris: an 8-year follow-up. Journal of electrocardiology 2010; 43 (Suppl. 04) 310-317.
• 18 Dellborg M, Malmberg K, Ryden L, Svensson A, Swedberg K. Dynamic on-line vectorcardiography improves and simplifies in-hospital ischemia monitoring of patients with unstable angina. Journal of the American College of Cardiology 1995; 26 (Suppl. 06) 1501-1507.
• 19 Hall P, Hallén B, Selander H. Linear discriminatory analysis: a patient classifying method for research and production control. Methods Inf Med 1971; 10 (Suppl. 02) 96-102.
• 20 Flores JG, Jiménez EG, Gómez GR. Análisis discriminante. La Muralla; 2001
• 21 Powers DM. Evaluation: From Precision, Recall and F-Factor to ROC, Informedness, Markedness & Correlation. Adelaide: School of Informatics and Engineering, Flinders University of South Australia; 2007
• 22 Sokolova M, Lapalme G. A systematic analysis of performance measures for classification tasks. Information Processing & Management 2009; 45 (Suppl. 04) 427-437.