Thursday, March 20, 2008

Gradient Calibration - 1D MRI

When people think about magnetic resonance imaging (MRI), they often think about the huge whole body imagers in hospitals. NMR spectroscopists use one dimensional MRI in a specially prepared sample to calibrate the Z gradient strength of their spectrometers. The sample consists of a plastic disk with a precisely known thickness immersed in a column of water (see the figure below). 1D MRI is conceptually quite simple - a linear field gradient is applied during the collection of an FID. Since the magnetic field varies across the sample and the NMR frequency is proportional to the magnetic field strength, the resulting NMR spectrum represents a one dimensional image of the sample. The spectrum has a "notch" missing as a result of the plastic disk. The width of the "notch" is proportional to the applied gradient strength and the thickness of the plastic disk. If the thickness of the plastic disk is precisely known, then the strength of the applied gradient can be calculated. The spectrum is rounded at the edges as a result of the gradient strength falling off away from the center of the sample. A modified version of this experiment using an echo is used as a routine method of calibrating gradient strengths.

17 comments:

Anonymous said...

Glenn, some people say that the best way to calibrate gradient strength is via a diffusion measurement using a sample with a well defined diffusion rate constant. Others say to use a phantom as in your post. Do you have an opinion?

Mike

Glenn Facey said...

Mike,

You can certainly use both methods to calibrate the gradients. If you choose to calibrate them based on a diffusion constant, then the temperature inside the probe must be very well calibrated, as diffusion constants are very highly temperature dependant. A good temperature calibration is not always available.

The problem with a phantom sample is that its dimensions must be known to very high precision. A good machinist can provide this. Besides, don't you think that the 1D imaging method is cool?

Glenn

Anonymous said...

Oh yes, there's no question in my mind that the measurement using the phantom sample is the coolest of the two!

Mike

Anonymous said...

Can a phantom be bought, or must one have it machined locally?

Glenn Facey said...

I have never seem a commercial phantom sample meant for the calibration of gradients.

Glenn

Anonymous said...

Glenn,

Do you have any tips on how to make the NMR phantom sample? I have the disk made, but I keep running into issues of airbubbles under the disc. Thanks for you help.

Henry

Glenn Facey said...

Anonymous,

I had the machine shop drill a very small hole through the disk so that the air would be released as I slowly pushed the disk into the tube.

Glenn

Anonymous said...

You can calibrate the temperature using ethylene glycol or methanol and measure the chemical shift between the peaks this is related to temperature, but the temperature at the top of the sample can be different from the bottom, the only way around this it to use variable temperature gas ( air or n2 ) at a high flow rate to minimise these effects. I have found nylon m3 or m4 screws work nicely because you can see the effect of the threads on the 1D image.

Glenn Jones, Bruker

Anonymous said...

Hello,
how would the pulse program (for Bruker) that does what's shown in the picture look like?
Thanks

Glenn Facey said...

Anonymous,

I used a slightly more complicated pulse sequence (one using a Hahn echo rather than a single pulse) than the one pictured. It is as follows:

;gradcalib
;avance-version (02/04/11)
;1D MRI experiment to calibrate gradient strenghts
;Uses a Hahn echo to avoid phase corrections
;
;$CLASS=HighRes
;$DIM=1D
;$TYPE=
;$SUBTYPE=
;$COMMENT=


#include
#include
#include


"d11=30m"
"d7 = d6 + 0.5*p1 -d21 -de"


1 ze
;d11 UNBLKGRAD
2 d1
p1:f1 ph1
d6
p1*2:f1 ph2
d7
d21 gron2
ACQ_START(ph30,ph31)
aq DWELL_GEN:f1
5u groff
rcyc=2
400m wr #0
;d11 BLKGRAD
exit


ph1=0 1 2 3
ph2=0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3
ph30=0
ph31=0 3 2 1 2 1 0 3


;pl1 : f1 channel - power level for pulse (default)
;p0 : f1 channel - small flip angle [2 usec]
;d1 : relaxation delay
;d11: delay for disk I/O [30 msec]
;d15: echo time
;d21: gradient stabilisation delay
;d27: length of dephasing gradient (phase encode time)
;NS: 1
;DS: 0


;this program can be used for dmx as well



;$Id: imgegp1d,v 1.2 2005/11/10 12:17:00 ber Exp $

Anonymous said...

Are you sure that there will no a signal loss in the liquid-phantom interface? I mean are you sure that the "notch" exactly corresponds to the width of the disk, but not the disk plus some interface layer?
Thank you.

Glenn Facey said...

Anonymous,
Thank you for the question. Here I am assuming that the thickness of any interface layer between the water and the plastic disk is insignificant with respect to the thickness of the disk.

Glenn

Anonymous said...

Thank you Glenn,
Do you have an experience with metal discs instead of plastic? I would like to carry out a sort of electrophoretic NMR and in that case I will use lithium metal as electrodes. So may I assume, that in the image I will see whole distance between electrodes?
Best wishes,
Sergey

Glenn Facey said...

Sergey,
Let me first say that I am by no means an expert in MRI. I know that there are certainly problems when MR images are obtained of people with metal implants in that there are very obvious artefacts around the metal implants. I suspect you may have a problem seeing the entire space between the metal electrodes because of artefacts resulting from the large susceptibility difference between the metal and material between the electrodes.
Glenn

Anonymous said...

The rounding, or the general appearance, of the edges of the spectrum/gradient profile is defined by two factors.

First is the gradient uniformity. If the gradient coil generates a perfectly uniform field then one should get a perfect square (assuming everything else is perfect). However gradient coils usually generate non-uniform fields with the gradient strength being stronger in the middle and weaker at the edges. This makes the gradient profiles look like "Batman ears", with characteristic spikes going upward. This is because more spins at the edges experience a more similar field than the ones in the centre so the edges have higher intensity. One can measure the gradient coil uniformity by running a standard DOSY experiment where the observe period is run under a gradient. There is a publication by Gareth Morris describing very nicely how this is done.

Second is the RF coil B1 homogeneity. If the RF coil can deliver exactly the same power along its whole length then the profile appears as a square (if everything else is perfect). If the RF coil is delivering less power at the edges then you see the profile having reduced intensity at the edges. One can measure precisely the B1 homogeneity of an RF coil by comparing the values of the 90, 450 and 810 degrees pulses. One can also make a visual representation of the field by running a 2D experiment where the delay incremented is the pulse width.

In reality the appearance of gradient profiles is defined by a mixture of the two factors mentioned and sometimes a bad RF coil and a bad gradient coil compensate each other and produce a square profile. This is why it is important to measure gradient uniformity and RF homogeneity separately and not judge just by one profile.

Anonymous said...

Is there a publication describing this method for gradient calibration?

Glenn Facey said...

Anonymous,
Stefan Berger and Siegmar Braun, 200 and More NMR Experiments p.455-7 and references therein.
Glenn