A Pulse Sequence for T1rho Imaging of Liver in Human Subjects

Weitian Chen¹, Queenie Chan², Min Deng¹, and Yixiang Wang¹

¹Imaging and Interventional Radiology, the Chinese University of Hong Kong, New Territory, Hong Kong, ²Philips Healthcare, New Territory, Hong Kong

Synopsis

T1rho is a potential biomarker for evaluation of liver fibrosis. However, the application of T1rho imaging for liver disease in human subjects remains challenging. We propose a pulse sequence based on single shot Turbo (Fast) Spin Echo (SSTSE or SSFSE) acquisitions to address the problems.

Introduction

Chronic liver disease is a significant health problem worldwide. Liver fibrosis is a key feature in most chronic liver diseases. When identified early, liver fibrosis is treatable and can be reversed. Recently, researchers reported that T1rho is sensitive for evaluation of liver fibrosis (1). However, the application of T1rho imaging for liver disease in human subjects is challenging due to the factors including the presence of rich blood vessels, respiratory motion, B1 RF and B0 field inhomogeneity, and prolonged scan time. Here we propose a pulse sequence for T1rho imaging of liver based on single shot Turbo (Fast) Spin Echo (SSTSE or SSFSE) acquisitions to overcome these challenges.

Methods

One of the main advantages of using SSTSE for T1rho imaging of liver is inherent black-blood effect. The main factors affecting blood signal suppression in SSTSE include flip angle of refocusing pulses and echo time (2,3) We noticed that the T1rho quantification based on T1rho-preped SSTSE acquisition is independent of the flip angle modulation and the echo time of SSTSE, as long as SNR is sufficient and CPMG condition is satisfied (Figure 1). This unique feature of TSE provides us flexibility to optimize black-blood effect without compromising T1rho quantification accuracy. The pulse sequence (Figure 2) starts with resetting magnetization to zero (4). A long T1 recovery time is followed to allow longitudinal signal recovery before T1rho-prep. SSTSE alone may be insufficient to achieve through-plane blood suppression (5). DIR is used to improve through plane blood suppression. SPAIR is used for fat suppression. A RF pulse cluster which has simultaneous compensation of B1 RF and B0 field inhomogeneity is used for T1rho prep (6). A module of phase correction is inserted between SSTSE and T1rho to maintain CPMG condition for SSTSE imaging. The time gap after T1rho-prep is kept as short as possible to avoid T1 contamination of T1rho contrast. Bloch simulation is used to validate that T1 contamination during the given time gap in our pulse sequence is ignorable. The data is acquired using SSTSE with constant refocusing flip angles. Data sets were acquired from a Philips Achieva 3.0T system equipped with dual transmitter. A 32 channel cardiac coil was used as the receiver. The scan parameters include: TR/TE=2500ms/15ms, resolution 1.5mm x 1.5mm, slice thickness 6mm, SENSE acceleration 2, half scan factor 0.6, T1 recovery time 2220ms, NSA 1, four TSL = [0 10 30 50ms], and spinlock frequency 500Hz. All TSLs of a 2D slice was acquired within a single breathhold of 10 seconds. RF shimming was applied to reduce B1 inhomogeneity. The data sets were fitted to a mono-exponential model using Levenberg-Marquardt non-linear least square fitting method.

Results and Discussion

Figure 3 shows the acquired images and the corresponding T1rho map. The mean T1rho value of liver within ROI is 43ms, and the mean of the goodness-of-fit is 0.96. Note most blood signal has been suppressed. Further studies are needed to optimize parameters for blood suppression. The rapid scan time and high SNR efficiency make the proposed pulse sequence potentially useful for routine clinical use. However, reproducibility test is needed to validate the robustness of this pulse sequence. Additional clinical studies are also needed to further understand the value of T1rho imaging for liver disease.

Conclusion

A new pulse sequence for T1rho imaging of liver disease is proposed which can be used to acquire a 2D T1rho map of liver within a single breathhold of 10 seconds with black blood effect

Acknowledgements

We would like to acknowledge the help from Feng Huang, Chengguang Zhao, and Zhigang Wu from Philips Healthcare for the help on pulse sequence programming

References

1. Wang YX et al, Radiology 2011, p712 2. Busse R, ISMRM Proceeding 2006, p2430 3. Storey P et al, MRM 2010, p1098. 4. Li et al, MRM 2008, p298 5. Jara H et al, MRM 1999, p575 6. Witschey WRT et al, JMR 2007, p75

Figures

Figure 1: Left: The refocusing flip angle of the echo train used for simulation. VFA – variable flip angle. CFA – Constant flip angle. Right: The estimated T1rho value. Five groups correspond to the five different flip angle schemes. With each flip angle scheme, we simulate five cases. For each case we skip different number of echoes (0, 5, 10, 15, 20, respectively) at the beginning of echo train before we start acquire the center of k-space. The actual T1rho is 45ms.

Figure 2: The diagram of the proposed pulse sequence.

Figure 3: Four T1rho-weighted images were acquired with TSL = [0,10,30,50] ms within a single breath-hold of 10 seconds. a) and b) T1rho-weighted images with TSL=0 and 50ms, respectively. c) The estimated T1rho map. The T1rho value and goodness-of-fit within ROI are 43ms and 0.96, respectively.

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

2830