Fast 3D Magnetic Resonance Fingerprinting (MRF) For Whole Brain Coverage in Less Than 3 Minutes
Dan Ma1, Jesse Hamilton2, Yun Jiang2, Nicole Seiberlich2, and Mark Griswold1

1Radiology, Case Western Reserve University, Cleveland, OH, United States, 2Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States

Synopsis

The purpose of this study is to accelerate the acquisition time of 3D MRF scans. A simple acquisition scheme was applied to allow a total factor of 144 acceleration as compared to the Nyquist rate, such that 3D T1, T2 and proton density maps can be acquired from a whole brain scan at clinical resolution in 2.6 minutes.

Purpose

The goal of this study is to accelerate the acquisition time of 3D magnetic resonance fingerprinting (MRF)[1,2]. MRF is an acquisition and processing framework that utilizes pseudorandomized acquisition scheme to enable simultaneous quantification of multiple tissue parameters, such as T1 and T2. This study implemented a simple accelerated acquisition scheme to allow a total factor of 144 acceleration as compared to the Nyquist rate, such that 3D T1, T2 and proton density (M0) maps can be acquired from a whole brain scan at clinical resolution in 2.6 minutes.

Methods

3D slab selective MRF with FISP readout [3] was implemented using a 3D slab-selective RF pulse with an optimized excitation profile. Flip angle and TR patterns for each partition were the same as [1]. In-plane acceleration was achieved by using a single spiral arm from a variable density spiral trajectory[3] in each TR. To improve spatial incoherence, a 7.8⁰ rotation of the trajectory was applied from one TR to the next. This yields an in-plane acceleration factor of 48. Through-partition acceleration was achieved by uniformly undersampling the partitions in an interleaved fashion by a factor of 3. In this way the sampling density of each partition, as well as the sampling density of each trajectory were the same after the entire acquisition. Further acceleration came from reducing the waiting time between neighboring partitions. Although traditional relaxometry methods require spins to be fully relaxed before sampling each partition, MRF can account for partial relaxation of the spins by simulating the known waiting time between neighboring partitions in the dictionary. Before image reconstruction, an SVD method [4] was applied in k-space, such that instead of reconstructing thousands of 3D images, only 15 3D volumes were reconstructed. This allowed for shortening of the reconstruction and pattern recognition time.

Acquisition

3D MRF were applied in both phantom and in vivo studies, with a FOV of 300x300x48 mm3 and a matrix size of 256x256x16 for the phantom study, and FOV of 300x300x120 mm3 and a matrix size of 256x256x24 for in vivo study. For each partition, 1500 time-points were acquired. A 3-second waiting time was added in between neighboring partitions in the in vivo study to allow for relaxation and to improve SNR. The total acquisition time is 2.6 minutes for a 3D brain scan.

Results

Figure 1 shows the results from 10 cylindrical phantoms with different T1 and T2 combinations. By using a 3D pulse, the excitation profile is relative flat in the central 10 partitions, which provides consistent T1 and T2 values across these partitions. Figures 2 and 3 display the T1 and T2 maps from different partitions of an in vivo study, in transversal and coronal views, respectively.

Discussion

This study showed a fast acquisition scheme for a fully quantitative 3D MRF scan in a 2.6 minute acquisition time demonstrating T1, T2 and M0 maps with 1.1x1.1x5 mm3 resolution. Further improvement of image resolution and acquisition time is particularly interesting, as no parallel imaging or compressed sensing methods have been used.

Acknowledgements

The authors would like to acknowledge funding from Siemens Healthcare and NIH grants NIH 1R01EB016728-01A1 and NIH 5R01EB017219-02

References

1. Ma D. ,Gulani V, Seiberlich, N, et al. Magnetic resonance fingerprinting. Nature (2013) 187-192

2. Ma D., Pierre E., Jiang Y., et al, Three-Dimensional MR Fingerprinting (MRF) and MRF-Music AcquisitionsISMRM (2015) , 3390

3. Jiang Y., Ma D., Seiberlich N., et al, MR Fingerprinting Using Fast Imaging with Steady State Precession (FISP) with Spiral Readout. MRM (2014)

4. McGivney D., Pierre E., Ma D. et al, SVD Compression for Magnetic Resonance Fingerprinting in the Time Domain. IEEE TMI. 2014. 10.1109

Figures

Figure 1:T1 and T2 values from all partitions in a phantom study

Figure 2: In vivo T1 (a) and T2 (b) maps in axial views

Figure 3: In vivo T1 (a) and T2 (b) maps in coronal views



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
3180