Najat Salameh1,2,3, Mathieu Sarracanie1,2,3, Loyd Waites4, David Waddington1,3,5, and Matthew Rosen1,2,3
1MGH/HST Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, 2Harvard Medical School, Boston, MA, United States, 3Department of Physics, Harvard University, Cambridge, MA, United States, 4Rensselaer Polytechnic Institute, Troy, NY, United States, 5ARC Center for Engineered Quantum Systems, School of Physics, University of Sydney, Sydney, Australia
Synopsis
Radiologists routinely use contrast-enhanced MRI with applications
mainly in oncology and abdominal imaging. Over the last decade, researchers
have put significant efforts in developing new probes for molecular imaging
where contrast agents would target only specific cells and/or regions. In all
cases, one main question remains: what
is the potential toxicity of this new contrast agent?
We propose here a safe approach to contrast-enhanced MRI, using pre-polarized biocompatible saline combined with imaging at ultra-low
field (0.0065 T).Purpose
Contrast-enhanced
MRI relies on changing the relaxivity of tissue, either by reduction in
T1
allowing an enhanced signal on
T1-weighted images, or by a reduction in
T2*
resulting in negative contrast on
T2 or
T2* weighted images to
track the uptake of the contrast agent. At ultra-low field (ULF), normal
biological
T1 values are significantly shorter compared to those at high
field, and the ULF regime is generally immune to
T2* effects suggesting
that conventional contrast agents will not be of tremendous utility for ULF
imaging. We describe here a new approach in ULF imaging by using saline as
contrast medium. Saline is biologically safe, and with the use of a small,
strong pre-polarizing permanent magnet, enables contrast-enhanced MRI at 0.0065
T.
Methods
Imaging was performed at 0.0065 T using a custom-built electromagnet-based
MRI scanner
1 (Fig.1). Saline hyperpolarization was performed using a 1.3 T
permanent magnet placed in the Faraday cage of the scan room, 1 m away from the
NMR coil (Fig.1). The permanent magnet is a 5.5 cm deep cylinder of diameter 5.5 cm, with a 2 cm axial
hole (K&J Magnetics, Pipersville, USA). It was placed inside a custom-built
steel shielded box, designed by simulation with COMSOL Multiphysics (Burlington,
USA) to minimize the effect of the permanent magnet on the homogeneity of our
scanner. A 10 mL plastic syringe was modified to form a cylinder with luer-lock
connectors on both ends and was placed inside of the permanent magnet. One end was
connected to a 60 mL syringe filled with water and placed on an infusion pump
outside the Faraday cage. The other end was connected to a PE50 capillary which
in turn connected to our phantom located in the imaging coil. The phantom itself consists of a modified 60
mL syringe allowing for continuous flow (with input from the 1.3 T magnet side,
and output to a waste container/jar). Constant flow of 20 mL per minute was started
a few seconds before the acquisition began. A 17 s b-SSFP sequence with 50%
undersampling and number of average NA=5 for a 2×2×10 mm
3 spatial resolution
was used as a reference before repeating the same acquisition while injecting
hyperpolarized water at a flow rate of 20 mL/min. The same scan with NA=20 was used as a reference. Data were processed using MATLAB (MathWorks, Natick, USA)
with scripts written in house.
Results
Simulations with COMSOL showed that the contribution to the local
magnetic field due to the permanent magnet were efficiently removed using our shielded box (Fig.2). The impact of the permanent magnet on
T2* was, however, strong
(
T2* dropped to 52 ms and was 690 ms without the 1.3 T magnet), but did not affect too strongly image quality. Figure 3
shows the image with NA=5 and no hyperpolarized water, whereas b) shows
the results for contrast-enhanced imaging. The enhanced water jet was overlaid on the
reference image with NA=20 c) and the result is shown in d).
Discussion
Our results show that we are able to perform contrast-enhanced MR
imaging at ultra-low magnetic field by using a strong permanent magnet placed close to our detection setup to
hyperpolarize water. We show that the presence of a permanent magnet
inside the Faraday cage affects
T2* but not enough to prevent good imaging quality. The transfer to
in vivo applications remains challenging for two main
reasons. The first limitation comes from the weak imaging gradient strength of our MRI
scanner (~ 1mT/m), which prevents from either better spatial resolution or
faster duty cycles and thus shorter acquisition times. The second limitation
comes from the short
T1 of hyperpolarized water (a few seconds) in our ULF
regime. A combination of stronger gradients and optimized NMR coils for receive operation would certainly help, as well as an efficient way to inject hyperpolarized
water in this setting (in particular by significantly reducing the dead
volume in the 1m tubing to the our phantom).
Conclusion
We have shown that hyperpolarized water can be used to perform
contrast-enhanced MRI at ultra-low magnetic field. This result, once optimized
and tested
in vivo, would open opportunities for MR angiography in a
cost-effective portable MRI scanner using fully biocompatible saline. Of
particular interest is the possible early diagnostic of acute phase of stroke.
Acknowledgements
D.E.J. Waddington was supported by ANSTO and the Australian-American Fulbright Commisson.
References
1. Sarracanie M, LaPierre C, Salameh N, et al. Low-cost high-performance MRI. Sci
Rep 2015;5:15177