Introduction

Recent studies report importance of the magnetization transfer (MT) effect [1-2] on quantitative magnetic resonance imaging [3-10]. Although many studies have been reported on the MT effect in multislice imaging [11-14], to the best of our knowledge, the MT effect on the QRAPMASTER sequence [15] has not yet been reported.

Materials and methods

The phantom consisted of seven test tubes (16mm OD; 14mm ID). Six test tubes were filled with MnCl₂ aqueous solution (T₁ ~ 500-1000 ms). One test tube was filled with raw chicken breast meat. The phantom was imaged with a small horizontal-bore 1.5 T MRI system [16]. Pulse sequences were multislice multiple spin-echo (MSMSE) and QRAPMASTER sequences. The QRAPMASTER sequence was developed by inserting saturation pulses (FA=120°, SLR design) into the sequential-order MSMSE sequence. The sequence parameters were; number of slices = 11, slice thickness = 5 mm, slice gap = 0 mm, FOV = (64 mm)², image matrix = 256², TR = 2200 ms. The number of echoes and echo spacing of the MSMSE were 8 and 12 ms. Effective TE of the QRAPMASTER sequence was 12 and 60 ms. The shape of the 90° and 180° pulses was a hamming-windowed sinc with seven side lobes. The duration times of the RF pulses were both 1 ms and the theoretical peak intensities were 47.083 and 94.166 μT. The excitation frequency varied from -40 to +40 in 8 kHz steps. The delay time (TD) for the QRAPMASTER sequence varied from 420 to1420 in 200 ms steps. Relaxation times of the phantom were calculated using a data matching between the series of image datasets experimentally acquired and dictionary datasets calculated by Bloch simulations. The Bloch image simulations were performed using a Bloch simulator (BlochSolver) [16]. The MT effect was evaluated by numerically solving modified Bloch equations for a two-pool model [17].

Results

Figs.1(a)-(c) show the first-echo real-part images of the central section acquired with the single-slice multiple SE, interleave MSMSE, and sequential MSMSE sequences. Figs.1(d)-(f) are the difference images of Figs.1(a)-(c). The image intensity ratio of the chicken breast sample shown in Figs.1(a)-(c) was 1:0.82:0.73. Figs.2(a)-(c) show the first-echo real-part images of the central section simulated with the same sequences as shown in Figs.1(a)-(c). The calculation time for one MSMSE image were about 4 hours using 91,750,240 isochromats and RTX2080Ti GPU. Gaussian noise of the same level as experiments was added on the simulated images. The image intensity difference between the multislice and the single-slice images was not observed for the central sample because the MT effect was not incorporated in the Bloch image simulation. Fig.3 shows the absolute value first-echo images of the central section acquired by the QRAPMASTER sequence at TD = 420-1620 ms.
Fig.4(a) shows the T₁ maps obtained by the QRAPMASTER sequences. Fig.4(b) shows the T₁ values averaged over the ROI in the T1 maps plotted against those measured by the standard method. Blue and orange dots indicate the T₁ values of the MnCl₂ solution and the chicken breast sample. Fig.4(c) shows the T₂ maps obtained by the MSMSE sequence. Fig.4(d) shows the T₂ values averaged over the ROI in the T₂ maps plotted against those measured by the standard method. Blue and orange dots indicate the T₂ values of the MnCl₂ solution and the chicken breast sample.
Figs.5(a)-(b) show the temporal variation of M_z of the chicken breast sample (sample 7) calculated for the central slice with the MSMSE sequences with and without the MT effect. Fig.5(c) shows the temporal variations of M_z of sample 7 calculated with the single-slice and sequential MSMSE sequences. The image intensity ratio of sample 7 shown in Figs.5(a)-(c) was 1:0.73:0.67. Figs.5(d)-(e) show the temporal variation of M_z of sample 7 calculated for the central slice with the QRAPMASTER sequences. The points on the graphs represent M_z just before the 90° pulses for each slice. Fig.5(f) shows M_z with and without the MT effect plotted against TD. These curves were fitted to exponential functions with single time constants of 929 ms and 1021 ms. Figs.5(g)-(h) show the temporal variation of M_z of sample 7 calculated for the central slice with the QRAPMASTER sequences having 3.2 times pulse width. The FA of the refocusing pulse was 180°. Fig.5(i) shows M_z with and without MT effects plotted against TD. The recovery curves were fitted to exponential functions with single time constants of 862, 810, and 702 ms for 90°, 120°, and 180° flip angles of the refocusing pulses, and 1024 ms calculated without the MT effect.

