Zero‐Shot Self‐Supervised Learning of Single Breath‐Hold Magnetic Resonance Cholangiopancreatography ( MRCP ) Reconstruction
Magnetic Resonance in Medicine · 2026-10 · 96 (4) · 1666-1681
DOI: 10.1002/mrm.70467 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 12.06.2026. Verfasst von Jinho Kim und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany. Hauptthemen laut MEDLINE: Cholangiopancreatography, Magnetic Resonance, Breath Holding, Image Processing, Computer-Assisted, Supervised Machine Learning. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft, Friedrich-Alexadner-Universität Erlangen-Nürnberg, Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung - BMBF).
Abstract (Originalsprache)
Purpose
To investigate the feasibility of zero‐shot self‐supervised learning reconstruction for reducing breath‐hold times in magnetic resonance cholangiopancreatography (MRCP).
Methods
Breath‐hold MRCP was acquired from 11 healthy volunteers on 3T scanners using an incoherent k ‐space sampling pattern, leading to a 14‐s acquisition time and an acceleration factor of R = 25. Zero‐shot reconstruction was compared with parallel imaging of respiratory‐triggered MRCP (338 s, R = 3) and compressed sensing reconstruction. For two volunteers, breath‐hold scans (40 s, R = 6) were additionally acquired and retrospectively undersampled to R = 25 to compute peak signal‐to‐noise ratio (PSNR). To address long zero‐shot training time, the full stages of the zero‐shot learning were divided into two parts to reduce backpropagation depth during training: (1) frozen stages initialized with ‐stage pretrained network and (2) trainable stages initialized either randomly or ‐stage pretrained network. Efficiency of our approach was assessed by varying initialization strategies and the number of trainable stages using the retrospectively undersampled data.
Results
Zero‐shot reconstruction significantly improved visual image quality over compressed sensing, particularly in SNR and ductal delineation, and achieved image quality comparable to that of successful respiratory‐triggered acquisitions with regular breathing patterns. Improved initializations enhanced PSNR and reduced reconstruction time. Adjusting frozen/trainable configurations demonstrated that PSNR decreased only slightly from 38.25 dB (0/13) to 37.67 dB (12/1), while training time decreased up to 6.7‐fold.
Conclusion
Zero‐shot learning delivers high‐fidelity MRCP reconstructions with reduced breath‐hold times, and the proposed partially trainable approach offers a practical solution for translation into time‐constrained clinical workflows.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.
- Research and Clinical Translation, Magnetic Resonance, Siemens Healthineers AG, Erlangen, Germany.
ORCID-Profile: Jinho Kim, Florian Knoll
Förderung
- Deutsche Forschungsgemeinschaft
- Friedrich-Alexadner-Universität Erlangen-Nürnberg
- Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung - BMBF)
Zitieren
APA
Kim, J., Nickel, M. D., & Knoll, F. (2026). Zero‐Shot Self‐Supervised Learning of Single Breath‐Hold Magnetic Resonance Cholangiopancreatography ( MRCP ) Reconstruction. Magnetic Resonance in Medicine, 96(4), 1666-1681. https://doi.org/10.1002/mrm.70467
Vancouver
Kim J, Nickel MD, Knoll F. Zero‐Shot Self‐Supervised Learning of Single Breath‐Hold Magnetic Resonance Cholangiopancreatography ( MRCP ) Reconstruction. Magnetic Resonance in Medicine. 2026;96(4):1666-1681. doi:10.1002/mrm.70467
BibTeX
@article{Kim2026,
author = {Kim, Jinho and Nickel, Marcel Dominik and Knoll, Florian},
title = {Zero‐Shot Self‐Supervised Learning of Single Breath‐Hold Magnetic Resonance Cholangiopancreatography ( MRCP ) Reconstruction},
journal = {Magnetic Resonance in Medicine},
year = {2026},
volume = {96},
number = {4},
pages = {1666--1681},
doi = {10.1002/mrm.70467}
}
EUVIMED