Technology Category
Machine Learning: Multi-class/Multi-label Learning & Extreme ClassificationComputer Vision: Multi-modal VisionHumans and AI: Explainable AI (XAI) for Human Understanding
EXGnet: a single-lead explainable-AI guided multiresolution network with train-only quantitative features for trustworthy ECG arrhythmia classification
🤖 AI Summary
To address the poor interpretability and low clinical trustworthiness of arrhythmia classification models for single-lead portable ECG devices, this paper proposes an XAI-guided multi-resolution deep learning framework. Methodologically: (1) a multi-scale convolutional architecture is designed to jointly extract time-frequency features; (2) Grad-CAM is integrated into the training phase to dynamically guide feature learning and generate interpretable diagnostic attribution maps; (3) quantitative ECG features (e.g., RR intervals, QRS width) are embedded exclusively during training—imposing no inference overhead. The key innovations are the first-ever “XAI-guided multi-resolution architecture” and a “train-only quantitative feature mechanism.” Evaluated on the Chapman and Ningbo datasets, the model achieves five-fold cross-validated average accuracies of 98.76% and 96.93%, and F1-scores of 97.91% and 95.53%, respectively—significantly outperforming existing single-lead approaches while ensuring both high accuracy and clinical interpretability.
Technology Category
Application Category
📝 Abstract
Background: Deep learning has significantly advanced ECG arrhythmia classification, enabling high accuracy in detecting various cardiac conditions. The use of single-lead ECG systems is crucial for portable devices, as they offer convenience and accessibility for continuous monitoring in diverse settings. However, the interpretability and reliability of deep learning models in clinical applications poses challenges due to their black-box nature. Methods: To address these challenges, we propose EXGnet, a single-lead, trustworthy ECG arrhythmia classification network that integrates multiresolution feature extraction with Explainable Artificial Intelligence (XAI) guidance and train only quantitative features. Results: Trained on two public datasets, including Chapman and Ningbo, EXGnet demonstrates superior performance through key metrics such as Accuracy, F1-score, Sensitivity, and Specificity. The proposed method achieved average five fold accuracy of 98.762%, and 96.932% and average F1-score of 97.910%, and 95.527% on the Chapman and Ningbo datasets, respectively. Conclusions: By employing XAI techniques, specifically Grad-CAM, the model provides visual insights into the relevant ECG segments it analyzes, thereby enhancing clinician trust in its predictions. While quantitative features further improve classification performance, they are not required during testing, making the model suitable for real-world applications. Overall, EXGnet not only achieves better classification accuracy but also addresses the critical need for interpretability in deep learning, facilitating broader adoption in portable ECG monitoring.
Problem
Research questions and friction points this paper is trying to address.
Enhances interpretability of ECG arrhythmia classification models
Improves reliability for single-lead portable ECG monitoring
Integrates explainable AI with multiresolution feature extraction
Innovation
Methods, ideas, or system contributions that make the work stand out.
Single-lead ECG with XAI for trust
Multiresolution feature extraction method
Train-only quantitative features enhance accuracy
🔎 Similar Papers
No similar papers found.