Discussion

T₁ values of the chicken breast samples were far from the regression lines of T₁ of MnCl₂ solution measured by the QRAPMASTER sequence. The T₁ shortening of the chicken breast sample was semi-quantitatively explained by the Bloch simulation for the two-pool model. The difference between the observed value (~750 ms) and calculated value (~930 ms) may be due to the theoretical absorption line shape (super-Lorentzian) of the sample, assumed two-pool model parameters, and neglection of the slice width effect. In conclusion, this study suggested that the MT effect shortens the T₁ values of biological samples obtained by the QRAPMASTER sequence.

Acknowledgements

No acknowledgement found.

References

[1] Wolff SD, Balaban RS. Magnetization transfer contrast (MTC) and tissue water proton relaxation in vivo. Magn. Reson. Med. 1989;10:135–144. [2] Henkelman RM, Huang X, Xiang QS, Stanisz GJ, Swanson SD, Bronskill MJ. Quantitative interpretation of magnetization transfer. Magn Reson Med. 1993;29:759-66. [3] Wood TC, Malik SJ. Magnetization Transfer. In: Seiberlich N, Gulani V, editors. Quantitative Magnetic Resonance Imaging, Elsevier; 2020, p. 839-856. [4] Ou X, Gochberg DF. MT effects and T1 quantification in single-slice spoiled gradient echo imaging. Magn Reson Med. 2008;59:835-45. [5] Crooijmans HJ, Gloor M, Bieri O, Scheffler K. Influence of MT effects on T2 quantification with 3D balanced steady-state free precession imaging. Magn Reson Med. 2011;65:195-201. [6] van Gelderen P, Jiang X, Duyn JH. Effects of magnetization transfer on T1 contrast in human brain white matter. Neuroimage. 2016;128:85-95. [7] Liu F, Block WF, Kijowski R, Samsonov A. Rapid multicomponent relaxometry in steady state with correction of magnetization transfer effects. Magn Reson Med. 2016;75:1423-33. [8] Radunsky, D, Blumenfeld-Katzir, T, Volovyk, O, Tal, A, Barazany, D, Tsarfaty, G, Ben-Eliezer, N. Analysis of magnetization transfer (MT) influence on quantitative mapping of T2 relaxation time. Magn Reson Med. 2019;82:145–158. [9] Teixeira RPAG , Malik SJ , Hajnal JV . Fast quantitative MRI using controlled saturation magnetization transfer. Magn Reson Med. 2019;81:907-920. [10] Teixeira RPAG, Neji R, Wood TC, Baburamani AA, Malik SJ, Hajnal JV. Controlled saturation magnetization transfer for reproducible multivendor variable flip angle T1 and T2 mapping. Magn Reson Med. 2020 84:221-236. [11] Dixon WT, Engels H, Castillo M, Sardashti M. Incidental magnetization transfer contrast in standard multislice imaging. Magn Reson Imaging 1990;8:417–422. [12] Melki PS, Mulkern RV. Magnetization transfer effects in multislice RARE sequences. Magn Reson Med 1992;24:189–195. [13] Santyr GE. Magnetization transfer effects in multislice MR imaging. Magn Reson Imaging 1993;11:521–532. [14] Weigel M, Helms G, Hennig J. Investigation and modeling of magnetization transfer effects in two-dimensional multislice turbo spin echo sequences with low constant or variable flip angles at 3 T. Magn Reson Med. 2010;63:230-4. [15] Warntjes JBM, Leinhard OD, West J, Lundberg P. Rapid magnetic resonance quantification on the brain: optimization for clinical usage. Magn Reson Med 2008;60:320–9. [16] Kose R, Kose K. BlochSolver: a GPU-optimized fast 3D MRI simulator for experimentally compatible pulse sequences. J Magn Reson 2017;281:51–65. [17] Graham SJ, Henkelman RM. Understanding pulsed magnetization transfer. J Magn Reson Imaging. 1997;7:903-912.