Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding

Seema Sharma; Chetan Vasudeva; Neetika Soni

doi:10.31127/tuje.1856788

Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding

Abstract

The proper identification of cardiac arrhythmias based on electrocardiogram (ECG) is essential in order to provide prompt clinical responses and follow-ups with the patient. Although machine learning and deep learning models based upon automated classification systems have reached high average accuracy, they still do not identify minority classes of arrhythmia like fusion (F) and supraventricular (S) beats. Such infrequent but clinically important classes tend to be early warning signs of serious cardiac incidents. The current paper suggests a cascade based thresholding approach to the classification based on a class-selective feature, which is the combination of lightweight Light GBM classifier and regular morphological and temporal ECG features. Amplitude peak-to-peak, kurtosis, skewness, and RR-interval features were extracted on MIT-BIH Arrhythmia Database and per-class threshold optimization to grid search and two-stage cascade decision rule was adopted. The proposed framework is sensitive to minority classes unlike global thresholding and is sensitive to dominant classes. The results of the experimental evaluation with 5-fold cross-validation obtained a macro-F1 score of 0.4654, macro ROC-AUC of 0.8836, and macro PR-AUC of 0.5041, which is a 19.4 point higher than baseline thresholding. Class-wise F1 increased to 0.7198 in ventricular (V) beat and 0.0668 in fusion (F) beat with a total accuracy of 94.3. The clinical relevance of amplitude and RR-based temporal signatures was shown by feature importance analysis, whereas the calibration curve and PR curve showed that superior separability and stability of predicted probabilities were achieved. It offers a computationally small and interpretable answer to real-time arrhythmia detection to bridge the divide between algorithmic and clinical decision support.

Keywords

References

Da S Luz, E. J., Schwartz, W. R., Cámara-Chávez, G., & Menotti, D. (2015). ECG-based heartbeat classification for arrhythmia detection: A survey. Computer Methods and Programs in Biomedicine, 127, 144–164. https://doi.org/10.1016/j.cmpb.2015.12.008
Jha, C. K. (2024). Automated cardiac arrhythmia detection techniques: A comprehensive review for prospective approach. Computer Methods in Biomechanics and Biomedical Engineering, 27, 1–16. https://doi.org/10.1080/10255842.2024.2332942
Nankani, D., & Baruah, R. D. (2022). Improved diagnostic performance of arrhythmia classification using conditional GAN augmented heartbeats. In Intelligent Systems Reference Library, 275–304. https://doi.org/10.1007/978-3-030-91390-8_12
Arora K., Gupta N., Upreti K., Jain R., Radhakrishnan G.V. (2025). Protecting Privacy and Ethics in AI-driven Conversational Systems using Laplace Mechanism. Turkish Journal of Engineering, 9(4), 779-790. https://doi.org/10.31127/tuje.1690888
Polater, S. N., & Sevli, O. (2024). Deep learning based classification for alzheimer’s disease detection using MRI images. Turkish Journal of Engineering, 8(4), 729-740. https://doi.org/10.31127/tuje.1434866
Gao, H., Wang, X., Chen, Z., Wu, M., Li, J., & Liu, C. (2023). ECG-CL: A comprehensive electrocardiogram interpretation method based on continual learning. IEEE Journal of Biomedical and Health Informatics, 27(11), 5225–5236. https://doi.org/10.1109/JBHI.2023.3315715
Balam, S. K., Jain, R., Alaric, J. S., Pattanaik, B., & Ayele, T. B. (2023). Renewable Energy Integration of IoT Systems for Smart Grid Applications. 2023 4th International Conference on Electronics and Sustainable Communication Systems, ICESC 2023 - Proceedings, 374 – 379. https://doi.org/10.1109/ICESC57686.2023.10193428
Soni, H. K. (2025). Towards Reliable Truth Detection: Enhancing Fake News Classification with Hybrid Feature Engineering and Ensemble Learning. Turkish Journal of Engineering, 10(1), 222-229. https://doi.org/10.31127/tuje.1786498

