Comparison of Dimensionality Reduction Approaches and Logistic Regression for ECG Classification
Source abstract
Abstract
This study aims to analyze electrocardiogram (ECG) data for the classification of five cardiac rhythms: sinus bradycardia (SB), sinus rhythm (SR), atrial fibrillation (AFIB), supraventricular tachycardia (SVT), and sinus tachycardia (ST). While SR is considered normal, the other four represent types of cardiac arrhythmias. A range of methods is utilized, including the supervised learning technique K-Nearest Neighbors (KNNs), combined with dimensionality reduction approaches such as Principal Component Analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP), a modern method based in Riemannian topology. Additionally, logistic regression was applied using both maximum likelihood and Bayesian methods, with two distinct prior distributions: an informative normal prior and a non-informative Jeffreys prior. Performance was assessed using evaluation metrics such as positive predictive value (PPV), negative predictive value (NPV), specificity, sensitivity, accuracy, and F1-score. Ultimately, the UMAP-KNN method demonstrated the best overall performance.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: ECG Monitoring and Analysis · Non-Invasive Vital Sign Monitoring · EEG and Brain-Computer Interfaces
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mathematics and Statistics, Université de Moncton, Moncton, NB E1A 3E9, Canada
Centre de Formation Médicale and École de Psychologie, Université de Moncton, Moncton, NB E1A 7R1, Canada
Funding
ResearchNB—Strategic Initiative Grant · AI Pre-Voucher New Brunswick Innovation Fund
References (22)
Show source-provided references
1. Enhanced electrocardiogram machine learning-based classification with emphasis on fusion and unknown heartbeat classes
2. Atrial fibrillation in critical illness: Epidemiology and clinical significance
3. A review on automatic detection and classification of arrhythmias using ECG signals
4. Mantravadi, R., Kim, H., and Moon, J. (2024). Explainable AI for ECG classification: Challenges and future directions. Comput. Biol. Med., 168.
5. A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients10.1038/s41597-020-0386-x ↗
6. Aziz, A., Al-Ali, A.R., and Al-Nashash, H. (2021). Deep learning for electrocardiogram (ECG) analysis: A review. Biomed. Signal Process. Control, 68.
7. A survey of deep learning methods for ECG classification
8. A novel wavelet sequence based on deep bidirectional LSTM network model for ECG signal classification10.1016/j.compbiomed.2018.03.016 ↗
9. Hastie, T., Tibshirani, R., Friedman, J.H., and Friedman, J.H. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Springer.10.1007/978-0-387-84858-7 ↗
10. McInnes, L., Healy, J., and Melville, J. (2018). UMAP: Uniform manifold approximation and projection for dimension reduction. arXiv.10.21105/joss.00861 ↗
11. Application of Aligned-UMAP to longitudinal biomedical studies10.1016/j.patter.2023.100741 ↗
12. Dimensionality Reduction Using PCA and t-SNE for Analysis and Prediction of Cardiovascular Disease10.1088/1742-6596/2161/1/012003 ↗
13. Meehan, C., Ebrahimian, J., Moore, W., and Meehan, S. (2025, June 11). Uniform Manifold Approximation and Projection (UMAP) [MATLAB Code]. 2023. MATLAB Central File Exchange. Available online: https://www.mathworks.com/matlabcentral/fileexchange/71902.
14. Kononenko, I., and Kukar, M. (2007). Chapter 10—Statistical Learning. Machine Learning and Data Mining, Woodhead Publishing.10.1533/9780857099440 ↗
15. Saporta, G. (2006). Probabilité, Analyse des Données et Statistique, Éditions Technip. [2nd ed.].
16. MCCullagh, P. (2019). Generalized Linear Models, Routledge.10.1201/9780203753736 ↗
17. Properties and Implementation of Jeffreys’s Prior in Binomial Regression Models10.1198/016214508000000779 ↗
18. Robert, C.P. (2007). The Bayesian Choice: From Decision-Theoretic Foundations to Computational Implementation, Springer. [2nd ed.].
19. Aziz, S., Haider, S.U., Yousaf, M.H., Rauf, H.A., and Rehman, A. (2021). Automated cardiac arrhythmia classification using deep learning techniques. Sci. Rep., 11.
20. Arrhythmia classification using residual attention CNN on 12-lead ECGs10.1016/j.heliyon.2024.e26787 ↗
21. Ozpolat, E., and Yildiz, O. (2023). A temporal CNN with dilated convolutions for ECG arrhythmia classification. Diagnostics, 13.10.3390/diagnostics13061099 ↗
22. Yildirim, O., Talo, M., Baloglu, U.B., Aydin, G., and Acharya, U.R. (2020). Arrhythmia detection using deep bidirectional LSTM network. Comput. Methods Programs Biomed., 185.