Deep learning-based detection of acute pancreatitis on abdominal contrast-enhanced CT
European Radiology Experimental · 31.07.2026 · 10 (1)
DOI: 10.1186/s41747-026-00775-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Radiology Experimental, online erschienen am 31.07.2026. Verfasst von Oleksandra Seidel und 11 weiteren Autor:innen. Beteiligte Einrichtung: Clinic for Diagnostic and Interventional Radiology, University of Bonn, University Hospital Bonn, Bonn, Germany. Hauptthemen laut MEDLINE: Pancreatitis, Deep Learning, Tomography, X-Ray Computed, Contrast Media. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: RACOON.
Abstract (Originalsprache)
Objective
We developed and evaluated a deep learning (DL) model for image-based detection of acute pancreatitis (AP) on abdominal contrast-enhanced CT (CECT).
Methods
A total of 552 patients from two university centers (January 2010–January 2026) were included. The internal dataset comprised 207 patients with clinically and radiologically confirmed AP (499 scans) and 250 control patients with suspected AP (368 scans). An independent external validation cohort included 95 patients. Convolutional neural network–based models were trained using monophasic and biphasic CECT data. The final model was evaluated on a 20% patient-level hold-out test set from the internal cohort and on the external cohort. Performance was assessed using the F1 score and area under the receiver operating characteristic curve (AUROC).
Results
A single-input multiphase model incorporating arterial and portal venous phase scans achieved the best performance, with ensembling applied for biphasic studies. On the internal hold-out test set ( n = 116), the model achieved an F1 score of 0.83 (95% confidence interval 0.75–0.89) and an AUROC of 0.89 (0.82–0.95). Performance remained robust on external validation ( n = 95), with an AUROC of 0.99 (0.96–1.00) and an F1 score of 0.92 (0.86–0.97).
Conclusion
DL enabled accurate CECT-based identification of AP in this retrospective multicenter cohort, with performance maintained in an independent external dataset. Prospective validation using broader and independently adjudicated clinical populations remains necessary. Relevance statement The model showed promising performance for CECT-based acute pancreatitis detection but was not designed or tested as a triage system. Key Points Diagnostic uncertainty in acute pancreatitis often arises from nonspecific abdominal symptoms and inter-reader variability in CECT interpretation. The DL model achieved high internal accuracy (AUROC 0.89) and maintained robust performance in an independent external validation cohort (AUROC 0.99).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Clinic for Diagnostic and Interventional Radiology, University of Bonn, University Hospital Bonn, Bonn, Germany.
- Clinic for Radiation Therapy and Radiooncology, University of Bonn, University Hospital Bonn, Bonn, Germany.
- Clinic for Neuroradiology, University of Bonn, University Hospital Bonn, Bonn, Germany.
- Department of Diagnostic and Interventional Radiology, University Medical Center of the Johannes Gutenberg-University, Mainz, Germany.
- Clinic for Diagnostic and Interventional Radiology, University of Bonn, University Hospital Bonn, Bonn, Germany. Dmitrij.kravchenko@ukbonn.de.
Förderung
- RACOON
Zitieren
APA
Seidel, O., Theis, M., Nowak, S., Garajová, L., Wulff, B., Block, W., Müller, L., Emrich, T., Luetkens, J. A., Hadizadeh, D. R., Sprinkart, A. M., & Kravchenko, D. (2026). Deep learning-based detection of acute pancreatitis on abdominal contrast-enhanced CT. European Radiology Experimental, 10(1). https://doi.org/10.1186/s41747-026-00775-2
Vancouver
Seidel O, Theis M, Nowak S, Garajová L, Wulff B, Block W, et al. Deep learning-based detection of acute pancreatitis on abdominal contrast-enhanced CT. European Radiology Experimental. 2026;10(1). doi:10.1186/s41747-026-00775-2
BibTeX
@article{Seidel2026,
author = {Seidel, Oleksandra and Theis, Maike and Nowak, Sebastian and Garajová, Laura and Wulff, Benjamin and Block, Wolfgang and Müller, Lukas and Emrich, Tilmann and Luetkens, Julian A and Hadizadeh, Dariusch R and Sprinkart, Alois M and Kravchenko, Dmitrij},
title = {Deep learning-based detection of acute pancreatitis on abdominal contrast-enhanced CT},
journal = {European Radiology Experimental},
year = {2026},
volume = {10},
number = {1},
doi = {10.1186/s41747-026-00775-2}
}
EUVIMED