QTcNet: a deep learning model for direct heart rate corrected QT interval estimation
Europace · 2025
DOI: 10.1093/europace/euaf274 ↗Kurz gefasst
Validierungsstudie in der Zeitschrift Europace, 2025. Verfasst von Lucas Plagwitz und 8 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Medical Informatics, University of Münster, Albert-Schweitzer-Campus 1/Building A11, Münster 48149, Germany. Hauptthemen laut MEDLINE: Deep Learning, Electrocardiography, Heart Rate, Signal Processing, Computer-Assisted. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Center for Clinical Research.
Abstract (Originalsprache)
Aims
Automated QTc measurements from commercial ECG systems often diverge from expert readings. We developed QTcNet, a deep learning model trained and validated on multiple large ECG datasets to improve automated QTc measurement accuracy. Methods and results QTcNet employs a regression-based convolutional neural network architecture. It was trained on 120 300 algorithm-labelled ECGs (60 150 from an internal hospital cohort and 60 150 from the MIMIC-IV dataset) after correction for a vendor-specific +15 ms bias. Performance was evaluated against expert QTc measurements in three independent datasets: PTB Diagnostic ECG Database (n = 100 ECGs in validation set), QTcMS (n = 210), and ECGRDVQ (n = 5219). The effect of fine-tuning on cardiologist-annotated ECGs was tested in the PTB database (n = 449 in fine-tuning set). Model explainability analyses were performed with Integrated Gradient maps. QTcNet reduced cross-cohort mean absolute error (MAE) from 23.4 to 13.4 ms and root mean square error (RMSE) from 40.1 to 22.1 ms, almost halving large (>50 ms) outliers. Fine-tuning only reduced errors in the PTB dataset but did not improve cross-cohort performance. Integrated Gradient maps confirmed that the model concentrated on QRS onset and T wave offset, supporting physiological plausibility.
Conclusion
QTcNet, trained on large-scale algorithmically labelled data, consistently outperformed conventional algorithms across three independent, external validation datasets. Fine-tuning of QTcNet may adapt the model to the characteristics of specific cohorts but reduces external validity in other cohorts. We openly release the full model and code, along with a ready-to-use online implementation at https://qtcnet.uni-muenster.de, facilitating further research and community-driven improvement.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Medical Informatics, University of Münster, Albert-Schweitzer-Campus 1/Building A11, Münster 48149, Germany.
- Clinic for Cardiology II: Electrophysiology, University Hospital Münster, Albert-Schweitzer-Campus 1/Building A1, Münster 48149, Germany.
- Institute of Medical Data Science, Otto-von-Guericke University Magdeburg, Leipziger Str. 44/Building 2, Magdeburg 39120, Germany.
ORCID-Profile: Lucas Plagwitz, Florian Doldi, Maxim Zotov, Lucas Bickmann, Dominik Heider, Julian Varghese, Lars Eckardt, Antonius Büscher
Förderung
- Center for Clinical Research
Zitieren
APA
Plagwitz, L., Doldi, F., Magerfleisch, J., Zotov, M., Bickmann, L., Heider, D., Varghese, J., Eckardt, L., & Büscher, A. (2025). QTcNet: a deep learning model for direct heart rate corrected QT interval estimation. Europace. https://doi.org/10.1093/europace/euaf274
Vancouver
Plagwitz L, Doldi F, Magerfleisch J, Zotov M, Bickmann L, Heider D, et al. QTcNet: a deep learning model for direct heart rate corrected QT interval estimation. Europace. 2025. doi:10.1093/europace/euaf274
BibTeX
@article{Plagwitz2025,
author = {Plagwitz, Lucas and Doldi, Florian and Magerfleisch, Jannes and Zotov, Maxim and Bickmann, Lucas and Heider, Dominik and Varghese, Julian and Eckardt, Lars and Büscher, Antonius},
title = {QTcNet: a deep learning model for direct heart rate corrected QT interval estimation},
journal = {Europace},
year = {2025},
doi = {10.1093/europace/euaf274}
}
EUVIMED