BACKGROUND: Early detection and diagnosis of myocardial infarction (MI) help physicians save lives through timely treatment. Vectorcardiography (VCG) is an alternative to the 12-lead electrocardiography, providing not only characteristic changes in cardiac electrical activity in MI patients but also unique spatial information often overlooked by traditional methods. Despite its potential, comprehensive comparative studies applying machine learning (ML) techniques specifically to VCG data remain limited. METHODS: This study proposes a novel VCG processing methodology using a comparative analysis of machine learning-based algorithms for the automated detection of MI patients from VCG recordings, utilizing extracted domain knowledge VCG features that monitor morphological changes in cardiac activity. For this purpose, records from the PTB Diagnostic dataset were used. The extracted domain knowledge dataset of morphological features was then fed into a diverse set of 210 machine learning configurations, including K-nearest neighbor, Support Vector Machine, Discriminant Analysis, Artificial Neural Network, Decision Tree, Random Forest, Naive Bayes, Logistic Regression, and Ensemble Methods. To further improve classification performance, we combined analyzed high-performing models using a stacking ensemble strategy, which integrates multiple base classifiers into a meta-classifier. RESULTS: The stacking-based decision-level fusion achieved high accuracy of 95.55%, sensitivity of 97.70%, specificity of 86.25%, positive predictive value of 96.86%, negative predictive value of 89.61% and f1-score of 97.27%. CONCLUSION: The results demonstrate that decision-level fusion via stacking improves classification performance and enhances the reliability of MI detection from VCG recordings, supporting cardiologists in decision-making.

Faculty of Electrical Engineering and Computer Science VSB Technical University of Ostrava Ostrava Czechia

Zobrazit více v PubMed

Acharya U. R., Fujita H., Oh S. L., Hagiwara Y., Tan J. H., Adam M. (2017). Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals. Inf. Sci. 415, 190–198. 10.1016/j.ins.2017.06.027 DOI

Alexandropoulos S.-A. N., Aridas C. K., Kotsiantis S. B., Vrahatis M. N. (2019). Stacking strong ensembles of classifiers. Springer, 545–556.

Amini M., Zayeri F., Salehi M. (2021). Trend analysis of cardiovascular disease mortality, incidence, and mortality-to-incidence ratio: results from global burden of disease study 2017. BMC Public Health 1, 12–412. 10.1186/s12889-021-10429-0 PubMed DOI PMC

Aranda Hernandez A., Bonizzi P., Peeters R., Karel J. (2023). Automatic localization of myocardial infarction using vectorcardiography.

Balakrishnama S., Ganapathiraju A. (1998). “Linear discriminant analysis-a brief tutorial,”Inst. Signal Information Process. 18. 1–8.

Barmpoutis P., Dimitropoulos K., Apostolidis A., Grammalidis N. (2019). Multi-lead ECG signal analysis for myocardial infarction detection and localization through the mapping of Grassmannian and euclidean features into a common hilbert space. Biomed. Signal Process. Control 52, 111–119. 10.1016/j.bspc.2019.04.003 DOI

Bartall H., Desser K. B., Benchimol A., Massey B. J. (1978). Echocardiographic left atrial enlargement. Comparison of vectorcardiogram and electrocardiogram for detection. J. Electrocardiol. 11, 355–359. 10.1016/s0022-0736(78)80141-1 PubMed DOI

Belgiu M., Dr Drăfuţ L. (2016). Random forest in remote sensing: a review of applications and future directions. ISPRS Journal Photogrammetry Remote Sensing 114, 24–31. 10.1016/j.isprsjprs.2016.01.011 DOI

Bousseljot R., Kreiseler D., Schnabel A. (1995). Nutzung der EKG-Signaldatenbank CARDIODAT der PTB über das Internet. Biomed. Technik/Biomedical Eng. 40, 317–318. 10.1515/bmte.1995.40.s1.317 DOI

Chuang Y.-H., Lee C.-Y., Chen Y.-H., Chang W.-W. (2023). A hybrid system for myocardial infarction classification with derived vectorcardiography IEEE 468–473. 10.1109/icufn57995.2023.10200755 DOI

