[Freesurfer] FW: talairach and scullstrip, order and control points, questions

From: Hoekstra, Liesbeth Sent: Wed 12/8/2010 11:11 To: freesurfer@nmr.mgh.harvard.edu Subject: talairach and scullstrip, order and control points, questions L.S., Let me start by saying that you've got a wonderful webpage! I'm a beginner in the Freesurfer

[Freesurfer] Integrate FDG-PET with longitudinal data

Hi! I have ran through the recon-all --long for a series of subjects. For each of the two timepoints, I also have FDG-PET. I want to integrate the PET-data with the longitudinal data. Cross-sectionally I would just coregister T1 and PET, but now that I have the longitudinal runs I wonder how this

[Freesurfer] Two images vs. one image for recon-all?

Dear freesurfers, I believe it is recommended to have two high-resolution images for one subject (same acquisition parameters) for recon-all (i.e. 001.mgz and 002.mgz in /mri/orig). I'm sure recon works, however, with just one image (001.mgz). What is the relative advantage of having two images?

Re: [Freesurfer] Two images vs. one image for recon-all?

Hi Joon, it depends on your field strength, receive coil and amount of acceleration. For example, at 3t with a 12 channel and 2x acceleration 1 acquisition is probably better than two as the blurring from motion correction hurts the CNR more than the noise reduction helps it. At 1.5T with a

Re: [Freesurfer] funcroi-table-sess: average baseline activity?

Hi Yuko, The contrast only allows you to constrain the anatomical ROI. If you don't spec a contrast, then it uses the whole anat ROI. Nothing gets averaged until you run funcroi-table-sess, and that only averages over the given contrast (unless you use -vol in which case it just reports the

Re: [Freesurfer] taskreg in version 5

It depends on what you want to test exactly. You can just weight each presentation and test against fixation as you suggest. This implies that the response amp is 0 when your weight is 0. If that's not the case, then you need to make a modification. Let me know if you need that. doug Katie

Re: [Freesurfer] Integrate FDG-PET with longitudinal data

There's nothing special that we have set up to do this. I would register both TPs to the base. doug Per Selnes wrote: Hi! I have ran through the recon-all --long for a series of subjects. For each of the two timepoints, I also have FDG-PET. I want to integrate the PET-data with the

Re: [Freesurfer] taskreg in version 5

I'm not sure if that's exactly what I am saying. Basically, there is an area in the brain (which I am trying to localize) that tracks the number of items you are holding in memory so that the more items you hold in memory, the higher the bold activation. We have behavioral results saying how

Re: [Freesurfer] taskreg in version 5

It comes down to whether you think there will be 0 activation in this area if there are 0 items held in memory. If so, then what you have is fine. doug Katie Bettencourt wrote: I'm not sure if that's exactly what I am saying. Basically, there is an area in the brain (which I am trying to

Re: [Freesurfer] asegstats2table in version 4.2.0 error?

Hi Jesse it seems the standard Python is not installed with your CentOS installation. asegstats2table is a Python script and needs it. You could either ask your sysadmin to install the Python. or if you have the permissions.. do it yourself -- some variant of sudo yum install python Best,

Re: [Freesurfer] asegstats2table in version 4.2.0 error?

Hi Krish, Thank you very much for your time, I will install the Python. Jesse On Wed, Dec 8, 2010 at 1:43 PM, Krish Subramaniam kr...@nmr.mgh.harvard.edu wrote: Hi Jesse it seems the standard Python is not installed with your CentOS installation. asegstats2table is a Python script and

Re: [Freesurfer] taskreg in version 5

You would represent each presentation as two different conditions, one with a weight that is always 1, the other weighted according to your scale. So this doubles the number of conditions. You would then test the weighted conditions. The weight=1 conditions represent an intercept, and the

[Freesurfer] CSF masks

Hi, We are trying to create a CSF volume mask from structural MEMPRAGE T1 data (we also have ME FLASH 5/30 -PD if needed). We are interested in two CSF compartments: - superficial CSF outside grey matter and inside inner skull - deep CSF inside ventricles Is there a way to simply extract these

Re: [Freesurfer] CSF masks

Hi Tommi you can use mri_extract_label to pull out the labels you want, but the sulcal CSF labels may not be great since the manual labeling that is the basis of the aseg was done on only T1 images. cheers Bruce On Wed, 8 Dec 2010 r...@nmr.mgh.harvard.edu wrote: Hi, We are trying to

Re: [Freesurfer] CSF masks

Thanks Bruce! However, as you suspected this does not seem to work: mri_extract_label mri/aseg.mgz 24 mri/mask_code24.mgz gives a tiny volume of a couple of hundred voxels in the middle of the brain (part of 3rd ventricle). I assume here that code 24 is the correct code for all CSF in aseg.mgz.