Deep learning CAIPIRINHA-accelerated 3D MRI of the knee cartilage at 7 Tesla

Adrian Alexander Marth, Thomas Marth, Georg Constantin Feuerriegel, Constantin von Deuster, Dominik Paul, Marcel Dominik Nickel, Daniel Nanz, Reto Sutter,
Deep learning CAIPIRINHA-accelerated 3D MRI of the knee cartilage at 7 Tesla,
European Journal of Radiology,
Volume 191,
2025,
112287,
ISSN 0720-048X,
https://doi.org/10.1016/j.ejrad.2025.112287.
(https://www.sciencedirect.com/science/article/pii/S0720048X25003730)
Abstract: Purpose
To investigate the feasibility and degree of scan-time reduction that can be achieved by deep learning (DL)-reconstruction in 3D dual-echo steady-state (DESS) knee-cartilage imaging accelerated with CAIPIRINHA (controlled aliasing in parallel imaging results in higher acceleration) at 7 Tesla (T).
Methods
35 knee joints of 18 volunteers were imaged using an isotropic fat-suppressed DL-reconstructed CAIPIRINHA-accelerated DESS sequence (CAIPIRINHADL) with an isotropic resolution of 0.25 mm. CAIPIRINHADL acquisitions were accelerated four-, six-, eight-, and twelve-fold (CAIPIRINHADL4-CAIPIRINHADL12), and compared to a state-of-the-art, fourfold accelerated GRAPPA (generalized autocalibrating partially parallel acquisitions) DESS acquisition (GRAPPA4) with matched spatial resolution. Three readers rated image-quality criteria. Rating differences were characterized with Friedman’s analysis of variance. Inter-reader agreement was assessed for grading of incidental cartilage lesions.
Results
CAIPIRINHADL6 had the shortest acquisition time (4:04 min) among sequences whose images were rated in the best category, with no statistically significant difference (p > 0.999) to GRAPPA4 (5:53 min). The overall quality of CAIPIRINHADL8 and CAIPIRINHADL12 images was rated lower than GRAPPA4 images, while artifacts were rated significantly more severe (p < 0.001, each). Inter-reader agreement for cartilage lesion grading ranged from almost perfect for CAIPIRINHADL4 and GRAPPA4 (κ = 0.90, each) to moderate for CAIPIRINHADL12 (κ = 0.52).
Conclusions
The quality of DL-reconstructed 3D DESS images acquired at 7 T with four- and six-fold CAIPIRINHA-based acceleration was equivalent to that of 3D-DESS images acquired with four-fold GRAPPA-based acceleration. These findings suggest the feasibility of DL-reconstructions for imaging of the knee cartilage at 7 T field strength.
Keywords: Magnetic resonance imaging; Ultra-high field MRI; 7 Tesla; Deep-learning reconstruction; Knee