Correa R., Arini P. D., Correa L., Valentinuzzi M. E., Laciar E. (2013a). Acute myocardial ischemia monitoring before and during angioplasty by a novel vectorcardiographic parameter set. J. Electrocardiology 46, 635–643. 10.1016/j.jelectrocard.2013.06.001 PubMed DOI

Correa R., Arini P. D., Valentinuzzi M. E., Laciar E. (2013b). Novel set of vectorcardiographic parameters for the identification of ischemic patients. Med. Engineering and Physics 35, 16–22. 10.1016/j.medengphy.2012.03.005 PubMed DOI

Correa R., Arini P. D., Correa L. S., Valentinuzzi M., Laciar E. (2016). Identification of patients with myocardial infarction. Vectorcardiographic and electrocardiographic analysis. Methods Information Medicine 55, 242–249. 10.3414/ME15-01-0101 PubMed DOI

Cowdery C. D., Wagner G., Starr J., Rogers G., Greenfield Jr J. (1980). New vectorcardiographic criteria for diagnosing right ventricular hypertrophy in mitral stenosis: comparison with electrocardiographic criteria. Circulation 62, 1026–1032. 10.1161/01.cir.62.5.1026 PubMed DOI

De la Garza Salazar F., Egenriether B. (2024). Exploring vectorcardiography: an extensive vectocardiogram analysis across age, sex, BMI, and cardiac conditions. J. Electrocardiol. 82, 100–112. 10.1016/j.jelectrocard.2023.12.004 PubMed DOI

Dehnavi A. R. M., Farahabadi I., Rabbani H., Farahabadi A., Mahjoob M. P., Dehnavi N. R. (2011). Detection and classification of cardiac ischemia using vectorcardiogram signal PubMed PMC

Dellborg M., Malmberg K., Ryden L., Svensson A.-m., Swedberg K. (1995). Dynamic on-line vectorcardiography improves and simplifies in-hospital ischemia monitoring of patients with unstable angina. J. Am. Coll. Cardiol. 26, 1501–1507. Publisher: Journal of the American College of Cardiology. 10.1016/0735-1097(95)00361-4 PubMed DOI

Dhawan A., Wenzel B., George S., Gussak I., Bojovic B., Panescu D. (2012). Detection of acute myocardial infarction from serial ECG using multilayer support vector machine. IEEE 2012, 2704–2707. 10.1109/EMBC.2012.6346522 PubMed DOI

Dietterich T. G. (2000). Ensemble methods in machine learning. Springer, 1–15.

Diker A., Cömert Z., Avci E., Velappan S. (2018). Intelligent system based on genetic algorithm and support vector machine for detection of myocardial infarction from ECG signals. IEEE 1–4, 1–4. 10.1109/siu.2018.8404299 DOI

Dima S.-M., Panagiotou C., Mazomenos E. B., Rosengarten J. A., Maharatna K., Gialelis J. V., et al. (2013). On the detection of myocadial scar based on ECG/VCG analysis. IEEE Trans. Biomed. Eng. 60, 3399–3409. 10.1109/TBME.2013.2279998 PubMed DOI

Dohare A. K., Kumar V., Kumar R. (2018). Detection of myocardial infarction in 12 lead ECG using support vector machine. Appl. Soft Comput. 64, 138–147. 10.1016/j.asoc.2017.12.001 DOI

Dreiseitl S., Ohno-Machado L. (2002). Logistic regression and artificial neural network classification models: a methodology review. J. Biomedical Informatics 35, 352–359. 10.1016/s1532-0464(03)00034-0 PubMed DOI

Fatimah B., Singh P., Singhal A., Pramanick D., Pranav S., Pachori R. B. (2021). Efficient detection of myocardial infarction from single lead ECG signal. Biomed. Signal Process. Control 68, 102678. 10.1016/j.bspc.2021.102678 DOI

Filos D., Chouvarda I., Tachmatzidis D., Vassilikos V., Maglaveras N. (2017). Beat-to-beat P-wave morphology as a predictor of paroxysmal atrial fibrillation. Comput. Methods Programs Biomed. 151, 111–121. 10.1016/j.cmpb.2017.08.016 PubMed DOI

