2662

A 10-element receive-only RF coil array for imaging the brain of awake marmosets
Wen-Yang Chiang1,2, Cecil Chern-Chyi Yen2, Mary P. McDougall1, and Afonso C. Silva2

1Department of Biomedical Engineering, Texas A&M University, College Station, TX, United States, 2Cerebral Microcirculation Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, United States

Synopsis

A 10-element head RF coil array was developed for imaging the brain of awake marmosets. A soccer ball design was used to improve whole brain coverage and parallel imaging acceleration when compared to our previous designs. Coil clip and PLA cement were introduced to help place the small coil elements on surfaces of irregular shape and optimize geometric decoupling. The matching network provided independent adjustments of tuning, matching, active detune and preamp decoupling, greatly simplifying the construction of coil arrays in space-limited applications. Molded-in foam padding was also introduced to provide additional comfort during awake marmoset imaging.

Purpose

The common marmoset (Callithrix jacchus), is gaining increasing popularity in neuroscience and pre-clinical research.1-6 The use of awake animals in functional neuroimaging studies is preferred to allow the animal to actively attend to the functional tasks, and to avoid the confounding effects of anesthesia on neurovascular coupling 7-11. For example, anesthesia-free animals show stronger and faster hemodynamic responses.12, 13 We previously report the development of individualized mechanical restraints and coil arrays for awake marmoset brain imaging 14-16. However, when more array elements are used in the limited space inside an animal scanner, the use of traditional coil fabrication methods becomes increasingly challenging. Here we describe streamlined design techniques to fabricate the coil elements, their matching network circuits and molded-in foam pad to alleviate these common problems in the development of coil arrays for awake marmoset imaging.

Methods

Circuit diagram of the receive-only coil elements is proposed in Figure 1. The proposed design allowed independent development of tuning, matching, active detune traps and phase shifters. This independency eliminated the tedious iterative work that is typical throughout the fabrication of the array elements. The physical length of the new circuit design is also shorter, and the reduced length helped with fitting extra coil elements in the limited space.

The array was built on a coil former that was designed to attach to our previously reported head-restraining helmets.14-16 Soccer ball pattern 17 of the receiving coil array was projected onto the coil former as fiducial marks in SolidWorks, and the coil former was 3D printed using polylactic acid (PLA) material. 3D-printed PLA coil clips were used to mount the coils by fusing the clips onto the coil former with tetrahydrofuran solvent. The ~0.2 mm tolerance of the coil clips allowed fine adjustment for the optimization of geometric decoupling. PLA cement was made by dissolving recycled PLA material into tetrahydrofuran, and the cement was applied on the coil clips to permanently secure the coils after the geometric decoupling is optimized (Figure 2).

An individualized helmet with molded-in foam padding was made for each marmoset. Based on the MRI of a marmoset, a mold was designed in SolidWorks and 3D printed using water-soluble polyvinyl alcohol (PVA) material. The PVA mold was placed under the helmet, and two-part urethane foam was injected into the space in between using a mixing syringe. Once the foam padding was cured, the entire assembly was soaked in water to remove the PVA mold. After the foam was dried and trimmed, the conforming 2-mm thick foam padding was embedded under the custom-made helmet.

A head-and-brain phantom (Figure 3) was designed and filled with 4.9 g/L saline and 0.5 mM Gd to help with the development and testing of the head coil array. MR images of both the phantom and of the awake marmoset were used to evaluate the performance of the coil array.

Results

All 10 coil elements were tuned to 300.128 MHz and noise matched to 50 Ohm (S11 < - 21 dB) when the array was loaded with the phantom. The active detune trap achieved an uncompromised performance of 45 dB isolation when a 22 nH inductor was used to resonate with the 12 pF tuning capacitor. Phase shifters were installed to achieve 24±5.7 dB of preamp decoupling. Unloaded to loaded Q ratios were measured (Channel 1 to 10): 1.19, 1.15, 1.13, 1.10, 1.07, 1.09, 1.10, 1.14, 1.13 and 1.2.

Phantom images showed that the 10-channel coil array present 1.83, 1.90, 2.02, 2.54 and 2.62 times SNR gain over the transmitting volume coil at the center of the brain, frontal, temporal, occipital and parietal lobes, respectively. G-factor maps and GRAPPA images showed minimal SNR reduction when 3X acceleration was used along the right-left or the anterior-posterior direction (Figure 4). In vivo images (Figure 5) showed good sensitivity and contrast over the entire marmoset brain.

Discussion

Because phased array coils for small-animal imaging are usually built in tight spaces, iterating capacitor/inductor values throughout the fabrication of the array is difficult. With the streamlined design developed here, replacing components throughout the process of coil development is no longer necessary, and the performance of each individual element is not compromised. The new molded-in foam padding does not shrink and provides perfect fit for the marmoset every time, and increases the marmosets’ comfort.

As a result, the proposed design showed better decoupling among coil elements and thus enabled better accelerated parallel imaging compared to that of our previous coil array.14 This work will greatly improve the quality of awake marmoset brain imaging for ongoing neurovascular studies.

Acknowledgements

Stephen Dodd, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, United States.

