Print

Print
No, what I mean is that if each subject group was scanned on different scanners, then when doing a group analysis, you would not be able to determine whether any effect is due to group membership or scanner.

If you had a mix of patients & controls on both scanners A & B, then a group analysis would be fine.
On 01/26/2015 12:30 PM, Seongtaek Lee wrote:
[log in to unmask]" type="cite"> 
Thank you for your comment.

Sorry I am a little confused. Do you mean that if all subjects' scans were not collected from the same scanner, then I won't be able to do a proper group analysis? Is this what you are saying? Could I listen more detail on this like what things would be problem or something like that?

Best regards,

Jason
________________________________________
From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Watson, Christopher <[log in to unmask]>
Sent: Saturday, January 17, 2015 7:42 PM
To: [log in to unmask]
Subject: Re: [FSL] About different sequences in TBSS analysis

If there is no overlap in group membership and scanner, then you will not be able to do a proper group analysis (e.g. control FA > patient FA).
________________________________________
From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of Jason S. Lee [[log in to unmask]]
Sent: Saturday, January 17, 2015 5:41 PM
To: [log in to unmask]
Subject: [FSL] About different sequences in TBSS analysis

Hi Fslers,

Right now, I am trying to figure out the effect of the difference of DTI sequences.
To answer it, my boss wants me to compare 23 randomly selected epilepsy patients with 23 healthy controls.
The twenty three patients' DTI scans were collected from a scanner (let's say Scanner A) and healthy controls' DTIs were collected from the other scanner (Scanner B). So, the scan parameters for patient and control groups are not identical.
The patients group has diverse seizure-occurring locations such as left/right of temporal, frontal, parietal lobes, and left/right hemisphere. So basically, I tried to mix diverse seizure-occurring spots as much as I can.

The thing is that honestly I don't know how I can check the effect of the sequence differences in the way my boss suggested.

I assume that the TBSS result is supposed to present changes in FA or MD in "both sides."??
However, I am not confident with my assumption. Would anyone please give me some advice in this issue?
Your help in this would be greatly appreciated in advance.

FYI, I am using randomise with n=5000, and unpaired t-test for group comparison.

Thank you!