1508
Performance evaluation of a B0-shim multi-coil system for small animal temperature mapping at 3T1Lauterbur Imaging Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, 2Shenzhen Key Laboratory for MRI, Shenzhen, China, 3Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 4UCSF/UC Berkeley Joint Graduate Group in Bioengineering, San Francisco, CA, United States
Synopsis
The magnetic field variation is a critical factor affecting the accuracy of temperature measurement in MRT. In this study, a 5-channel B0-shim coil was constructed for small animal temperature mapping in the MRI guided high intensity focused ultrasound (HIFU) at 3T. Firstly, the shimming ability was evaluated by the phantom study with a result that the standard deviation (STD) value of the offset magnetic field has reduced to 69% after currents optimized. Secondly, the relationship between T2* and SNR improvement has been studied. The results demonstrate that the temperature measurement accuracy is improved by 8% with the local multiple B0-shim coils.
Introduction
The Proton Resonance Frequency Shift (PRFS) based Magnetic Resonance Thermometry (MRT) method is the most popular technique in the guidance and monitoring of many interventional procedures such as tumor ablation1. However, the accurate thermometry is hampered by temporal magnetic field changes, one of which sources is the variation of magnetic susceptibility distribution of tissue for the temperature dependent2. To mitigate the effect, the ideal solution would be to reduce the magnetic field variations at their source. In this study, we have built a local shim coil to compensate for B0 inhomogeneity and then evaluated its performance by phantom experiments, as a preliminary study for the improvement of temperature measurement accuracy in vivo experiments.Methods
A 5-channel shim coil is designed based on the setup for small animal temperature mapping at a 3T Siemens MRI system, as shown in Figure 1. The coil former with a 6 cm diameter hole on the bottom was made by polycarbonate material to minimize its impact on the B0 field. Each coil element with a 4 cm diameter circle shape was place on the upper half part surface of the former and the shim controller was built in our previous work3. To reduce the impact of eddy currents on phase image, the B0 maps were acquired with a 2D multi-echo gradient echo (GRE) sequence, of which parameters were: echo times (TE) = [2.66 … 18.94] ms, echo train length=5, echo spacing=4.07 ms, repetition time (TR) = 200 ms, flip angle=10o, FOV=100 mm, matrix size =64 x 64, slice thickness=5 mm, slices=1, bandwidth=390 Hz/pixel. Two maps were acquired with 100 mA current and zero current for each channel coil, of which phase images subtraction can be represented as the field sensitivity map. The optimized B0 maps were calculated with a linear regression of the unwrapping phase images of each channel coil at different TEs and scaled to Hz in a 22 mm x 22 mm region of interest (ROI) by minimizing the standard deviation (STD) value of the offset frequency. The T2* related to the offset magnetic field ΔB0 is given by equation 1 and it can be calculated by exponential fit with equation 2, as following:
1/T2*=1/T2+γ·ΔB0 (1)
S(n)=M0·exp(-TE(n)/T2*) (2)
The T2* maps were obtained by a 12-echo 2D GRE sequence with parameters: TE= [2.66 … 47.43] ms, echo spacing=4.07 ms, TR = 500 ms, flip angle=60o, while the other parameters were the same. Here, SNR calculated with equation: sqrt(2)·signal/σ ( σ is the standard deviation of noise image) was also evaluated by a 2D GRE sequence (TR/TE=500 ms/20 ms, flip angle=60o). For PRFS temperature mapping, the calculated equation of the change temperature can be given as1:
ΔT=(φ-φbaseline)/(γ·α·B0·TE) (3)
Where α=-0.01 ppm·oC-1 is the PRFS coefficient and γ is the gyromagnetic ratio. φ and φbaseline are the current phase images acquired and the phase images acquired before heating, respectively. It was acquired by a 2D GRE sequence (TR/TE=15 ms/10 ms, flip angle=10o, measurement=100, each scan time=1 s).
Phantom results
Figure 1 shows the optimized (with shim) and basic (without shim) offset frequency distributions, in which the mean and the STD values in the ROI are also depicted. As a result, the STD value of the optimized field has been reduced to 69% of the basic field STD value. The T2* maps related to and the SNR maps are displayed in figure 2, from which it can be known that the mean T2* value in the optimized ROI is greatly enhanced with 44% while the SNR improvement is 13%. The signal variation trends of the ideal and practical curves at different TEs with the measured T2* are well accordant, as shown in figure 4. Most importantly, the temperature measurement accuracy of room temperature without heating, as demonstrated in figure 5, has been improved by 8%.Discussion
The relationship between T2* and SNR improvement is difficult to quantify for that T2* value is easily affected by the magnetic field changes and SNR is related to the other factors, such as the received coil and the transmit efficiency. In this study, we mainly focus on the shimming performance of the 5-channel B0-shim coil and the improvement of the temperature measurement accuracy. Further work will be implemented with vivo experiments.Conclusion
A 5-channel B0-shim coil for small animal temperature mapping has been designed and evaluated. A better field homogeneity with the STD value decrease to 69% and an 8% improvement of temperature measurement accuracy have been achieved with the shim coil, which shows a promising development in temperature mapping.Acknowledgements
This work was supported in part by NSFC under Grant No. 61571433, 81527901, 81527901, 81327801 and 81627901, Guangdong Province grants 2014A030313691, 2015B020214006 and 2014B030301013. Youth Innovation Promotion Association of CAS No. 2017415, city grants KQJSCX20160301143250 and JCYJ20170413161314734 and a Pengcheng Scholar Award.References
1. Zou C, et al. Referenceless MR thermometry—a comparison of five methods [J]. Physics in medicine and biology, 2016, 62(1):1.
2. Sprinkhuizen S M, Konings M K, van der Bom M J, et al. Temperature‐induced tissue susceptibility changes lead to significant temperature errors in PRFS‐based MR thermometry during thermal interventions [J]. Magnetic resonance in medicine, 2010, 64(5): 1360-1372.
3. Jo Lee, et al. Design of 5-channel On-coil Shimming Coil for Rat Brain MRI Guided High Intensity Focused Ultrasound: a preliminary study. 25th ISMRM, 2017, 2602.
Figures
