[Mristudio-users] Rotate gradient if applicable

[susumu mori]

Hi all,

Because the gradient rotation is a very tricky issue, I want to revisit it.

1) First of all, this is a complicated issue. Manufacturer may change the
way they perform diffusion weighting and store the information during
upgrade. So, as I mentioned before, you need to do several testings by
yourself. It is very difficult to provide a tool that is guaranteed to work
in all situations.

2) You don't have to worry about this issue if you are not using oblique
imaging, meaning you always use simple sagittal/coronal/axial scans without
FOV (field of view) rotations.

3) When FOV is rotated (e.g. AC-PC alignment or to compensate brain tilt),
there are two potential options. Suppose a DWI is acquired with [1, 0, 0]
(diffusion along X ). In the first option, after the FOV rotation (say, 5
degree about Z axis), [1, 0, 0] remains "physical" X gradient. So, oblique
rotation has no impact on the gradient application. In the second option,
[1, 0, 0] means diffusion weighting along FOV (image matrix) X axis, which
is no longer the same as the physical gradient X axis. In this case, the
actual diffusion-weighting is [cos(5), sin(5), 0]. Namely, the diffusion
table rotates with the FOV.

4) As fur as I know, Philips and GE currently do the latter ([cos(5),
sin(5), 0]) and Siemens does the former ([1, 0, 0]). However, for Philips,
if an option called "GradientOverplus" is ON, it does the former ([1, 0,
0]).

5) If your scanner does the dynamic diffusion table rotation, following the
FOV rotation, you do NOT want to click the "rotate table" option. This
means, Philips and GE data usually do not have to rotate table.

6) If your scanner doesn't do the table rotation ([1,0,0] remains physical
[1, 0, 0] regardless of the FOV rotation), you NEED to click the "rotate
table" option. This means, we usually choose this option if you are using
Siemens.

7) I am not sure what would happen if you use "user-defined" options in
these scanners. Also, the way the gradient table information is stored in
the header files (does "X" means physical gradient X or image matrix X)
 may not be consistent among manufacturers as well as different files types
with in the same manufacturer (e.g. DICOM and PAR/REC formats of Philips).

7) I want to stress that it is very difficult for us to give you the
correct answers because there are so many variables and potential changes
after scanner upgrades. Therefore, we always suggest you to do small tests.

Test
A> obtain one set of DTI. Usually you don't see noticeable difference
between "rotate gradient" ON/OFF
B> Then obtain another set of DTI with a severe oblique angle (like 30
degrees). The brain looks rotated. Now you see difference between "rotate
gradient" ON/OFF
C> (optional) you can ask the subject to rotate the head significantly
(like 30 degrees) and rotate the FOV to make the brain looks upright in the
image. Again, you should see difference between "rotate gradient" ON/OFF.

If you are processing your data correctly,
The color map of B should look very different from A.
C should give you a color map similar to A.

In all cases, tractography should work and produce similar results.
If A color map has similar appearance as B color map, or C color map
doesn't look like A color map, then your DTI calculation is not working.
Please check if it can be solved by checking or unchecking the "rotate
gradient" option.

I would appreciate users to share your findings or report any errors in my
statement. We have an access to only a limited number of scanners and it is
difficult to test different scanners.

Susumu


On Fri, Feb 22, 2013 at 9:16 PM, Durai Arasan <durai23 at gmail.com> wrote:

> and this
>
> http://lists.mristudio.org/pipermail/mristudio-users/2010/001359.html
>
>
> On Fri, Feb 22, 2013 at 9:13 PM, Durai Arasan <durai23 at gmail.com> wrote:
>
>> this might help
>>
>> http://lists.mristudio.org/pipermail/mristudio-users/2010/001868.html
>>
>> On Fri, Feb 22, 2013 at 11:28 AM, Vanessa Douet <douet at hawaii.edu> wrote:
>>
>>>  Hi,
>>>
>>> I am a very new user of the MRI studio. I am learning how to use your
>>> software. Could you explain to me when and why the "rotate gradient if
>>> applicable" has to be checked ?
>>>
>>> Thank you very much
>>>
>>> Vanessa
>>>
>>> --
>>> Vanessa Douet, PhD
>>> Department of Medicine
>>> John A. Burns School of Medicine
>>> The Queen's Medical Center
>>> 1356 Lusitana Street, UH Tower, 7th floor
>>> Honolulu, HI 96813
>>> (808) 586-7468
>>> _______________________________________________
>>> mristudio-users mailing list
>>> mristudio-users at mristudio.org
>>> http://lists.mristudio.org/mailman/listinfo/
>>> Unsubscribe, send a blank email to:
>>> mristudio-users-unsubscribe at mristudio.org
>>>
>>>
>>
>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20130223/98dce1d7/attachment-0001.html