Improved signal‐to‐noise performance of MultiNet GRAPPA 1H FID MRSI reconstruction with semi‐synthetic calibration data

• Published in: Magnetic resonance in medicine 2022-10, Vol.88 (4), p.1500-1515
• Main Authors: Chan, Kimberly L., Ziegs, Theresia, Henning, Anke
• Format: Article
• Language: English
• Subjects: acceleration; Brain slice preparation; Calibration; GRAPPA; Image acquisition; Image processing; Image reconstruction; Lipids; metabolite mapping; Metabolites; MRSI; neural networks; Noise; Reliability; Training
• ISSN: 0740-3194; 1522-2594
• DOI: 10.1002/mrm.29314

Purpose To further develop MultiNet GRAPPA, a neural‐network‐based reconstruction, for lower SNR proton MRSI (1H MRSI) data using adapted undersampling schemes and improved training sets. Methods 1H FID‐MRSI data and an anatomical image for GRAPPA reconstruction were acquired in two slices in the human brain (n = 6) at 7T. MRSI data were retrospectively undersampled for a 4×, 6×, and 7× acceleration rate. Signal‐to‐noise, relative error (RE) between accelerated and fully sampled metabolic maps, RMS of the lipid artifacts, and fitting reliability were compared across acceleration rates, to the fully sampled data, and with different kinds and amounts of training images. Results Training with semi‐synthetic images resulted in higher SNR and lower lipid RMS relative to training with acquired images from one or several subjects. SNR increased with the number of semi‐synthetic training images and the 4× accelerated data retains ∼30% more SNR than other accelerated data. Spectra reconstructed with 20 semi‐synthetic averages retained ∼100% more SNR and had ∼5% lower lipid RMS than those reconstructed with the center k‐space points of one image as was originally proposed for very high SNR MRSI data and had higher fitting reliability. The metabolite RE was lowest when training with 20‐semi‐synthetic training images and highest when training with the center k‐space points of one image. Conclusion MultiNet GRAPPA is feasible with lower SNR 1H MRSI data if 20‐semi‐synthetic training images are used at a 4× acceleration rate. This acceleration rate provided the best trade‐off between scan time and spectral SNR.