Simultaneous estimation of PD, T₁, T₂, T₂^*, and ∆B₀ using magnetic resonance fingerprinting with background gradient compensation

Abstract

Purpose This study aims to estimate PD, T1, T2, T2*, and normalΔB0 simultaneously using magnetic resonance fingerprinting (MRF) with compensation of the linearly varying background field. Methods MRF based on fast imaging with steady‐state precession (FISP) and multi‐echo spoiled gradient (SPGR) schemes are alternatively used, which encode T2 and T2*, respectively. Simulations are performed to determine the appropriate ratio of the FISP and SPGR sections with respect to the T2 and T2* accuracy. Additionally, background field inhomogeneity (Gz) compensation using z‐shim gradients are incorporated into the SPGR section and the dictionary. The background field compensation is tested in the phantom experiment under well‐shimmed and poor‐shimmed conditions. An in vivo experiment is performed and the estimated parameters are compared before and after Gz compensation. Results The T1, T2, and T2* values from the phantom results are in good agreement with the reference methods under well‐shimmed condition. The underestimated T2 and T2* values under poor‐shimmed condition are recovered by Gz compensation and the parameters are also in good agreement with the reference methods. In the human brain, T2 and T2* values are restored by Gz compensation in regions where the magnetic field is particularly inhomogeneous, such as near the sinus and ear canals. Conclusions The proposed FISP and SPGR combined MRF provides a simultaneous estimation of PD, T1, T2, T2*, and normalΔB0. By incorporating field inhomogeneity as a gradient term into both the sequence and dictionary, T2 and T2* values can be restored where field inhomogeneity exists.

B0 field homogeneity recommendations, specifications, and measurement units for MRI in radiation therapy

“…29 Pkpk field inhomogeneity values of 1.6 ppm in the brain, 4.8 ppm in the breast, 5.6 ppm in the liver, and 4.8 ppm in the spinal cord were reported with active shimming. 27,[30][31][32] Hence, there are four objectives that will be addressed herein. First, we will relate MRI quality recommendations for RT to B 0 field homogeneity for systems designed for diagnostic MRI, RT simulation, and MR-IGRT.…”

Magnetic Resonance Fingerprinting with Combined Gradient- and Spin-echo Echo-planar Imaging: Simultaneous Estimation of T1, T2 and T2* with integrated-B1 Correction

“…This leads to faster dictionary generation, lower storage requirements and faster dictionary matching. For example as a rough comparison, in our case, the dictionary size is over three orders of magnitude smaller than a previously reported study (16) that used spirals to estimate the same tissue relaxation parameters (T1, T2, T2*).…”

“…Wyatt et al explored a variable-echo-time (TE) scheme to incorporate T2* sensitivity into the originally proposed MRF pulse sequence (15). Hong et al did so by combining a fast imaging with steady-state precession (FISP) sequence (sensitive to T2) and a multi-echo spoiled gradient echo (SPGR) sequence (sensitive to T2*) (16). Lastly, Wang et al used a quadratic-RF phase increment pattern to create T2* sensitivity (17).…”

Magnetic resonance fingerprinting review part 2: Technique and directions

“…In a breast MRF study using MRF-FISP, 37 a dictionary with 20,059 columns representing possible T 1 , T 2 combinations was used, whereas in a brain tumor study using MRF-bSSFP, 72 the additional dimension of off-resonance increases the dictionary size to 287,709 columns. In the case where the sequence is also sensitized to quantify T 1 , T 2 , off-resonance, and T 2 *, the number of columns in the dictionary was reported to be over 30 million in Wang et al, 61 and 64 million in Hong et al 63 Inner product pattern matching has been shown to be accurate and robust to the high degree of aliasing artifacts due to undersampling in several of the initial MRF studies, including Ma et al 1 and Jiang et al 5 Also shown in both, 1,5 the number of timepoints used in the sequence will have a direct impact on the quality and accuracy of the T 1 and T 2 maps. Therefore, for the sequences in these initial studies, the number of timepoints was generally between 1000 to 3000.…”