 |
 |
Hi Suchada, If you use the defaults, then what you suggest is correct. For step 2 you can use the original or eddy corrected dataset. In general, you need to run BET on the volume used as reference in the eddy current correction step. This is by default the first volume in the dataset, which is not changed by eddy correct. So, it won't matter if you BET this volume using the original or the corrected dataset. Cheers, Stam On 11 Oct 2011, at 14:15, Suchada Tantisatirapong wrote: > Dear Stam, > > So I should do DTI processing as: > 1. Eddy current correction on original 4D data > 2. BET on eddy current corrected 4D data to get binary brain mask. > 3. FDT: for diffusion weighted data file, select the eddy current corrected image on the original data, not the brain extracted image. > > Thank you so much for very quick response. > Regards, > Suchada > > > > -----Original Message----- > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Stamatios Sotiropoulos > Sent: 11 October 2011 14:05 > To: [log in to unmask] > Subject: Re: [FSL] Errors: data and bvals/bvecs do not contain the same number of entries > > Hi Suchada, > > The only thing that you need from BET is the binary brain mask. You should use the 4D data (after eddy correct) as input to the FDT pipelines. > > Cheers, > Stam > > > > On 11 Oct 2011, at 12:09, Suchada Tantisatirapong wrote: > >> Dear Stam, >> >> Yes, >> the original data: the volumes is 42. Voxels: 104x108x55. Dimension:2.22 x 2.22 x 2.5 mm >> after eddy correction: the volumes remained the same. But after BET, the volumes changed to 1. >> >> Why is that so?. >> Many thanks >> Suchada >> >> >> -----Original Message----- >> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Stamatios Sotiropoulos >> Sent: 11 October 2011 11:43 >> To: [log in to unmask] >> Subject: Re: [FSL] Errors: data and bvals/bvecs do not contain the same number of entries >> >> Hi Suchada, >> >> the number of entries in bvals and bvecs should match the number of diffusion-weighted *volumes* in your dataset, it has nothing to do with the number of slices. Your data file should be 4D, with the 4th dimension indicating the diffusion-weighted volumes. If you open the data file with fslview and press the 'i' button, the number of volumes is provided. Is this 42? >> >> Stam >> >> >> On 11 Oct 2011, at 11:05, Suchada Tantisatirapong wrote: >> >>> Hi, >>> >>> I'm running FDT diffusion and got the error saying 'Errors: data and bvals/bvecs do not contain the same number of entries'. >>> There are 42 DTI slices/subject. I used dcm2nii to convert dicom to nii.gz. dcm2nii generates 42 values in bvals file and 3 x 42 matrices in bvecs file. But FDT still complains about mismatch entries. >>> >>> Besides, in some of my data set having 55 dicom slices, the dcm2nii could not extract bvals and bvecs from these data set. No bvals and bvecs file generated at all. >>> >>> Could you please tell me why and how to solve this? >>> Thank you so much >>> Suchada >>> >> >