Synopsis

Inhomogeneous magnetization transfer (ihMT) is a new MRI technique that shows promise for myelin imaging. A significant boost of the ihMT signal intensity is achievable when using a concentrated energy deposition scheme. Here we characterized the power dependency of ihMT for various energy deposition schemes (distributed vs. concentrated RF) and we identified distinct regimes with linear, sub-linear and saturated B1 dependency. This has a strong impact on the ihMT response to variable sequence configurations and on the most suitable conditions for different field strengths, e.g. to enhance sensitivity or provide immunity to B1⁺ inhomogeneities.

Introduction

Methods

Whole brain ihMT imaging was carried out using a pulsed ihMT module combined with a 3D GRE readout [3,5]. Fixed ihMT parameters were: ±Δf=7kHz, p_w=0.5 ms , Δt=1ms (see definitions in Fig.1), N_p=12 Hann-shaped pulses (except otherwise specified), bandwidth= 100Hz/pixel, 4.8ms TE and 2.5mm isotropic resolution. Readout segmentation and partial k-space saturation were used to reduce acquisition time, and investigate higher B1 intensities [3].

Experiment 1 was performed on a whole body 1.5-T MRI system (Avanto, Siemens, Erlangen, Germany) with a 12-channel receive-only head coil on healthy volunteers. IhMT was studied as function of the intensity of the saturation pulses, for different TR settings. Three configurations were chosen based on the characteristic of the boost effect vs TR (Fig.2c,f): 1/ TR=24ms corresponding to the most distributed energy deposition achievable with the fixed ihMT and readout parameters, 2/ TR=67ms corresponding to the optimal boosted configuration where the energy deposition was optimized to maximize the sensitivity of ihMT, 3/ TR=140ms (using higher-power Tukey pulses) corresponding to the most concentrated energy deposition considering sequence parameters and hardware limitations (peak B1). Similar experimental conditions were simulated by stepwise numerical resolution of the differential equations describing the ihMT dynamics [3] using Matlab (The MathWorks Inc., Natick, MA, USA). All results are expressed as a function of the average power defined over TR (B1rms), and the saturation power defined over the ihMT pulse train (B1sat=B1rms x √TR/(NPxΔt), Fig.1). B1rms was varied from 2uT to 5.5uT. IhMT ratios, ihMTR (fig1), were calculated in the Pyramidal Tract, PT.

Experiment 2 was performed on a whole body 3-T MRI system (Verio, Siemens) with a 32-channel receive-only head coil on healthy volunteers. IhMTR values were measured in the whole brain for B1rms = 2.2 μT and Np=6 Tukey pulses for both concentrated (TR=260ms, B1sat = 14.7 μT) and distributed energy deposition (TR=19ms, B1sat = 3.9 μT) and their dependency with the B1⁺ distribution was evaluated.

Results and discussion

Experiment 1: Fig.2 a-b shows experimental ihMTR as a function of B1rms and B1sat for TR=24, 67 and 140 ms. Three different patterns were found: For TR=24 ms (distributed RF power) ihMT signal slowly and almost linearly increased with B1rms over the entire range. For TR=67ms (optimized RF power), the increase was faster and tended to reach a maximum. For TR 140 ms (concentrated RF power), the ihMT signal increases even faster but rapidly saturates for B1rms > 4uT. When expressed as function of B1sat, common dependency regimes can be observed between the three configurations, with linear increase followed by strong inflexion for B1sat > 10-12uT, highlighting the dominance of saturation power (B1sat) in the ihMT effect. Simulations satisfyingly predict these observations (Fig.2 d-e).

Experiment 2: In Fig 3, the ihMTR measured in all white matter pixels is reported as a function of the local B1⁺ (obtained with a prepared turbo FLASH, Siemens WIP 543, and normalized to its nominal value). In addition to the ihMT boost obtained for the long TR of 260ms here, a trend for reduced B1 sensitivity was observed as compared to the distributed approach TR of 19 ms, consistent with the characteristics in B1 measured at 1.5T (Fig. 2)

Conclusion

Acknowledgements

References

[1] G. Varma, G. Duhamel, C. de Bazelaire, and D. C. Alsop, “Magnetization Transfer from Inhomogeneously Broadened Lines: A Potential Marker for Myelin,” Magn. Reson. Med., vol. 73, no. 2, pp. 614–622, Feb. 2015.

[2] O. M. Girard, V. H. Prevost, G. Varma, P. J. Cozzone, D. C. Alsop, and G. Duhamel, “Magnetization transfer from inhomogeneously broadened lines (ihMT): Experimental optimization of saturation parameters for human brain imaging at 1.5 Tesla,” Magn. Reson. Med., vol. 73, no. 6, pp. 2111–2121, juin 2015.

[3] O. M. Girard et al., “Theoretical and Experimental Optimization of a 3D Steady-State Inhomogeneous Magnetization Transfer (ihMT) Gradient Echo Sequence: Boosting the ihMT Sensitivity with Sparse Energy Deposition,” in Proceedings of the ISMRM, Singapore, 2016, p. 2892.

[4] G. Varma, O. M. Girard, V. H. Prevost, G. Duhamel, and D. C. Alsop, “Increasing the inhomogeneous magnetization transfer (ihMT) signal in vivo with high amplitude, low duty cycle irradiation,” in Proceedings of the ISMRM, Singapore, 2016, p. 2890.

[5] O. M. Girard et al., “Whole Brain inhomogeneous MT using an ihMT prepared 3D GRE sequence at 1.5T,” in Proceedings of the ISMRM, Toronto, Canada, 2015, p. 3356.