Introduction

Rhesus monkey is a common model used functional neuro MR imaging (fMRI) [1,2]. Due to the required stereotaxic instrument in monkey MR imaging, the commercially available RF coils do not fit the experiment setup. Consequently the imaging signal-to-noise ratio (SNR) and base resolution are dramatically degraded. In this work, we designed, constructed and tested a dedicated eight-channel monkey RF coil array perfectly fitting the MRI stereotaxic instrument and rhesus monkey’s head. The results show that the proposed monkey coil obtained over 66% higher SNR than that of the commercially available Siemens flex small4 in phantom studies, and achieves 0.4×0.4×0.4mm³ high resolution images in in-vivo study at 3T.

Methods

The outer appearance of the customized rhesus monkey head coil (monkey coil) was built to perfectly fit the structure of the MRI stereotaxic instrument (KOPF 1430M), as shown in Figure 1(a). From the front view could see that the monkey coil design has two inclined planes at the right and lift side to match the support rod of the inclined plane. This design not only can ensure the monkey coil placed steadily on the instrument. The whole coil shell was made by polycarbonate. The inner view of monkey coil is illustrated in Figure 1(c) and (d). The coil structure was designed in eight coil arrays with single loop size 75×65 mm2 in average. Coil arrays were made by 6mm thick chopper wire, and were arranged in three rows from head to feet direction with following orders: 3,3,2. Each coil array included three parts of elements layout, separated coil array into three equal length. The three-element layout include one active detune board and two tuning capacitances. All coil arrays were tuned at 123.2MHz, and electromagnetic decoupling among the coil elements was achieved by the overlapping method and low input impedance preamplifiers. Phantom study: All on-system studies were scanned by using Siemens 3T Tim Trio system, and all comparison experiments were performed with the 4 channels small flexible coil array (flex small 4) from Siemens. A customized phantom with 432cm² content volume was made to mimic the adult rhesus monkey head by using photosensitive resin 3D printer, and was filled by (3.75g NISO₄×6H₂O + 5g NACL)/1000g H₂O. The phantom was placed and fixed in the position the rhesus monkey’s head was located in the stereotaxic instrument. Gradient echo (GRE) sequence with following parameters was used for obtaining signals：TR/TE= 300/10ms, slice thickness = 3mm, FOV = 150×150mm², acquisition matrix = 256×256, flip angle = 60 degree, bandwidth = 130 Hz/pixel, number of average = 1. While the power set to zero was used to acquire the noise. Images in all three orientations were required. MATLAB was used for the post analysis. The method called root sum of squares (sos) from reference[3] was used for calculating SNR. The inverse g-factor map was calculated with acceleration factor set to 2 and the accelerate direction from right to left, anterior to posterior and head to feet for transverse, sagittal and coronal orientation in order. In-vivo study: An adult rhesus monkey (10 years old, weight 11 kg, male) was scanned for the anatomic images, by T1-weighted 3D MPRAGE sequence with following parameters: TR/TE/TI = 2100ms /2.8 ms/ 1100 ms, FOV = 140×140×77.44 mm³, matrix = 320×320×176, isotropic resolution 0.4×0.4×0.4mm³, flip angle = 8 °, bandwidth = 200 Hz/pixel, echo spacing = 8.8ms, number of average = 1.

results

The SNR maps and SNR-curve charts were shown in Figure 2. The 1D profiles of SNR along the dash-lines in the images were plotted in the lower inset of Figure 2. The area of the brain was measured from the adult rhesus monkey in the in-vivo study. From the SNR-curves, monkey coil was 66%, 173% and 114% in average better than flex small 4 in transverse, sagittal and coronal orientations. The inverse g-factor maps were represented in Figure 3. The white dash line represents the brain zone, and the calculated g-factor data were written under. The results indicate the parallel imaging capability of the monkey coil is better than flex small 4. The high spatial resolution (0.4×0.4×0.4mm³) of 3D rhesus monkey brain structural images is demonstrated in Figure 4. The images reveal that the proposed monkey coil can achieve higher quality images even in the deep brain area.

Discussion/Conclusion

A custom-designed 8-channel monkey coil which well fits the specified stereotaxic instrument in the monkey imaging was constructed and tested for 3T MR imaging. The monkey coil is characterized with high sensitivity, robustness and easy to use, and is able to provide high resolution 3D images of whole rhesus monkey brain, better acceleration ability, and high SNR in all three directions.

Acknowledgements

This work was supported in part by NSFC under Grant No. 61571433,81527901，81527901, 81327801and 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.