Re: Standard space in ExploreASL

[Mutsaerts, H.J.M.M. (Henk Jan)]

Thanks, Tayebe, your confusion is understandable. The standard space that we use in ExploreASL is the same as CAT12 uses, which is the IXI that you referred to, but resampled to 1.5 mm isotropic, which indeed has fewer voxels if the original space had a higher spatial resolution (eg 1mm isotropic). All atlases and templates that we use are also resampled to the same spatial resolution. 1.5 mm seems to be a nice balance between computing resources and orientation differences between subjects and studies. 

Hope this helps, have a good weekend, 

Henk

---

From: tayebeh ebrahimi <tayebehe...@yahoo.com>
Sent: Friday, November 6, 2020 4:44:31 PM
To: Mutsaerts, H.J.M.M. (Henk Jan) <h.j.mu...@amsterdamumc.nl>
Subject: Standard space in ExploreASL

 

Dear Dr. Mutsaerts,

As far as I have understood, rT1_sub**.nii and also qCBF_sub**.nii outputs, which are in the population folder, both have 121* 145* 121 dimensions. I also have read the main ExploreASL article entirely. It has been written there that the standard space in ExplreASL is IXI555. But regarding what I searched, the dimensions of IXI555 is 150* 256* 256.

I really got confused about these files and their dimensions. Could you do me favor and clarify this issue too?

Thank you in advance.

Best regards,

Tayyebe

______________________________________________________
VUmc disclaimer : www.vumc.nl/disclaimer
AMC disclaimer : www.amc.nl/disclaimer

[Mutsaerts, H.J.M.M. (Henk Jan)]

Hi Tayebeh, 

Indeed, ensuring that you inspect and/or analyse the correct hemisphere is not trivial, since the hemispheres have a too similar appearance for algorithms to automatically detect that they have not been flipped during the process.

Any other flips (e.g. superior/inferior or anterior/posterior) are easy to observe by a human or machine.
What ExploreASL does (and many others):

1) The MRI scanner saves the original orientation of the image (including the left-right orientation) as it comes from the scanner (in the NIfTI header this is stored as .mat0).

2) After image processing, there is a test that compares if there has been a left-right flip in the new orientation (after image processing, which is stored as .mat in the NIfTI header), with the original orientation (.mat0). This would issue a warning and can be checked in the QC, e.g. in the PDF report you have the parameter “TL1_LR_flip_YesNo” for the T1 (and similarly for other scan types), which should be 0 (no flip).

3) MRICron is certainly a helpful visualization tool. There is an option to view the image in the “true orientation”, which is referred to as the “world coordinates”. You can activate this by checking >HELP>Preferences>Reorient images when loading.

Hope this helps!

Get Outlook for Android

---

From: tayebeh ebrahimi <tayebehe...@yahoo.com>
Sent: Friday, November 6, 2020 8:19:33 PM

To: Mutsaerts, H.J.M.M. (Henk Jan) <h.j.mu...@amsterdamumc.nl>

Subject: Re: Standard space in ExploreASL

 

Dear Dr. Musaerts,

Thank you so much indeed for your prompt reply. Your explanations were quite helpful. I just have another question. It is kind of you to answer it. I have a nii file in which the lesion is on the left side according to T1 images.  But when I open that nii with MRIcron (because in WalkThrough, you recommended using MRIcron for inspection visually), That lesion is on the right side. Could you please explain a bit little about this also? 

Best,

Tayebe