3836

Minimum-Duration Parallel Transmit Pulses for Simultaneous Multislice Imaging
Mihir Pendse1 and Brian K Rutt1

1Stanford University, Stanford, CA, United States

Synopsis

We describe a procedure for obtaining minimum duration parallel transmit pulses for simultaneous multislice imaging. The strategy involves a combination of the IMPULSE parallel transmit pulse design algorithm for mitigating local SAR hotspots and enforcing flip angle homogeneity; an optimal control algorithm for intelligent design of slice selective subpulse shape with minimum power; and time-optimal VERSE for reducing peak power with minimum increase in pulse duration. Results indicate a reduction in pulse duration by a factor of 4.9 compared to a pTx-SMS pulse designed with conventional techniques.

Introduction

Simultaneous multislice (SMS) imaging is an effective technique for reducing acquisition time or increasing slice coverage. The design of SMS RF pulses to excite multiple slices simultaneously is challenging because these pulses have high peak power [1]. At high fields, RF pulse design is further complicated by B1+ inhomogeneity and increased local SAR concerns that can be addressed through SAR-aware parallel transmission (pTx), typically resulting in multi-spoke pulses [2,3]. In addition to having high peak power, the duration of pTx pulses with multiple spokes can be long, making them very sensitive to B0 inhomogeneity. Here we describe a technique for designing minimum-duration pTx-SMS pulses that satisfy peak power, excitation accuracy, and local SAR constraints. Our results demonstrate significant reduction in pulse duration compared to existing pTx-SMS pulse designs.

Theory

We modified the previously described IMPULSE-SMS design algorithm [4] to find minimum-duration RF and gradient waveforms following the same principles as minimum-SAR pTx pulse design. The design procedure is as follows. First, IMPULSE pTx optimization is performed to find spokes locations and channel weightings for each slice in order to minimize SAR while enforcing a specified flip angle inhomogeneity tolerance over all excited slices. Next, the SMS pulse to excite multiple slices is designed using optimal control theory [5] to minimize a cost function consisting of terms corresponding to both excitation accuracy and pulse power. The regularization term in the cost function can be chosen such that the resulting excitation accuracy is an accordance with the specified flip angle inhomogeneity tolerance. Finally, if the peak power limit on any channel is exceeded, time-optimal variable rate selective excitation (VERSE) [6] is applied to enforce the peak power constraint with minimum pulse duration by reshaping the RF and gradient waveforms without altering the excitation profile.

Methods

Small-tip pTx-SMS RF and gradient waveforms were designed to simultaneously excite 6 coronal brain slices using a 16 channel transmit coil and pulse constraints as described in [7]. A minimum duration pulse was designed while satisfying all of the following constraints: peak local SAR < 10W/kg, global SAR < 3.2W/kg, and in-slice RMS flip angle inhomogeneity < 6.7%. We compared these results to those obtained with a conventional SAR-unaware pTx-SMS design algorithm using naïve stretching of the pulse to satisfy peak power and SAR constraints.

Results

We found that it was possible to achieve an SMS factor of 6 with pulse duration of less than 5ms while meeting all constraints. The conventional pTx-SMS pulse design resulted in a duration of 24.4 ms (Figure 1 ). The proposed method succeeded in reducing this duration to 4.96 ms (Figure 2) which is a factor of 4.9 improvement . As seen in Figures 3 and 4, this results in flip angle maps that have the same degree of homogeneity while also satisfying all local SAR, global SAR, peak RF power, and maximum gradient amplitude/slew constraints.

Discussion

The proposed approach allows for significant reduction of the duration of pTx-SMS pulses by using a SAR-aware pTx optimization to better satisfy safety constraints and by using time-optimal VERSE to enforce peak power constraints in a more efficient manner than naïvely stretching the pulse. The shorter pulses will be more robust to B0 inhomogeneity and thus can be expected to be useable in invivo imaging without resulting in major artifacts. One limitation of this method is that VERSE is sensitive to accuracy in the gradient field and errors in the timing between the RF and gradient waveforms but the dramatic improvement seen here is motivation for investing in more advanced hardware with better management of these concerns.

Acknowledgements

Research support from the NIH (P41 EB015891, 1 S10 RR026351-01A1), GE Healthcare, and Zurich MedTech AG (Sim4Science program)

Dr. William Grissom and organizers of 2015 ISMRM RF Pulse Design Challenge for motivation behind the methods developed in this work.

References

[1] Eichner et al. Magn Reson Med. 2014 Oct;72(4):949-58 [2] M. Pendse and B. Rutt. IMPULSE: A generalized and scalable algorithm for joint design of minimum SAR parallel transmit RF pulses. Proceedings ISMRM 23:5008, 2015 [3] Hoyos-Idrobo, Andres, et al. "On variant strategies to solve the magnitude least squares optimization problem in parallel transmission pulse design and under strict SAR and power constraints." IEEE transactions on medical imaging 33.3 (2014): 739-748. [4] Pendse MR and Rutt BK. IMPULSE-SMS: Local SAR and peak power optimized pTx pulse design for simultaneous multislice imaging at high fields. Proceedings ISMRM 2016 [5] Aigner CS, Clason C, Rund A, and Stollberger R. RF pulse design for low SAR simultaneous multislice (SMS) excitation using optimal control. Proceedings ISMRM 2014 [6] Lee D, Lustig M, Grissom WA, Pauly JM. Time-optimal design for multidimensional and parallel transmit variable-rate selective excitation. Magn Reson Med 2009;61:1471–1479. [7] Grissom, W. A., Setsompop, K., Hurley, S. A., Tsao, J., Velikina, J. V. and Samsonov, A. A. (2016), Advancing RF pulse design using an open-competition format: Report from the 2015 ISMRM challenge. Magn. Reson. Med.

Figures

Figure 1: RF and gradient waveforms using a conventional pulse design algorithm

Figure 2: RF and gradient waveforms using the proposed pulse design algorithm

Figure 3: Flip angle maps using a conventional pulse design algorithm

Figure 4: Flip angle maps using the proposed pulse design algorithm

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
3836