References

1. Kishi, N., et al., Common marmoset as a new model animal for neuroscience research and genome editing technology. Dev Growth Differ, 2014. 56(1): p. 53-62.

2. Okano, H., et al., The common marmoset as a novel animal model system for biomedical and neuroscience research applications. Semin Fetal Neonatal Med, 2012. 17(6): p. 336-40.

3. Solomon, S.G. and M.G. Rosa, A simpler primate brain: the visual system of the marmoset monkey. Front Neural Circuits, 2014. 8: p. 96.

4. Abbott, D.H., et al., Aspects of common marmoset basic biology and life history important for biomedical research. Comp Med, 2003. 53(4): p. 339-50.

5. Leibovitch, E., et al., Novel marmoset (Callithrix jacchus) model of human Herpesvirus 6A and 6B infections: immunologic, virologic and radiologic characterization. PLoS Pathog, 2013. 9(1): p. e1003138.

6. Ohta, S., et al., Isolation and characterization of dendritic cells from common marmosets for preclinical cell therapy studies. Immunology, 2008. 123(4): p. 566-74.

7. Attwell, D., et al., Glial and neuronal control of brain blood flow. Nature, 2010. 468(7321): p. 232-43.

8. Iadecola, C., Neurovascular regulation in the normal brain and in Alzheimer's disease. Nat Rev Neurosci, 2004. 5(5): p. 347-60.

9. Kleinfeld, D., et al., A guide to delineate the logic of neurovascular signaling in the brain. Front Neuroenergetics, 2011. 3: p. 1.

10. Lauritzen, M., Relationship of spikes, synaptic activity, and local changes of cerebral blood flow. J Cereb Blood Flow Metab, 2001. 21(12): p. 1367-83. 11. Masamoto, K. and I. Kanno, Anesthesia and the quantitative evaluation of neurovascular coupling. J Cereb Blood Flow Metab, 2012. 32(7): p. 1233-47.

12. Liu, J.V., et al., fMRI in the awake marmoset: somatosensory-evoked responses, functional connectivity, and comparison with propofol anesthesia. Neuroimage, 2013. 78: p. 186-95.

13. Pisauro, M.A., et al., Fast hemodynamic responses in the visual cortex of the awake mouse. J Neurosci, 2013. 33(46): p. 18343-51.

14. Papoti, D., et al., Design and implementation of embedded 8-channel receive-only arrays for whole-brain MRI and fMRI of conscious awake marmosets. Magn Reson Med, 2016.

15. Papoti, D., et al., An embedded four-channel receive-only RF coil array for fMRI experiments of the somatosensory pathway in conscious awake marmosets. NMR Biomed, 2013. 26(11): p. 1395-402.

16. Silva, A.C., et al., Longitudinal functional magnetic resonance imaging in animal models. Methods Mol Biol, 2011. 711: p. 281-302.

17. Wiggins, G.C., et al., 32-channel 3 Tesla receive-only phased-array head coil with soccer-ball element geometry. Magn Reson Med, 2006. 56(1): p. 216-23.

Figures

Figure 1. Comparison of our previous (a) 14 and the proposed (b) circuit diagram of the coil. Previously, Lm is used for both matching and active detuning trap. However, because Lm is usually very small (~3 nH) when the coil is matched to 50 Ohms, performance of the active detuning circuit is low. A common solution is to sacrifice matching by increasing the value of Lm for better active detuning. In the new design, tuning, matching, active detuning and phase shifter can be optimized independently, so that no compromises are necessary and optimal coil performance can be achieved.

Figure 2. (a) Photo of 10-channel head coil array with large floating cable traps placed at every quarter-wave. (b) The lid of the coil enclosure is removed to show the soccer ball construction. Small floating cable traps were wrapped by blue heat-shrink tubes. Each receiving coil was mounted by 3D printed cable clips. Each clip was first fused with the coil former by tetrahydrofuran solvent, and the gap between the clip and the coil was filled by PLA cement after geometric decoupling was optimized. This technique allowed streamlined construction of coil array in the space-limited application.

Figure 3. (a-c) MR images of the custom phantom. When the ball pattern is aligned in 3 pilot views, the brain area is along AC-PC line. (d) Cross sectional view of the CAD design of the custom phantom. A “head shell” and a “brain shell” were derived from an MRI of a marmoset, and body-center-cubic structure was used to support the brain shell in the phantom. This phantom allowed us to localize the position of the brain area consistently and help evaluate SNR distribution inside of the brain area without the use of a marmoset.

Figure 4. Axial view of g-factor maps with 2, 3 and 4 acceleration factors along right-left and anterior-posterior directions are shown on the top row. The corresponding GRAPPA reconstructions of the custom phantom are shown on the bottom row. 3X acceleration along R-L and A-P directions show a satisfactory compromise between acceleration and minimal SNR reduction.

Fig 5. Spin-echo sequence (RARE factor/TR/TE/NEX/matrix size/resolution = 16/9000/64/16/200x160x60/0.2x0.2x0.5 mm3) of an awake marmoset show excellent sensitivity and homogeneity over the entire marmoset brain.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
2662