What is the clinical evidence from this study?

Study design: Other. Population: Cardiac arrhythmias. Intervention: Fine-Tuned CNN (FT-CNN) vs. Fourteen benchmark models including standard CNN, ResNet, LSTM, and GRU. Primary outcome: Classification accuracy.

What does this research mean for the field?

An optimized Fine-Tuned Convolutional Neural Network (FT-CNN) achieves 98.51% accuracy in ECG-based multiclass arrhythmia classification, outperforming standard benchmark models while providing quantitative explainability. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

June 3, 2026Open Access

Integration of Machine Learning Techniques in ECG-Based Multiclass Arrhythmia Classification with Explainability Analysis

Key Result

The Fine-Tuned CNN (FT-CNN) achieved 98.51% accuracy in ECG-based arrhythmia classification, outperforming standard CNN (97.20%), ResNet (96.88%), and other benchmark models.

Structured PICO

Does a Fine-Tuned CNN improve automated ECG-based multiclass arrhythmia classification compared to standard neural architectures?

Population

ECG signals from the MIT-BIH Arrhythmia Database

Intervention

Fine-Tuned CNN (FT-CNN) optimized for ECG signal characteristics with adaptive kernel sizing, multi-faceted regularization, cosine annealing learning rate, and custom loss function

Comparator

Fourteen benchmark models including standard CNN, ResNet, LSTM, GRU, and traditional classifiers

Outcome

Classification accuracysurrogate

A fine-tuned CNN optimized for ECG characteristics achieves 98.51% accuracy in multiclass arrhythmia classification, outperforming standard deep learning architectures while providing model interpretability.

Main Result

Absolute Event Rate: 98.51% vs 97.2%

Limitations

Classification of supraventricular ectopic beats and fusion beats remains challenging

Abstract

Electrocardiogram (ECG) analysis is a cornerstone non-invasive diagnostic technique for detecting cardiac arrhythmias, which remain a leading cause of mortality worldwide. While recent advances in deep learning have significantly improved automated arrhythmia classification, the current literature lacks systematic, fair comparisons of fundamental neural architectures under unified experimental conditions, and very few studies provide model interpretability. This study addresses these gaps by first providing a rigorous comparative analysis of three representative architectures—Artificial Neural Network (ANN), Convolutional Neural Network (CNN), and Residual Network (ResNet)—on the MIT-BIH Arrhythmia Database under identical preprocessing, training, and evaluation protocols. We then propose an efficient Fine-Tuned CNN (FT-CNN) optimized for ECG signal characteristics through adaptive kernel sizing for P-QRS-T morphological extraction, multi-faceted regularization including L2, dropout, and batch normalization, cosine annealing learning rate, and a custom loss function combining weighted categorical cross-entropy with focal loss with gamma equal to 2.0 to address severe class imbalance. The FT-CNN achieves an accuracy of 98.51%, outperforming fourteen benchmark models, including standard CNN with an accuracy of 97.20%, ResNet with 96.88%, LSTM with 96.50%, GRU with 96.30%, and traditional classifiers. Comprehensive ablation studies confirm an improvement of 6.17% over the baseline. Class-wise analysis reveals excellent performance for normal beats with an F1-score of 0.99, ventricular ectopic beats with 0.95, and unknown beats with 0.98, while supraventricular ectopic beats with an F1-score of 0.79 and fusion beats with 0.70 remain challenging. Unlike most prior works, we integrate Grad-CAM and Integrated Gradients for explainability, quantitatively evaluating attribution faithfulness, sanity checks, and noise robustness.