Quantitative comparison of GRAPPA and RAKI simultaneous multi-slice reconstruction algorithms
Magnetic Resonance Imaging · 2026-09 · 131 · 110697
DOI: 10.1016/j.mri.2026.110697 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Magnetic Resonance Imaging, 2026-09. Verfasst von Tyler Gallun und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA. Hauptthemen laut MEDLINE: Algorithms, Image Processing, Computer-Assisted, Magnetic Resonance Imaging. Die Arbeit nennt 30 Literaturangaben.
Abstract (Originalsprache)
The aims of this study were to (1) quantitatively compare various k-space interpolation-based simultaneous multislice (SMS) reconstruction algorithms, including linear GRAPPA and nonlinear, data-driven RAKI methods, and (2) provide an open-source toolbox for GPU-accelerated SMS reconstructions. For a single phantom, fully sampled k-space reference data and SMS-accelerated, RSMS, data were collected for RSMS factors of 2 through 5, with in-plane acceleration, Rip, added retrospectively. Slice-GRAPPA, split-slice-GRAPPA, readout-SENSE-GRAPPA, and their hyperparameter-tuned analogous deep learning-based RAKI reconstructions were performed at different combinations of RSMS and Rip factors. Performance of the reconstruction methods were compared quantitatively by testing for significant differences in whole-image SSIM and regions of interest-based measurements of coefficient of variation (CV). The number of epochs, hidden layer size, and the interaction between the two were statistically significant factors for a Type II ANOVA test with SSIM performance (p=0.00005, p=0.00687, and p=0.000281, respectively) for RAKI reconstructions with post hoc tests suggesting that 500 epochs, 3 hidden layers, and 128 neurons per layer strike an ideal balance of speed and performance for this dataset. CV testing through Kruskal-Wallis tests did not yield any significant differences between reconstruction algorithms at any given RSMS×Rip. Limited testing indicated that a RAKI method generally resulted in larger SSIM values at any net acceleration factor, (Rnet=RSMS×Rip), and all RAKI methods had a larger SSIM value than their respective GRAPPA counterparts at Rnet>4, with the exception of ROSR at RSMS=5 and Rnet=10, and SPSR at RSMS=2,Rip=5. The difference between GRAPPA and RAKI was typically large enough (ΔSSIM of >0.02) to be obvious at Rnet≥8. Slice-GRAPPA, slice-RAKI, or readout-SENSE-GRAPPA performed well for net accelerations Rnet≤6, and readout SENSE-RAKI or split-slice-RAKI performed well for Rnet≥8. Caution should be taken when using these generalizations as observations were dataset-specific and collected on a single vendor, coil array, pulse sequence, and phantom. They may, however, serve as a useful starting point for hyperparameter tuning. More experiments need to be conducted before these results can be translated into in vivo observations.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA.
- Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA. Electronic address: nmickevicius@mcw.edu.
Zitieren
APA
Gallun, T., & Mickevicius, N. J. (2026). Quantitative comparison of GRAPPA and RAKI simultaneous multi-slice reconstruction algorithms. Magnetic Resonance Imaging, 131, 110697. https://doi.org/10.1016/j.mri.2026.110697
Vancouver
Gallun T, Mickevicius NJ. Quantitative comparison of GRAPPA and RAKI simultaneous multi-slice reconstruction algorithms. Magnetic Resonance Imaging. 2026;131:110697. doi:10.1016/j.mri.2026.110697
BibTeX
@article{Gallun2026,
author = {Gallun, Tyler and Mickevicius, Nikolai J},
title = {Quantitative comparison of GRAPPA and RAKI simultaneous multi-slice reconstruction algorithms},
journal = {Magnetic Resonance Imaging},
year = {2026},
volume = {131},
pages = {110697},
doi = {10.1016/j.mri.2026.110697}
}
EUVIMED