Frank E. (1956). An accurate, clinically practical system for spatial vectorcardiography. Circulation 13, 737–749. 10.1161/01.cir.13.5.737 PubMed DOI

Ghosal T., Hembram A., Ahmed I., Goswami D. P., Bandyopadhyay A., Sengupta A. (2024). Exploring cardiac vector propagation in acute myocardial infarction: a spatial velocity perspective. Future Cardiol. 20, 1–10. 10.1080/14796678.2024.2440257 PubMed DOI PMC

Goernig M., Hoeffling B., Lau S., Figulla H. R., Haueisen J. (2015). T vector and loop characteristics improve detection of myocardial injury after infarction. J. Med. Biol. Eng. 35, 381–386. 10.1007/s40846-015-0041-8 DOI

Goldberger A., Amaral L., Glass L., Hausdorff J., Ivanov P. C., Mark R., et al. (2000). Components of a new research resource for complex physiologic signals. PhysioBank, PhysioToolkit, Physionet. 10.1161/01.cir.101.23.e215 PubMed DOI

Gragnaniello M., Borghese A., Marrazzo V. R., Maresca L., Breglio G., Irace A., et al. (2024). Real-time myocardial infarction detection approaches with a microcontroller-based Edge-AI device. Sensors 24, 828. 10.3390/s24030828 PubMed DOI PMC

Habibzadeh F., Habibzadeh P., Yadollahie M. (2016). On determining the Most appropriate test cut-off value: the case of tests with continuous results. Biochem. Medica 26, 297–307. 10.11613/BM.2016.034 PubMed DOI PMC

Hafshejani N. J., Mehridehnavi A., Hajian R., Boudagh S., Behjati M. (2021). Identification of myocardial infarction using morphological features of electrocardiogram and vectorcardiogram. IET Signal Process. 15, 674–685. 10.1049/sil2.12072 DOI

Hammad M., Maher A., Wang K., Jiang F., Amrani M. (2018). Detection of abnormal heart conditions based on characteristics of ECG signals. Measurement 125, 634–644. 10.1016/j.measurement.2018.05.033 DOI

Hammad M., Chelloug S. A., Alkanhel R., Prakash A. J., Muthanna A., Elgendy I. A., et al. (2022). Automated detection of myocardial infarction and heart conduction disorders based on feature selection and a deep learning model. Sensors 22, 6503. 10.3390/s22176503 PubMed DOI PMC

Han C., Shi L. (2019). Automated interpretable detection of myocardial infarction fusing energy entropy and morphological features. Comput. Methods Programs Biomedicine 175, 9–23. 10.1016/j.cmpb.2019.03.012 PubMed DOI

He C., Liu M., Xiong P., Yang J., Du H., Xu J., et al. (2022). Localization of myocardial infarction using a multi-branch weight sharing network based on 2-D vectorcardiogram. Eng. Appl. Artif. Intell. 116, 105428. 10.1016/j.engappai.2022.105428 DOI

Hearst M. A., Dumais S. T., Osuna E., Platt J., Scholkopf B. (1998). Support vector machines. IEEE Intelligent Syst. Their Applications 13, 18–28. 10.1109/5254.708428 DOI

Hernandez A. A., Bonizzi P., Karel J., Peeters R. (2018). Myocardial ischemia diagnosis using a reduced lead system IEEE 5302–5305. 10.1109/embc.2018.8513511 PubMed DOI

Hsieh C.-J., Chang K.-W., Lin C.-J., Keerthi S. S., Sundararajan S. (2008). A dual coordinate descent method for large-scale linear SVM, 408–415.

Jambukia S. H., Dabhi V. K., Prajapati H. B. (2015). “Classification of ECG signals using machine learning techniques: a survey,” in 2015 international conference on advances in computer engineering and applications (IEEE; ), 714–721.

