Collaborative Reconstruction of PROPELLER‐EPI Data Using POCSMUSE (CORPUSE) for High‐Fidelity Diffusion MRI
Magnetic Resonance in Medicine · 2026-10 · 96 (4) · 1929-1946
DOI: 10.1002/mrm.70448 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 30.05.2026. Verfasst von Hailin Xiong und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, The Chinese University of Hong Kong, Hong Kong, China. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Brain, Diffusion Magnetic Resonance Imaging, Echo-Planar Imaging. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Research Grants Council (RGC) of Hong Kong, Innovation and Technology Commission (ITC), The Government of Hong Kong Special Administrative Region, Innovation and Technology Commission.
Abstract (Originalsprache)
Purpose
To develop a reconstruction framework for DW‐PROPELLER‐EPI that improves image quality and SNR efficiency under per‐blade acceleration while minimizing EPI‐related artifacts, enabling high‐resolution diffusion‐tensor imaging (DTI) with fewer blades.
Methods
We propose CORPUSE, a joint reconstruction framework adapted from POCSMUSE and tailored for multiblade DW‐PROPELLER‐EPI. The method integrates distortion‐correction operators into a joint‐blade reconstruction model and exploits the inherent redundancy of multiblade sampling. Leveraging self‐extracted composite 2D phase errors and blade‐specific field maps as physics‐based constraints, CORPUSE improves reconstruction performance. By using wider blades with higher per‐blade acceleration, the framework increases scan efficiency while preserving image quality and geometric fidelity. The framework was evaluated in healthy volunteers on a 1.5T MRI scanner across varying numbers of blades and with two PROPELLER‐EPI trajectories: long‐axis (LAP) and short‐axis (SAP), under different per‐blade accelerations and scan conditions.
Results
In vivo experiments showed that CORPUSE improved image sharpness, geometric fidelity, and overall reconstruction quality compared with conventional PROPELLER‐EPI reconstruction. These improvements enabled higher per‐blade acceleration and wider blades without compromising image quality, thereby supporting more flexible and efficient reconstruction for LAP and SAP trajectory. Additionally, CORPUSE demonstrated greater motion resilience than multishot EPI based on multiplexed sensitivity encoding (MUSE), preserving structural detail even under subtle motion conditions.
Conclusion
The CORPUSE framework enables high‐resolution, high‐quality DTI with fewer blades, improving the practicality of DW‐PROPELLER‐EPI. By maintaining SNR and geometric accuracy under high per‐blade acceleration, it offers a robust and efficient alternative to other multishot diffusion imaging approaches.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Biomedical Engineering, The Chinese University of Hong Kong, Hong Kong, China.
- Multi-Scale Medical Robotics Center, Hong Kong, China.
- Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong, China.
ORCID-Profile: Hing-Chiu Chang
Förderung
- Research Grants Council (RGC) of Hong Kong
- Innovation and Technology Commission (ITC), The Government of Hong Kong Special Administrative Region
- Innovation and Technology Commission
Zitieren
APA
Xiong, H., Liang, L., Chen, S., Li, Y., Yuan, C., Liu, T., Xu, X., & Chang, H. C. (2026). Collaborative Reconstruction of PROPELLER‐EPI Data Using POCSMUSE (CORPUSE) for High‐Fidelity Diffusion MRI. Magnetic Resonance in Medicine, 96(4), 1929-1946. https://doi.org/10.1002/mrm.70448
Vancouver
Xiong H, Liang L, Chen S, Li Y, Yuan C, Liu T, et al. Collaborative Reconstruction of PROPELLER‐EPI Data Using POCSMUSE (CORPUSE) for High‐Fidelity Diffusion MRI. Magnetic Resonance in Medicine. 2026;96(4):1929-1946. doi:10.1002/mrm.70448
BibTeX
@article{Xiong2026,
author = {Xiong, Hailin and Liang, Liyuan and Chen, Shihui and Li, Yi and Yuan, Chenglang and Liu, Tianbaige and Xu, Xiaorui and Chang, Hing-Chiu},
title = {Collaborative Reconstruction of PROPELLER‐EPI Data Using POCSMUSE (CORPUSE) for High‐Fidelity Diffusion MRI},
journal = {Magnetic Resonance in Medicine},
year = {2026},
volume = {96},
number = {4},
pages = {1929--1946},
doi = {10.1002/mrm.70448}
}
EUVIMED