Hey Ben,
Thanks for calling attention to my error.
One of the reasons I posted the original questions was to attempt to understand why the WFU_pickatlas and the Rex program produce different extraction values for voxel count, mean, etc. After taking a long look at the coding, I have come to the realization that the answer lies with the "hold" variable in the spm_sample_vol function. The "hold" variable (according to the code comments) sets the interpolation method for the resampling.
The WFU_pickatlas generate table function sets the "hold" value to 1 (first-order hold or trilinear interpolation), whereas the Rex program sets it to 0 (zero-order nearest neighbor). (Well actually the Rex program uses spm_get_data, which has 0 as the default "hold" variable setting).
Would you or any of the other exports out there care to comment on the merits of each approach? If one wants to extract volume information from a VBM image using a binary ROI, which method of interpolation would be most appropriate?
Thanks again for the responses,
Doug Merkitch
Neurological Sciences
Rush University Medical Center
Phone: (312) 563-3853
Fax: (312) 563-4660
Email: [log in to unmask]
|