Sergio González-Cabeza
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sergio González-Cabeza has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Marian Diaz-Vicente
Complutense University of Madrid, Madrid, Spain and Vision2.ai, Madrid, Spain
4 shared publications - Mario Sanz-Guerrero
Johannes Gutenberg University Mainz, Mainz, Germany
4 shared publications - Gonzalo Luis Alonso Salinas
Department of Cardiology, Hospital Universitario de Navarra, 31008 Pamplona, Spain
3 shared publications - Joaquín Recas
Universidad Complutense de Madrid
3 shared publications - Luis Piñuel
Computer Architecture and Automation Department, Complutense University of Madrid, Madrid, Spain
3 shared publications - Mauro Luis Buelga Suárez2 shared publications
Latest Publications
- Open Access
- 30 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Explainable Deep Learning for Multi-Label ECG Classification: Clinical Validation and Analysis of Signal Transformations
Deep learning models for automated electrocardiogram (ECG) interpretation have achieved high diagnostic accuracy, yet clinical adoption remains limited due to a lack of explainability and frequent reliance on private datasets and closed-source implementations. We present a clinically validated explainable AI (XAI) framework for multi-label ECG classification that shifts the focus from raw performance toward explainability, reproducibility, and clinical usability. Our framework is developed exclusively using publicly available resources—the PTB-XL database (21,799 ECGs across five diagnostic categories), an open-source model (Ribeiro et al. [14]), and open-source libraries. The model achieves an F1-score of 0.738 while providing gradient-based saliency maps evaluated by practicing cardiologists. We further compare time-domain and time–frequency representations (STFT and CWT). Direct time-series analysis achieves superior performance compared to transformed inputs with this model, while also requiring substantially lower computational cost. A key finding from a two-round validation process with clinicians is that explanation usefulness depends not only on the attribution method but also on presentation format: cardiologists could not reliably interpret saliency overlays until they were rendered on standard ECG grid paper used in daily clinical practice. Our contributions are: (1) a fully reproducible XAI pipeline with documented iterative clinical validation, (2) evidence that visualization format is integral to explanation quality in ECG analysis, (3) a GAM-compliant extension of the open-source ecg_plot library, and (4) a controlled analysis of signal transformation trade-offs.
- 30 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessExplainable Deep Learning for Multi-Label ECG Classification: Clinical Validation and Analysis of Signal Transformations
Deep learning models for automated electrocardiogram (ECG) interpretation have achieved high diagnostic accuracy, yet clinical adoption remains limited due to a lack of explainability and frequent reliance on private datasets and closed-source implementations. We present a clinically validated explainable AI (XAI) framework for multi-label ECG classification that shifts the focus from raw performance toward explainability, reproducibility, and clinical usability. Our framework is developed exclusively using publicly available resources—the PTB-XL database (21,799 ECGs across five diagnostic categories), an open-source model (Ribeiro et al. [14]), and open-source libraries. The model achieves an F1-score of 0.738 while providing gradient-based saliency maps evaluated by practicing cardiologists. We further compare time-domain and time–frequency representations (STFT and CWT). Direct time-series analysis achieves superior performance compared to transformed inputs with this model, while also requiring substantially lower computational cost. A key finding from a two-round validation process with clinicians is that explanation usefulness depends not only on the attribution method but also on presentation format: cardiologists could not reliably interpret saliency overlays until they were rendered on standard ECG grid paper used in daily clinical practice. Our contributions are: (1) a fully reproducible XAI pipeline with documented iterative clinical validation, (2) evidence that visualization format is integral to explanation quality in ECG analysis, (3) a GAM-compliant extension of the open-source ecg_plot library, and (4) a controlled analysis of signal transformation trade-offs.
- Open Access
- 19 February 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 119
Peak Attention U-Net: Enhancing ECG delineation with attention
- 19 February 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 119
Open AccessPeak Attention U-Net: Enhancing ECG delineation with attention
- Open Access
- 16 December 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 115
Performance evaluation of smartwatches: Can they match clinical standards for ECG analysis?
- 16 December 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 115
Open AccessPerformance evaluation of smartwatches: Can they match clinical standards for ECG analysis?
- Open Access
- 27 August 2025
- Published by IOP Publishing in Medical Engineering & Physics
- Vol. 145 (1) , 104419
Reducing leads, enhancing wearable practicality: A comparative study of 3-lead vs. 12-lead ECG classification
Inspired by recent advances in clinical research and the growing adoption of wearable ECG devices, this study explores the feasibility of using reduced-lead ECGs for automated detection of heart anomalies using deep learning, providing a more accessible and cost-effective alternative to traditional 12-lead ECGs. This research adapts and evaluates a state-of-the-art 12-lead deep learning model (from Ribeiro et al. [1]) for 3-lead configurations. The 12-lead ECG model architecture was trained from scratch on the public database PTB-XL. It was then modified to use 3 leads by only changing the input layer. Despite a 75% reduction in input data, the 3-lead model showed only a subtle 3% performance drop. To address this gap, the 3-lead model was further optimized using a novel strategy that combines transfer learning and a One-vs-All classification approach. Using PTB-XL's five-class setup (normal vs. four pathologies: myocardial infarction, ST/T change, conduction disturbance, and hypertrophy), we report the micro-averaged F1-score across all test samples. The new optimized 3-lead model achieves a global (micro-averaged) F1-score of 77% (vs. 78% for the 12-lead model). These findings highlight the potential of simplified and cost-effective reduced-lead classification models to deliver near-equivalent diagnostic accuracy. This advancement could democratize access to early cardiac diagnostics, particularly in resource-limited settings.
- 27 August 2025
- Published by IOP Publishing in Medical Engineering & Physics
- Vol. 145 (1) , 104419
Open AccessReducing leads, enhancing wearable practicality: A comparative study of 3-lead vs. 12-lead ECG classification
Inspired by recent advances in clinical research and the growing adoption of wearable ECG devices, this study explores the feasibility of using reduced-lead ECGs for automated detection of heart anomalies using deep learning, providing a more accessible and cost-effective alternative to traditional 12-lead ECGs. This research adapts and evaluates a state-of-the-art 12-lead deep learning model (from Ribeiro et al. [1]) for 3-lead configurations. The 12-lead ECG model architecture was trained from scratch on the public database PTB-XL. It was then modified to use 3 leads by only changing the input layer. Despite a 75% reduction in input data, the 3-lead model showed only a subtle 3% performance drop. To address this gap, the 3-lead model was further optimized using a novel strategy that combines transfer learning and a One-vs-All classification approach. Using PTB-XL's five-class setup (normal vs. four pathologies: myocardial infarction, ST/T change, conduction disturbance, and hypertrophy), we report the micro-averaged F1-score across all test samples. The new optimized 3-lead model achieves a global (micro-averaged) F1-score of 77% (vs. 78% for the 12-lead model). These findings highlight the potential of simplified and cost-effective reduced-lead classification models to deliver near-equivalent diagnostic accuracy. This advancement could democratize access to early cardiac diagnostics, particularly in resource-limited settings.