Karisik F., Baumert M. (2019). A long short-term memory network to classify myocardial infarction using vectorcardiographic ventricular depolarization and repolarization. IEEE–1. 10.22489/CinC.2019.272 DOI

Keshtkar A., Seyedarabi H., Sheikhzadeh P., Rasta S. H. (2013). Discriminant analysis between myocardial infarction patients and healthy subjects using wavelet transformed signal averaged electrocardiogram and probabilistic neural network. J. Medical Signals Sensors 3, 225–230. 10.4103/2228-7477.128316 PubMed DOI PMC

Ketu S., Mishra P. K. (2022). Empirical analysis of machine learning algorithms on imbalance electrocardiogram based arrhythmia dataset for heart disease detection. Arabian J. Sci. Eng. 47, 1–23. 10.1007/s13369-021-05972-2 DOI

Khan S. I., Pachori R. B. (2021a). Automated detection of posterior myocardial infarction from vectorcardiogram signals using fourier–bessel series expansion based empirical wavelet transform. IEEE Sensors Lett. 5, 1–4. 10.1109/lsens.2021.3070142 DOI

Khan S. I., Pachori R. B. (2021b). Derived vectorcardiogram based automated detection of posterior myocardial infarction using FBSE-EWT technique. Biomed. Signal Process. Control 70, 103051. 10.1016/j.bspc.2021.103051 DOI

Kijonka J., Vavra P., Zonca P., Penhaker M. (2022). A wavelet-based VCG QRS loop boundaries and isoelectric coordinates detector. Front. Physiology 13, 941827. 10.3389/fphys.2022.941827 PubMed DOI PMC

Kors J., Van Herpen G., Sittig A., Van Bemmel J. (1990). Reconstruction of the frank vectorcardiogram from standard electrocardiographic leads: diagnostic comparison of different methods. Eur. Heart Journal 11, 1083–1092. 10.1093/oxfordjournals.eurheartj.a059647 PubMed DOI

Malmivuo J., Plonsey R. (1995). Bioelectromagnetism: principles and applications of bioelectric and biomagnetic fields. USA: Oxford University Press.

Mantas C. J., Abellan J. (2014). Credal-C4. 5: decision tree based on imprecise probabilities to classify noisy data. Expert Syst. Appl. 41, 4625–4637. 10.1016/j.eswa.2014.01.017 DOI

Mantas C. J., Abellán J. (2014). Analysis and extension of decision trees based on imprecise probabilities: application on noisy data. Expert Syst. Appl. 41, 2514–2525. 10.1016/j.eswa.2013.09.050 DOI

McFee R., Parungao A. (1961). An orthogonal lead system for clinical electrocardiography. Am. Heart J. 62, 93–100. 10.1016/0002-8703(61)90488-4 DOI

Mozaffarian D., Benjamin E. J., Go A. S., Arnett D. K., Blaha M. J., Cushman M., et al. (2016). Heart disease and stroke statistics–2016 update: a report from the American heart association. Circulation 133, e38–e360. 10.1161/CIR.0000000000000350 PubMed DOI

Mucherino A., Papajorgji P. J., Pardalos P. M., Mucherino A., Papajorgji P. J., Pardalos P. M. (2009). “K-nearest neighbor classification,” in Data mining in agriculture, 83–106.

Murat F., Yildirim O., Talo M., Demir Y., Tan R.-S., Ciaccio E. J., et al. (2021). Exploring deep features and ECG attributes to detect cardiac rhythm classes. Knowledge-Based Syst. 232, 107473. 10.1016/j.knosys.2021.107473 DOI

Murphy K. P. (2006). Naive bayes classifiers, 18. University of British Columbia, 1–8.

Ng A., Jordan M. (2001). On discriminative vs. generative classifiers: a comparison of logistic regression and naive bayes. Adv. Neural Information Processing Systems 14.

Noble W. S. (2006). What is a support vector machine? Nat. Biotechnology 24, 1565–1567. Publisher: Nature Publishing Group UK London. 10.1038/nbt1206-1565 PubMed DOI

Nocedal J., Wright S. J. (1999). Numerical optimization. Springer.

