Pediatric pancreas segmentation from MRI scans with deep learning
Pancreatology · 2025
DOI: 10.1016/j.pan.2025.06.006 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pancreatology, 2025. Verfasst von Elif Keles und 13 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Northwestern University, IL, USA. Electronic address: elif.keles@northwestern.edu. Hauptthemen laut MEDLINE: Deep Learning, Magnetic Resonance Imaging, Pancreas, Pancreatitis. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NIDDK NIH HHS, National Institute of Diabetes and Digestive and Kidney Diseases Division of Diabetes Endocrinology and Metabolic Diseases.
Abstract (Originalsprache)
OBJECTIVE: Our study aimed to evaluate and validate PanSegNet, a deep learning (DL) algorithm for pediatric pancreas segmentation on MRI in children with acute pancreatitis (AP), chronic pancreatitis (CP), and healthy controls. METHODS: With IRB approval, we retrospectively collected 84 MRI scans (1.5T/3T Siemens Aera/Verio) from children aged 2-19 years at Gazi University (2015-2024). The dataset includes healthy children as well as patients diagnosed with AP or CP based on clinical criteria. Pediatric and general radiologists manually segmented the pancreas, then confirmed by a senior pediatric radiologist. PanSegNet-generated segmentations were assessed using Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff distance (HD95). Cohen's kappa measured observer agreement. RESULTS: Pancreas MRI T2W scans were obtained from 42 children with AP/CP (mean age: 11.73 ± 3.9 years) and 42 healthy children (mean age: 11.19 ± 4.88 years). PanSegNet achieved DSC scores of 88 % (controls), 81 % (AP), and 80 % (CP), with HD95 values of 3.98 mm (controls), 9.85 mm (AP), and 15.67 mm (CP). Inter-observer kappa was 0.86 (controls), 0.82 (pancreatitis), and intra-observer agreement reached 0.88 and 0.81. Strong agreement was observed between automated and manual volumes (R2 = 0.85 in controls, 0.77 in diseased), demonstrating clinical reliability. CONCLUSION: PanSegNet represents the first validated deep learning solution for pancreatic MRI segmentation, achieving expert-level performance across healthy and diseased states. This tool, algorithm, along with our annotated dataset, are freely available on GitHub and OSF, advancing accessible, radiation-free pediatric pancreatic imaging and fostering collaborative research in this underserved domain.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Northwestern University, IL, USA. Electronic address: elif.keles@northwestern.edu.
- Department of Radiology, Gazi University Faculty of Medicine, Ankara, Türkiye.
- Department of Radiology, Northwestern University, IL, USA.
- Division of Gastroenterology and Hepatology, University of Illinois at Chicago, Chicago, IL, USA.
- Division of Pediatrics, Gastroenterology, Hepatology and Nutrition, Washington University School of Medicine in St. Louis, MO, USA.
- Division of Pediatric Gastroenterology, Hepatology, Pancreatology and Nutrition, University of Iowa, Carver College of Medicine, Iowa, IA, USA.
Förderung
- NCI NIH HHS
- NIDDK NIH HHS
- National Institute of Diabetes and Digestive and Kidney Diseases Division of Diabetes Endocrinology and Metabolic Diseases
- National Institutes of Health
- National Cancer Institute Center for Biomedical Informatics and Information Technology
Zitieren
APA
Keles, E., Yazol, M., Durak, G., Hong, Z., Aktas, H. E., Zhang, Z., Peng, L., Susladkar, O., Guzelyel, N., Boyunaga, O. L., Yazici, C., Lowe, M., Uc, A., & Bagci, U. (2025). Pediatric pancreas segmentation from MRI scans with deep learning. Pancreatology. https://doi.org/10.1016/j.pan.2025.06.006
Vancouver
Keles E, Yazol M, Durak G, Hong Z, Aktas HE, Zhang Z, et al. Pediatric pancreas segmentation from MRI scans with deep learning. Pancreatology. 2025. doi:10.1016/j.pan.2025.06.006
BibTeX
@article{Keles2025,
author = {Keles, Elif and Yazol, Merve and Durak, Gorkem and Hong, Ziliang and Aktas, Halil Ertugrul and Zhang, Zheyuan and Peng, Linkai and Susladkar, Onkar and Guzelyel, Necati and Boyunaga, Oznur Leman and Yazici, Cemal and Lowe, Mark and Uc, Aliye and Bagci, Ulas},
title = {Pediatric pancreas segmentation from MRI scans with deep learning},
journal = {Pancreatology},
year = {2025},
doi = {10.1016/j.pan.2025.06.006}
}
EUVIMED