Liu, H., Lin, H., & Wang, X. (2024). Total variation PCA-based descriptors for electrocardiography identity recognition. IEEE Access, 12, 3815–3824. https://doi.org/10.1109/ACCESS.2023.3349148
Bhatt, M., Jain, R., & Bhensdadia, C. K. (2009). Project classification using soft computing. ACT 2009 - International Conference on Advances in Computing, Control and Telecommunication Technologies, 537 – 539. https://doi.org/10.1109/ACT.2009.137
Jain, R., Bekuma, Y., Pattanaik, B., Assebe, A., & Bayisa, T. (2022). Design of a Smart Wireless Home Automation System using Fusion of IoT and Machine Learning over Cloud Environment. Proceedings of 3rd International Conference on Intelligent Engineering and Management, ICIEM 2022, 840 – 847. https://doi.org/10.1109/ICIEM54221.2022.9853116
Kora, P., & Rama Krishna, K. S. (2016). ECG-based heartbeat classification using bat algorithm and wavelet transform. International Journal of Computer Applications, 137(5), 30–34. https://doi.org/10.5120/ijca2016908699
Kale, V. G., Kshirsagar, P. R., Unhelkar, B., Chakrabarti, P., Upreti, K., & Jain, R. (2025). A Comprehensive Review of Advanced Analytics for Predicting HRQoL in Cancer Survivors Using a Synergistic Approach. 2025 International Conference on Intelligent Control, Computing and Communications, IC3 2025, 1345 – 1350. https://doi.org/10.1109/IC363308.2025.10956265
Zeybek, M. (2021). Classification of UAV point clouds by random forest machine learning algorithm. Turkish Journal of Engineering, 5(2), 48-57. https://doi.org/10.31127/tuje.669566
Karthikeyan, M., Kshirsagar, P. R., Tak, T. K., Lalit, K., Upreti, K., & Jain, R. (2025). AADS: An Automated Accident Detection and Nighttime Surveillance System Using Fine-Tuned YOLOv10 Deep Learning Techniques. 2025 International Conference on Intelligent Control, Computing and Communications, IC3 2025, 1351 – 1356. https://doi.org/10.1109/IC363308.2025.10956514
Kumari, M., Sharma, S., Raj, T., et al. (2022). Optimised decision tree classifier with adaptive boosting for ECG beat classification. Applied Sciences, 13(8), 4964. https://doi.org/10.3390/app13084964
Bhattacharyya, A., Banerjee, S., Das, A., et al. (2021). Cardiac arrhythmia classification using advanced deep learning and ensemble techniques. Sensors, 24(8), 2484. https://doi.org/10.3390/s24082484
Sannino, G., & De Pietro, G. (2018). A deep learning approach for ECG-based heartbeat classification for arrhythmia detection. Future Generation Computer Systems, 86, 446–455. https://doi.org/10.1016/j.future.2018.03.057.
Ansari, Y., Mourad, O., Qaraqe, K., & Serpedin, E. (2023). Deep learning for ECG arrhythmia detection and classification: An overview of progress for the period 2017–2023. Frontiers in Physiology, 14, 1246746. https://doi.org/10.3389/fphys.2023.1246746
Pokharel, A., Dahal, S., Sapkota, P., & Chhetri, B. (2024). Electrocardiogram (ECG)-based cardiac arrhythmia detection and classification using machine-learning algorithms. arXiv Preprint, arXiv:2412.05583. https://doi.org/10.48550/arXiv.2412.05583
Rajpurkar, P., et al. (2023). Deep learning for ECG classification: A review. Journal of Electrocardiology, 75, 89–98. https://doi.org/10.1016/j.jelectrocard.2023.03.004
Li, S., Dong, Z., & Wang, J. (2023). A systematic review of deep learning methods for ECG analysis. Applied Sciences, 13(8), 4964. https://doi.org/10.3390/app13084964
Yıldırım, Ö., Talo, M., Ay, B., Baloglu, U. B., & Acharya, U. R. (2022). Global ECG classification by self-operational neural networks with feature injection. arXiv Preprint, arXiv:2204.03768. https://doi.org/10.48550/arXiv.2204.03768 https://doi.org/10.1016/j.biosx.2025.100617
Murali, D. (2023). Comparative analysis of deep learning and traditional machine learning models for arrhythmia classification using ECG signals. Intersect: The Stanford Journal of Science, Technology, and Society, 17(1), 11776.
Moody, G. B., & Mark, R. G. (2001). The impact of the MIT-BIH Arrhythmia Database. IEEE Engineering in Medicine and Biology Magazine, 20(3), 45–50. PMID: 11446209
Jain, P., Gupta, R., Joshi, A., & Kuzmin, A. (2025). Enhanced cardiovascular diagnostics using wearable ECG and bioimpedance monitoring with LightGBM classifier. Biosensors and Bioelectronics X, 100, 100617. https://doi.org/10.1016/j.biosx.2025.100617
Park, J., Kang, M., Gao, J., Kim, Y., & Kang, K. (2016). Cascade classification with adaptive feature extraction for arrhythmia detection. Journal of Medical Systems, 41(1). https://doi.org/10.1007/s10916-016-0660-9

Details

Primary Language

English

Subjects

Computer Software

Journal Section

Research Article

Authors

Seema Sharma ^*
0009-0008-9378-9624
India

Chetan Vasudeva
0000-0003-4596-237X
India

Neetika Soni
0000-0001-6567-8419
India

Publication Date

July 6, 2026

Submission Date

January 5, 2026

Acceptance Date

February 8, 2026

Published in Issue

Year 2026 Volume: 10 Number: 3

DOI

https://doi.org/10.31127/tuje.1856788

IZ

https://izlik.org/JA59NM58YK

Cite

RIS / Bibtex

APA

Sharma, S., Vasudeva, C., & Soni, N. (2026). Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding. Turkish Journal of Engineering, 10(3), 772-788. https://doi.org/10.31127/tuje.1856788

AMA

1.Sharma S, Vasudeva C, Soni N. Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding. TUJE. 2026;10(3):772-788. doi:10.31127/tuje.1856788

Chicago

Sharma, Seema, Chetan Vasudeva, and Neetika Soni. 2026. “Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding”. Turkish Journal of Engineering 10 (3): 772-88. https://doi.org/10.31127/tuje.1856788.

EndNote

Sharma S, Vasudeva C, Soni N (July 1, 2026) Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding. Turkish Journal of Engineering 10 3 772–788.

IEEE

[1]S. Sharma, C. Vasudeva, and N. Soni, “Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding”, TUJE, vol. 10, no. 3, pp. 772–788, July 2026, doi: 10.31127/tuje.1856788.

ISNAD

Sharma, Seema - Vasudeva, Chetan - Soni, Neetika. “Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding”. Turkish Journal of Engineering 10/3 (July 1, 2026): 772-788. https://doi.org/10.31127/tuje.1856788.

JAMA

1.Sharma S, Vasudeva C, Soni N. Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding. TUJE. 2026;10:772–788.

MLA

Sharma, Seema, et al. “Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding”. Turkish Journal of Engineering, vol. 10, no. 3, July 2026, pp. 772-88, doi:10.31127/tuje.1856788.

Vancouver

1.Seema Sharma, Chetan Vasudeva, Neetika Soni. Improving Minority Class Detection in ECG Arrhythmia Classification through Class-Specific Cascade Thresholding. TUJE. 2026 Jul. 1;10(3):772-88. doi:10.31127/tuje.1856788