Interpretable Deep Learning Models for Arrhythmia Classification Based on ECG Signals Using PTB-X Dataset
Diagnostics · 04.08.2025 · 15 (15) · 1950
DOI: 10.3390/diagnostics15151950 ↗Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 04.08.2025. Verfasst von Ahmed E Mansour Atwa und 5 weiteren Autor:innen. Beteiligte Einrichtung: Electronics and Communication Department, College of Engineering and Computer Science, Mustaqbal University, Buraydah 51411, Saudi Arabia. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Scientific Research at Northern Border 749 University, Arar, KSA.
Abstract (Originalsprache)
Background/Objectives: Automatic classification of ECG signal arrhythmias plays a vital role in early cardiovascular diagnostics by enabling prompt detection of life-threatening conditions. Manual ECG interpretation is labor-intensive and susceptible to errors, highlighting the demand for automated, scalable approaches. Deep learning (DL) methods are effective in ECG analysis due to their ability to learn complex patterns from raw signals.
Methods
This study introduces two models: a custom convolutional neural network (CNN) with a dual-branch architecture for processing ECG signals and demographic data (e.g., age, gender), and a modified VGG16 model adapted for multi-branch input. Using the PTB-XL dataset, a widely adopted large-scale ECG database with over 20,000 recordings, the models were evaluated on binary, multiclass, and subclass classification tasks across 2, 5, 10, and 15 disease categories. Advanced preprocessing techniques, combined with demographic features, significantly enhanced performance.
Results
The CNN model achieved up to 97.78% accuracy in binary classification and 79.7% in multiclass tasks, outperforming the VGG16 model (97.38% and 76.53%, respectively) and state-of-the-art benchmarks like CNN-LSTM and CNN entropy features. This study also emphasizes interpretability, providing lead-specific insights into ECG contributions to promote clinical transparency.
Conclusions
These results confirm the models’ potential for accurate, explainable arrhythmia detection and their applicability in real-world healthcare diagnostics.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Electronics and Communication Department, College of Engineering and Computer Science, Mustaqbal University, Buraydah 51411, Saudi Arabia.
- Department of Computer Science, College of Computer Science and Engineering, Taibah University, Yanbu 966144, Saudi Arabia.
- Computer Science Department, Faculty of Science, University of Tanta, Tanta 31527, Gharbia, Egypt.
- Information Technology Department, Faculty of Computers and Information, Menoufia University, Shibin El Kom 6131567, Egypt.
- Faculty of Medicine Kasr Al-Ainy, Cairo University, Cairo 11562, Egypt.
- Computer Science Department, College of Science, Northern Border University, Arar 91431, Saudi Arabia.
ORCID-Profile: El-Sayed Atlam, Mohamed Ahmed Atwa, Elsaid Md Abdelrahim, Ali I Siam
Förderung
- Deanship of Scientific Research at Northern Border 749 University, Arar, KSA
Zitieren
APA
Atwa, A. E. M., Atlam, E. S., Ahmed, A., Atwa, M. A., Abdelrahim, E. M., & Siam, A. I. (2025). Interpretable Deep Learning Models for Arrhythmia Classification Based on ECG Signals Using PTB-X Dataset. Diagnostics, 15(15), 1950. https://doi.org/10.3390/diagnostics15151950
Vancouver
Atwa AEM, Atlam ES, Ahmed A, Atwa MA, Abdelrahim EM, Siam AI. Interpretable Deep Learning Models for Arrhythmia Classification Based on ECG Signals Using PTB-X Dataset. Diagnostics. 2025;15(15):1950. doi:10.3390/diagnostics15151950
BibTeX
@article{Atwa2025,
author = {Atwa, Ahmed E Mansour and Atlam, El-Sayed and Ahmed, Ali and Atwa, Mohamed Ahmed and Abdelrahim, Elsaid Md and Siam, Ali I},
title = {Interpretable Deep Learning Models for Arrhythmia Classification Based on ECG Signals Using PTB-X Dataset},
journal = {Diagnostics},
year = {2025},
volume = {15},
number = {15},
pages = {1950},
doi = {10.3390/diagnostics15151950}
}
EUVIMED