Reduction of RF Pulse Duration using Dephased Transition
Seohee So1, HyunWook Park1, Dongchan Kim1, Hyunseok Seo1, JaeJin Cho1, Young Woo Park1, and Kinam Kwon1

1Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea, Republic of

Synopsis

Slice profile used for magnetic resonance (MR) imaging has transition region between passband and stopband. Sharper transition performs better slice selection. This abstract proposes method to design short-duration RF pulse without increasing transition region width. Additional phase is merged into a transition region of slice profile. Two RF pulses having different phases are used alternately. The proposed algorithm produces about 30% reduction of RF pulse duration without transition increase.

PURPOSE

Slice selection in magnetic resonance imaging is assumed as a desired region is selected with equal weight. It could be possible if the selection profile is a rectangular shape. But in practice, there exists transition region between passband and stopband in which the profile gradually changes. As duration of RF pulse decreases, this transition region width increases. The transition region in slice profile produces signal from unwanted position and could cause spatial blurring. In order to get more effective RF pulse excitation, this abstract proposes how to design a short RF pulse without increasing the transition width. The proposed method adopts RF pulses whose phase in transition region varies.

METHODS

For short RF pulse without increasing transition width, the proposed method uses two key concepts. First one is that the slice profile has a wider transition region so that the RF pulse can be shorter. Second one is that two RF pulses are alternately applied and each RF pulse has different phase in the transition region.

Three steps are required to produce the proposed RF pulse that has de-phased transition region. At first, an optimal polynomial is produced by using the Parks-McClellan algorithm for FIR filter design. At this step, a desired bandwidth is set to half-amplitude width of the profile like Shinar-Le Roux pulse design 1. Second step is to add phase to the transition region. The additional phase is merged into slice profile of the RF pulse. Final step is the inverse SLR transform process 1. The designed slice profile which contains phase variation is converted to RF pulse through the inverse SLR transform which is widely used.

When MR image is produced, two RF pulses are alternately applied, which have different phases in the transition region. One is involved to get odd phase encoding lines. Even phase encoding lines are acquired by the other RF pulse. This MR acquisition algorithm separates 2D MR image according to phase. If the first RF pulse has phase θ and the other pulse has phase ψ at one point $$$(x,y)$$$ of the transition region, the MR image contributed by the transition region could be expressed as follows,

$$\widehat{M}(x, y) = e^{i \frac{\theta - \psi}{2}}\left\{M(x,y)\cos\frac{\theta + \psi}{2} + iM(x, y-\frac{N_{y}}{2})\sin\frac{\theta + \psi}{2} \right\}$$

where $$$M(x,y)$$$ is a magnetization at the transition region.

Above equation shows that phase in transition region introduces magnitude reduction of transition region in original image position and additional FOV/2 shifted image compared to the conventional image. FOV/2-shifted image is removed by GRAPPA algorithm 2. Figure 1 explains the proposed image acquisition process briefly.

RESULTS & DISCUSSION

The proposed method can produce the same transition region with shorter RF pulse as the conventional RF pulse which is designed by PM FIR filter and SLR algorithm. Figure 2 shows experimental results of a resolution phantom and a human brain. In the experiment, phase added to transition region of slice profile is designed as linearly changing. The phase slope is π/(transition width) for one of the RF pulses and -π/2*(transition width) for the other. Phase difference between two pulses is a linear phase in transition region of resultant slice profile, which reduces remaining ripples in stop band by dephasing of the transition region.

In the phantom experiment, steps in left and right part of phantom represents slice profile. Steps near the center mean transition region. The conventional pulse with a width of 2.56ms and the proposed pulse with 1.96ms show nearly same transition region (number of steps in Fig. 2) and almost same brain images. It implies that the proposed method shows narrower transition than the conventional one for the same pulse duration. Figure 3 shows Bloch simulation result of RF pulse duration versus 10% to 90% transition region. According to Bloch simulation results, about 30.54% of RF pulse duration is reduced by the proposed method. In other words, the transition region of the proposed method decreases about 23.4% for the same RF duration.

CONCLUSION

The proposed method provides reduction of transition region in slice profile. The proposed method makes it possible to use shorter RF pulse for MR imaging without change of image quality. It could be used for fast imaging and short TE imaging.

Acknowledgements

This research was supported by the Brain Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2014M3C7033999), and by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry for Health and Welfare, Korea (HI14C1135).

References

1. Pauly, John, et al. Parameter relations for the Shinnar-Le Roux selective excitation pulse design algorithm [NMR imaging]. Medical Imaging, IEEE Transactions on, 1991, 10.1: 53-65.

2. Lu, Lan, et al. Lipid elimination with an echo-shifting N/2-ghost acquisition (LEENA) MRI. Magnetic Resonance in Medicine, 2015, 73.2: 711-717.

Figures

The proposed k-space data acquisition scheme

a : The conventional slice profile of SINC pulse with 2.56ms.

b,c : The proposed slice profiles with 1.96ms(b) and 2.56ms(c).

d, e, f : Resolution phantom images using pulses whose slice profiles are a, b and c, respectively.

g, h, I : In vivo human brain images using pulses whose slice profiles are a, b and c, respectively.


Bloch simulation results of RF pulse duration versus 10% to 90% transition width



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1887