UCTECG-Net: Uncertainty-aware Convolution Transformer ECG Network for Arrhythmia Detection

TL;DR

This paper introduces UCTECG-Net, a hybrid deep learning model combining convolutions and Transformers for ECG classification, which also provides uncertainty estimates to enhance reliability in critical applications.

Contribution

The paper presents a novel uncertainty-aware hybrid architecture for ECG classification that outperforms existing models and offers improved reliability through integrated uncertainty quantification methods.

Findings

UCTECG-Net achieves up to 98.58% accuracy on MIT-BIH dataset.

Uncertainty quantification methods improve reliability of predictions.

Ensemble methods provide the most reliable uncertainty estimates.

Abstract

Deep learning has improved automated electrocardiogram (ECG) classification, but limited insight into prediction reliability hinders its use in safety-critical settings. This paper proposes UCTECG-Net, an uncertainty-aware hybrid architecture that combines one-dimensional convolutions and Transformer encoders to process raw ECG signals and their spectrograms jointly. Evaluated on the MIT-BIH Arrhythmia and PTB Diagnostic datasets, UCTECG-Net outperforms LSTM, CNN1D, and Transformer baselines in terms of accuracy, precision, recall and F1 score, achieving up to 98.58% accuracy on MIT-BIH and 99.14% on PTB. To assess predictive reliability, we integrate three uncertainty quantification methods (Monte Carlo Dropout, Deep Ensembles, and Ensemble Monte Carlo Dropout) into all models and analyze their behavior using an uncertainty-aware confusion matrix and derived metrics. The results show that UCTECG-Net, particularly with Ensemble or EMCD, provides more reliable and better-aligned uncertainty estimates than competing architectures, offering a stronger basis for risk-aware ECG decision support.