you need to run recon-all on your T1.nii, then register your spectroscopy images to the recons (we use bbregister for this). Then you can use our segmentations to analyze the spectroscopy data.

cheers, Bruce On Tue, 2 Feb 2010, magnet scanner wrote:

Hi, Bruce,

Thanks for your quick response. here is what I am going to do with my spectroscopy data;

1. mri_convert to convert all my T1 and T2 images to nifti format mri_convert -it dicom {T1 filename} -ot nii {T1.nii} --out_orientation

RPS mri_convert -it dicom{T2 filename} -ot nii {T2.nii} 2. obtain affine matrix from T1.nii and T2.nii hdr. 3. register T1 to T2 and align with my spectroscopic data. 4. do segmentation 5. assign my spectroscopic data to segmented images

I used FSL bet command for segmentation before, and it does not change anything. However, I need to obtain the information about subcortical structure like Hippo, Thal, etc. Therefore, I used freesurfer segmentation tool. But since conform is the necessary step, the volsizes for the segmented region are changed, the coordinates are changed. So my affine matrix is different. My questions are here again:

1. In that case, do I need to conform T1.nii and T2.nii as well?
2. I did conform and T2.nii looks weird to me.

Thanks,

Jun

On Tue, Feb 2, 2010 at 12:31 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

...

Hi Mag,

no, the conform is a necessary step, but it wouldn't be applied to your spectroscopy data in any case. You would register that to the anatomicals (that were conformed). Is that what you mean? Bruce

On Tue, 2 Feb 2010, magnet scanner wrote:

Hi,

...

Please excuse me if this is an old question. I was wondering if there is a way of not doing conform when segmentation on the whole brain since I am dealing with spectroscopy data and need affine matrix. So I have to make sure the matrix is the same.

Thanks,

Mag