Quantitative Brain Mapping Using Magnetic Resonance Fingerprinting on a 50-Mt Portable MRI Scanner

Liang Xuan,Yuxiang Zhang,Jiamin Wu,Yucheng He,Zheng Xu
DOI: https://doi.org/10.1002/nbm.5077
2024-01-01
NMR in Biomedicine
Abstract:Ultralow-field magnetic resonance imaging (ULF-MRI) has broad application prospects because of its portable hardware system and low cost. However, the low B0 magnitude of ULF-MRI results in a reduced signal-to-noise ratio in qualitative images compared with that of commercial high-field MRI, which can affect the visibility and delineation of tissues and lesions. In this work, a magnetic resonance fingerprinting (MRF) approach is applied to a homemade 50-mT ULF-MRI scanner to achieve efficient quantitative brain imaging, which is an original and promising disease-diagnosis approach for portable MRI systems. An inversion recovery fast imaging with steady-state precession-based sequence is utilized for MRF through Cartesian acquisition. A microdictionary analysis method is proposed to select the optimal repetition time and flip angle variation schedule and ensure the best possible tissue discriminative ability of MRF. The T1 and T2 relaxation properties and the B1+ distribution are considered for estimation, and the results are compared with those of gold standard (GS) quantitative imaging or qualitative imaging methods. The phantom experiment indicates that the quantitative values obtained by schedule-optimized MRF show good agreement, and the bias from the GS results is acceptable. The in vivo experiment shows that the relaxation times of white and gray matter estimated by MRF are slightly lower than the reference data, and the relaxation times of lipid are within the range of the reference data. Compared with qualitative MRI under ULF, MRF can intuitively reflect various items of brain tissue information in a single scan, so it is a valuable addition to point-of-care imaging approaches. A magnetic resonance fingerprinting (MRF) approach is applied to a homemade 50-mT ULF-MRI scanner to achieve efficient quantitative brain imaging. The T1, T2 relaxation properties, as well as the B1+ distribution, are taken into account for estimation, and the results are compared with gold standard quantitative imaging or qualitative imaging methods.image
What problem does this paper attempt to address?