Thank you for the email, Chris!
Following up on the second point, I have several scans with very bad anatomicals that I rejected (much worse than the ones I attached in my previous message). I will make a note of it to ensure proper structural is acquired in case of motion, but most of the data has already been acquired. My question is, how can I preprocess the fMRI without the structural? My pipeline is to coregister the structural to the anatomical, segment the coregistered anatomical, and write using normalize:write with the segmentation parameters. If the anatomical is bad, wouldn't that break the pipeline? How would I go about approving and using the functionals with the structurals? Thank you.
'Sincerely,
- Ananth
On Tue, 3 Jun 2014 13:34:56 -0400, Chris Watson <[log in to unmask]> wrote:
>1. I think this is something you'll need to do on a case-by-case basis.
>The example you posted doesn't look *too* bad, and I think Segmentation
>will do fine. I can't give any specific rejection criteria; I always use
>my own judgment and if I think the anatomical scan was bad, I will
>repeat it at the end of the study.
>
>2. No, I wouldn't. If you reject the anatomical scan, you won't be able
>to do volumetrics (probably), but you can still keep the functionals.
>
>On 06/03/2014 11:54 AM, Ananth Narayanan wrote:
>> Hello all,
>>
>> I am encountering some mild to severe ringing and blurring in the anatomical T1 scans in my study. I have attached 2 files with screenshots of these observations (different subjects). My question is, at what point do these artifacts start to prevent proper coregistration (to functional) and segmentation/normalization to template (using TPMs in segmentation)? What is the rejection criteria for the anatomical scans with artifacts? Obviously we want to reject as little as possible and use a maximum subset of the data, but when do I draw the line? What are the variables that I consider for rejection?
>>
>> Secondly, I have some studies that have artifacts in the anatomical scan, but with pretty good functional scans, no venetian blinds or anything readily observable. If I disapprove the anatomicals, should I automatically disapprove the functionals? Is it possible to salvage them in some way? Thanks in advance!
>> 'Sincerely,
>> - Ananth
>
>
|