Panagiotou C., Dima S.-M., Mazomenos E. B., Rosengarten J., Maharatna K., Gialelis J., et al. (2013). Detection of myocardial scar from the vcg using a supervised learning approach. IEEE 2013, 7326–7329. 10.1109/EMBC.2013.6611250 PubMed DOI

Perkins S. J., Monovoukas D., Chopra Z., Kucharski K., Powell C., Vejalla A., et al. (2024). Vectorcardiography predicts heart failure in patients following ST elevation myocardial infarction. Ann. Noninvasive Electrocardiol. 29, e70013. Publisher: Wiley Online Library. 10.1111/anec.70013 PubMed DOI PMC

Prabhakararao E., Dandapat S. (2019). A weighted SVM based approach for automatic detection of posterior myocardial infarction using VCG signals. IEEE, 1–6. 10.1109/ncc.2019.8732238 DOI

Prabhakararao E., Dandapat S. (2020). Automated detection of posterior myocardial infarction from VCG signals using stationary wavelet transform based features. IEEE Sensors Lett. 4, 1–4. 10.1109/lsens.2020.2992760 DOI

Ripoll V. J. R., Wojdel A., Romero E., Ramos P., Brugada J. (2016). ECG assessment based on neural networks with pretraining. Appl. Soft Comput. 49, 399–406. 10.1016/j.asoc.2016.08.013 DOI

Rossello X., Wiegerinck R., Alguersuari J., Bardaji A., Worner F., Sutil M., et al. (2013). Improved diagnostic yielding of acute pericarditis and myocardial ischemia based on changes in QRS complex and QT interval duration. Eur. Heart J. 34, P5576. 10.1093/eurheartj/eht310.p5576 DOI

Schmitt O. H., Simonson E. (1955). Symposium on electrocardiography and vectorcardiography: the present status of vectorcardiography. AMA Archives Internal Medicine 96, 574–590. 10.1001/archinte.1955.00250160016002 PubMed DOI

Shalev-Shwartz S., Singer Y., Srebro N. (2007). Pegasos: primal estimated sub-gradient solver for svm, 807–814.

Sharma L. D., Sunkaria R. K. (2018). Inferior myocardial infarction detection using stationary wavelet transform and machine learning approach. Signal, Image Video Process. 12, 199–206. 10.1007/s11760-017-1146-z DOI

Sharma M., San Tan R., Acharya U. R. (2018). A novel automated diagnostic system for classification of myocardial infarction ECG signals using an optimal biorthogonal filter bank. Comput. Biology Medicine 102, 341–356. 10.1016/j.compbiomed.2018.07.005 PubMed DOI

Śmigiel S., Pałczyński K., Ledziński D. (2021). ECG signal classification using deep learning techniques based on the PTB-XL dataset. Entropy 23, 1121. 10.3390/e23091121 PubMed DOI PMC

Sun Q., Liang C., Chen T., Ji B., Liu R., Wang L., et al. (2022). Early detection of myocardial ischemia in 12-lead ECG using deterministic learning and ensemble learning. Comput. Methods Programs Biomed. 226, 107124. 10.1016/j.cmpb.2022.107124 PubMed DOI

Tadesse G. A., Javed H., Weldemariam K., Liu Y., Liu J., Chen J., et al. (2021). DeepMI: deep multi-lead ECG fusion for identifying myocardial infarction and its occurrence-time. Artif. Intell. Med. 121, 102192. 10.1016/j.artmed.2021.102192 PubMed DOI

Thygesen K., Alpert J. S., White H. D. Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction (2007). Universal definition of myocardial infarction. J. Am. Coll. Cardiol. 50, 2173–2195. 10.1016/j.jacc.2007.09.011 PubMed DOI

Tripathy R., Dandapat S. (2017). Detection of myocardial infarction from vectorcardiogram using relevance vector machine. Signal, Image Video Process. 11, 1139–1146. 10.1007/s11760-017-1068-9 DOI

van Bemmel J. H., Kors J. A., van Herpen G. (1992). Combination of diagnostic classifications from ECG and VCG computer interpretations. J. Electrocardiology 25, 126–130. 10.1016/0022-0736(92)90078-e PubMed DOI

