EUVIMEDEUVIMED entdecken
Crossref

Phase-constrained zero-shot self-supervised learning for BLADE liver MRI reconstruction

Uten Yarach, Sorravit Akrasirakul, Hendrik Mattern, Oliver Speck

Magnetic Resonance Materials in Physics, Biology and Medicine · 11.04.2026 · 39 (5) · 849-863

DOI: 10.1007/s10334-026-01359-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance Materials in Physics, Biology and Medicine, online erschienen am 11.04.2026. Verfasst von Uten Yarach und 3 weiteren Autor:innen. Beteiligte Einrichtung: Radiologic Technology Department, Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand. uten.yarach@cmu.ac.th. Hauptthemen laut MEDLINE: Liver, Supervised Machine Learning, Image Processing, Computer-Assisted, Magnetic Resonance Imaging. Die Arbeit nennt 43 Literaturangaben. Gefördert von: the Federal Ministry of Education and Research within the Research Campus STIMULATE.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: Liver MRI plays a critical role in the diagnosis and monitoring of liver disease; however, image quality is often degraded by respiratory motion and noise, particularly in high-resolution and diffusion-weighted imaging. Propeller-based sequences such as BLADE improve motion robustness, but advanced reconstruction strategies are required to fully exploit their potential under accelerated conditions. METHODS: We propose a phase-constrained zero-shot self-supervised learning (PC ZS-SSL) framework for BLADE liver MRI reconstruction. The method embeds BLADE forward and adjoint operators within an unrolled deep network together with an explicit phase estimation module. Unlike supervised approaches, PC ZS-SSL requires no external training data and instead leverages partitioned k-space. The framework was first evaluated using structured phantom experiments assessing spatial resolution and low-contrast detectability, and subsequently applied to in vivo 3 T T2-weighted (T2W, 20 blades) and diffusion-weighted imaging (DWI, 18 and 15 blades, b = 0, 800 s/mm2). Performance was compared with locally low-rank (LLR) and vendor-provided reconstructions. RESULTS: In phantom studies, PC ZS-SSL preserved fine structural details and demonstrated improved sharpness compared with fewer gradient updates and LLR. In vivo, PC ZS-SSL consistently reduced noise and ringing artifacts while maintaining anatomical fidelity. In T2W imaging, it achieved image quality comparable to LLR with fewer artifacts. In DWI, where noise is more pronounced, PC ZS-SSL provided clearer organ boundaries than both LLR and vendor reconstructions. Additional in vivo evaluation demonstrated that PC ZS-SSL remained robust under severe undersampling conditions (e.g., 4 blades) and challenging imaging scenarios, where LLR reconstructions exhibited residual artifacts. CONCLUSION: PC ZS-SSL enables high-quality, artifact-suppressed BLADE liver MRI without the need for external training data. Its strong performance in both phantom and in vivo experiments-particularly under high-noise diffusion conditions-highlights its potential for clinical translation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

LiverSupervised Machine LearningImage Processing, Computer-AssistedMagnetic Resonance ImagingPhantoms, ImagingHumansDiffusion Magnetic Resonance ImagingAlgorithmsImage Interpretation, Computer-AssistedMotionArtifactsLiver DiseasesSignal-To-Noise Ratio

Stichwörter der Autor:innen

BLADE imagingDeep LearningDiffusion-weighted imagingLiver MRIMRI reconstructionZero-shot self-supervised learning

Autor:innen und Institutionen

  • Radiologic Technology Department, Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand. uten.yarach@cmu.ac.th.
  • Radiologic Technology Department, Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand.
  • Department of Biomedical Magnetic Resonance, Otto-Von-Guericke University, Magdeburg, Germany.
  • German Center for Neurodegenerative Diseases (DZNE) Helmholtz Association, Magdeburg, Germany.
  • Center Behavioral Brain Sciences (CBBS), Magdeburg, Germany.
  • Research Campus STIMULATE, Otto Von Guericke University, Magdeburg, Germany.

Förderung

  • the Federal Ministry of Education and Research within the Research Campus STIMULATE

Zitieren

APA

Yarach, U., Akrasirakul, S., Mattern, H., & Speck, O. (2026). Phase-constrained zero-shot self-supervised learning for BLADE liver MRI reconstruction. Magnetic Resonance Materials in Physics, Biology and Medicine, 39(5), 849-863. https://doi.org/10.1007/s10334-026-01359-2

Vancouver

Yarach U, Akrasirakul S, Mattern H, Speck O. Phase-constrained zero-shot self-supervised learning for BLADE liver MRI reconstruction. Magnetic Resonance Materials in Physics, Biology and Medicine. 2026;39(5):849-863. doi:10.1007/s10334-026-01359-2

BibTeX
@article{Yarach2026,
  author = {Yarach, Uten and Akrasirakul, Sorravit and Mattern, Hendrik and Speck, Oliver},
  title = {Phase-constrained zero-shot self-supervised learning for BLADE liver MRI reconstruction},
  journal = {Magnetic Resonance Materials in Physics, Biology and Medicine},
  year = {2026},
  volume = {39},
  number = {5},
  pages = {849--863},
  doi = {10.1007/s10334-026-01359-2}
}
Kostenlos starten →