EUVIMEDEUVIMED entdecken
Crossref

DARCS: Memory-Efficient Deep Compressed Sensing Reconstruction for Acceleration of 3D Whole-Heart Coronary MR Angiography

Zhihao Xue, Fan Yang, Juan Gao, Zhuo Chen, Hao Peng, Chao Zou, Hang Jin, Chenxi Hu

IEEE Transactions on Medical Imaging · 2025

DOI: 10.1109/tmi.2024.3524717 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2025. Verfasst von Zhihao Xue und 7 weiteren Autor:innen. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Magnetic Resonance Angiography, Heart, Coronary Angiography. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 78 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Shanghai Science and Technology Commission.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Three-dimensional coronary magnetic res- onance angiography (CMRA) requires reconstruction algorithms that can significantly suppress the artifacts encountered in heavily undersampled acquisitions. While unrolling-based deep reconstruction methods have ach- ieved state-of-the-art performance on 2D image recons- truction, their application in 3D reconstruction is hindered by the large amount of memory required to train an unrolled network. In this study, we propose a memory-efficient deep compressed sensing method that employs a sparsifying transform based on a pre-trained artifact estimation network. The artifact image estimated by a well-trained network is expected to be sparse when the input image is artifact-free and less sparse when the input image has artifacts. Thus, the artifact estimation network can be used as an inherent sparsifying transform. The proposed method, De-Aliasing Regularization-based Compressed Sensing (DARCS), was compared with a patch-based low-rank method, de-aliasing generative adversarial network (DAGAN), 3D model-based deep learning (MoDL), plug-and-play, and AI-assisted compressed sensing (AI-CS) in terms of 3D CMRA acceleration. The results demonstrate that DARCS surpasses the reconstruction quality of the comparison methods, by approximately 2 dB in peak signal-to-noise ratio (PSNR). Furthermore, the proposed method generalizes well to different undersampling rates, patterns, and noise levels, with a memory usage of only 63% of that needed by 3D MoDL. In conclusion, DARCS improves reconstruction quality for 3D CMRA with reduced memory burden.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Imaging, Three-DimensionalMagnetic Resonance AngiographyHeartCoronary AngiographyHumansDeep LearningAlgorithmsCoronary VesselsArtifacts

Förderung

  • National Key Research and Development Program of China
  • National Natural Science Foundation of China
  • Shanghai Science and Technology Commission

Zitieren

APA

Xue, Z., Yang, F., Gao, J., Chen, Z., Peng, H., Zou, C., Jin, H., & Hu, C. (2025). DARCS: Memory-Efficient Deep Compressed Sensing Reconstruction for Acceleration of 3D Whole-Heart Coronary MR Angiography. IEEE Transactions on Medical Imaging. https://doi.org/10.1109/tmi.2024.3524717

Vancouver

Xue Z, Yang F, Gao J, Chen Z, Peng H, Zou C, et al. DARCS: Memory-Efficient Deep Compressed Sensing Reconstruction for Acceleration of 3D Whole-Heart Coronary MR Angiography. IEEE Transactions on Medical Imaging. 2025. doi:10.1109/tmi.2024.3524717

BibTeX
@article{Xue2025,
  author = {Xue, Zhihao and Yang, Fan and Gao, Juan and Chen, Zhuo and Peng, Hao and Zou, Chao and Jin, Hang and Hu, Chenxi},
  title = {DARCS: Memory-Efficient Deep Compressed Sensing Reconstruction for Acceleration of 3D Whole-Heart Coronary MR Angiography},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2025},
  doi = {10.1109/tmi.2024.3524717}
}
Kostenlos starten →