Vondrak J., Penhaker M. (2022a). Review of processing pathological vectorcardiographic records for the detection of heart disease. Front. Physiology 13, 535. 10.3389/fphys.2022.856590 PubMed DOI PMC

Vondrak J., Penhaker M. (2022b). Statistical evaluation of transformation methods accuracy on derived pathological vectorcardiographic leads. PubMed DOI PMC

Vondrak J., Penhaker M., Kubicek J. (2024). Detection of myocardial infarction using analysis of vectorcardiographic loops. Measurement 226, 114094. 10.1016/j.measurement.2023.114094 DOI

Weng J. T.-Y., Lin J.-J., Chen Y.-C., Chang P.-C. (2014). Myocardial infarction classification by morphological feature extraction from big 12-lead ECG data. Springer, 689–699.

Wright S. J., Nowak R. D., Figueiredo M. A. (2009). Sparse reconstruction by separable approximation. IEEE Trans. Signal Processing 57, 2479–2493. 10.1109/tsp.2009.2016892 DOI

Xu W. (2011). Towards optimal one pass large scale learning with averaged stochastic gradient descent. arXiv Preprint arXiv:1107, 2490). 10.48550/arXiv.1107.2490 DOI

Yoon T., Kang D. (2023). Multi-modal stacking ensemble for the diagnosis of cardiovascular diseases. J. Personalized Med. 13, 373. 10.3390/jpm13020373 PubMed DOI PMC

Zeng W., Yuan C. (2023). Myocardial infarction detection using ITD, DWT and deterministic learning based on ECG signals. Cogn. Neurodynamics 17, 941–964. 10.1007/s11571-022-09870-7 PubMed DOI PMC

Zeng W., Yuan J., Yuan C., Wang Q., Liu F., Wang Y. (2020). 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. Artif. Intell. Med. 106, 101848. 10.1016/j.artmed.2020.101848 PubMed DOI

Zhang J., Liu A., Gao M., Chen X., Zhang X., Chen X. (2020). ECG-Based multi-class arrhythmia detection using spatio-temporal attention-based convolutional recurrent neural network. Artif. Intell. Med. 106, 101856. 10.1016/j.artmed.2020.101856 PubMed DOI

Zhang J., Liu M., Xiong P., Du H., Zhang H., Lin F., et al. (2021). A multi-dimensional association information analysis approach to automated detection and localization of myocardial infarction. Eng. Appl. Artif. Intell. 97, 104092. 10.1016/j.engappai.2020.104092 DOI

Zhang J., Liu M., Xiong P., Du H., Yang J., Xu J., et al. (2022a). Automated localization of myocardial infarction from vectorcardiographic PubMed DOI

Zhang X., Wu H., Chen T., Wang G. (2022b). Automatic diagnosis of arrhythmia with electrocardiogram using multiple instance learning: from rhythm annotation to heartbeat prediction. Artif. Intell. Med. 132, 102379. 10.1016/j.artmed.2022.102379 PubMed DOI

Zhao C., Vij A., Malhotra S., Tang J., Tang H., Pienta D., et al. (2021). Automatic extraction and stenosis evaluation of coronary arteries in invasive coronary angiograms. Comput. Biol. Med. 136, 104667. 10.1016/j.compbiomed.2021.104667 PubMed DOI

Zhao X., Zhang J., Gong Y., Xu L., Liu H., Wei S., et al. (2022). Reliable detection of myocardial ischemia using machine learning based on temporal-spatial characteristics of electrocardiogram and vectorcardiogram. Front. Physiology 13, 854191. 10.3389/fphys.2022.854191 PubMed DOI PMC

Zhou Z.-H. (2025). Ensemble methods: foundations and algorithms. CRC Press.

Zhou S., Tan B. (2020). Electrocardiogram soft computing using hybrid deep learning CNN-ELM. Appl. Soft Comput. 86, 105778. 10.1016/j.asoc.2019.105778 DOI

Zhou S. H., Liebman J., Dubin A. M., Gillette P. C., Gregg R. E., Helfenbein E. D., et al. (2001). Using 12-lead ECG and synthesized VCG in detection of right ventricular hypertrophy with terminal right conduction delay PubMed DOI