Salty MRI

If you all have been following along, you know that MRI works with protons, predominantly hydrogen in water. Well, that’s not 100% true, some MRI scanners have multi-nuclear capability which I’ll explain more in a bit. Bangerter et al recently published their work on a sodium  (Na) MRI coil for breast imaging. The Science Daily article says:

..substantial improvement over anything that we’ve seen using this particular MRI technique for breast cancer imaging. (emphasis added) It’s important to note that this is a remarkable improvement for sodium based MRI but not breast MRI in general and I’ll explain why. There is some evidence that sodium increases in malignant tumors. There’s also some evidence that sodium iodine symporters (ion channels) co-localize with estrogen receptors in breast cancer. In the Bangerter article they state, the objective of this study was to determine whether a sodium phased array would improve sodium breast MRI at 3 T. The secondary objective was to create acceptable proton images with the sodium phased array in place. So let’s distill what that means.

Multi-nuclear

With MRI and NMR (nuclear magnetic resonance) you can tune the hardware (coil and electronic detection circuits) to different frequencies for different nuclei. Most scanners are tuned for protons and that’s it. It’s less expensive than multi-nuclear. A transmit/receive coil is pictured in figure 1 below. You can think of it as two coils in one because one loop is tuned for 23-Na and the other for 1-H. It’s also a phased array which I’ll explain later. One important thing about multi-nuclear is the gyromagnetic ratio. It’s a constant for any given nucleus that relates the magnetic resonant frequency of the external magnetic field. If you recall, MRI works by changing the direction of the magnetic field. The magnetization of the nucleus has to change with it. When you let the system equilibrate, you get information from that decay back to equilibrium. Since there are no external torques, the total angular momentum must be conserved. This mass rotation can be measured. The gyromagnetic ratio for protons is 42.58 MHz/T. Other gyromagnetic ratios for common “x” nuclei for multi-nuclear systems: 19-Fluorine 40.05 MHz/T, 31-Phosphorus 17.24 MHz/T, and 23-Sodium 11.26 MHz/T. You’ll note that sodium is the smallest of the bunch, which adds to its low signal-to-noise ratio (SNR).

SNR

I wrote about signal-to-noise ratio before. Reaching for some signal  http://goo.gl/bNMAVm  The problem with sodium imaging is it has inherently low SNR due to lower concentration of Na relative to hydrogen (H, protons). There is also a rapid biexponential signal decay for Na that you don’t get with H, so you have less time to capture the signal. In figure 6 and 10 below, the difference in SNR is apparent. However, in figure 10 (comparing top left with bottom left) you can see the remarkable improvement in SNR when using a phased array coil vs. non-phased array coil. 

Phased Array Coil

In computers you often hear about parallel processing. In MRI, especially clinical MRI, you can use parallel imaging to go faster and in some cases improve SNR. One of the easiest coils to make for imaging a whole sample, is a volume coil. It’s basically a tube with copper wires around the circumference like a solenoid. An even easier coil to make is a surface coil, basically a loop of copper that you can place on the surface of a sample. Surface coils have great SNR at the surface (hence the name) and the signal drops off as you image farther from the coil. So they aren’t so good for volume imaging. In a phased array coil you basically have many surface coils placed around the sample. They are out of phase so that you can image simultaneously without having them interfere with each other. Parallel imaging is very complex in terms of both the electronics and raw data reconstruction. There are a lot of software enhancements to get acceleration with phased array coils.

So Bangerters group did a great job making a multi-nuclear phased array coil for breast imaging. By combining the 23-Na loop with 1-H loop, you don’t have to move the sample to get high SNR proton images to use as an anatomic guide for the lower SNR sodium image.

h/t Filippo Salustri (http://goo.gl/5pmfkQ)

Image source and ref:

A 3 T sodium and proton composite array breast coil.

Kaggie JD, Hadley JR, Badal J, Campbell JR, Park DJ, Parker DL, Morrell G, Newbould RD, Wood AF, Bangerter NK.

Magn Reson Med. 2013 Sep 16. doi: 10.1002/mrm.24860

http://www.ncbi.nlm.nih.gov/pubmed/24105740

Quantitative immunohistochemical analysis reveals association between sodium iodide symporter and estrogen receptor expression in breast cancer.

Chatterjee S, Malhotra R, Varghese F, Bukhari AB, Patil A, Budrukkar A, Parmar V, Gupta S, De A.

PLoS One. 2013;8(1):e54055. doi: 10.1371/journal.pone.0054055. Epub 2013 Jan 14.

http://www.ncbi.nlm.nih.gov/pubmed/23342072

Happy #ScienceSunday  and  #HappyBirthdayHalfpintBuddy