adhd200preproc
adhd200preproc > RE: Getting Started (for non-neuroscientists)
Jun 6, 2011 10:06 PM | Pierre Bellec
RE: Getting Started (for non-neuroscientists)
Originally posted by Cameron Craddock:
Best,
Pierre
(1) The fMRI data is acquired in at least 3x3x3
mm space, (...) I balanced these tradeoffs and came up with 4x4x4.
(...) Pierre and I tend to disagree a bit on the issue of
what is the best resolution for fMRI analysis, so it will be
interesting to see what he has to say...
Haha I am glad this came up. Cameron and I disagree ... on about 1
mm. The NIAK release is resampled at 3*3*3 mm3. The rationale in my
case is that it produces nicer maps, and I don't care too much
about disk space. I also use the region growing trick to avoid
computing connectomes at the voxel resolution, so the initial
resampling resolution does not matter too much. I have never looked
at the impact of this parameter on the results of a clustering
analysis, but I guess it will be minor.Best,
Pierre
Threaded View
| Title | Author | Date |
|---|---|---|
| Niels van Galen Last | May 15, 2011 | |
| Niels van Galen Last | Jun 6, 2011 | |
| Cameron Craddock | Jun 6, 2011 | |
| Niels van Galen Last | Jun 7, 2011 | |
| Cameron Craddock | Jun 7, 2011 | |
| Pierre Bellec | Jun 6, 2011 | |
| Niels van Galen Last | Jun 1, 2011 | |
| Niels van Galen Last | May 26, 2011 | |
| Cameron Craddock | May 27, 2011 | |
| Niels van Galen Last | May 25, 2011 | |
| Cameron Craddock | May 25, 2011 | |
| Cameron Craddock | May 26, 2011 | |
| Pierre Bellec | May 16, 2011 | |
| Niels van Galen Last | May 16, 2011 | |
| Cameron Craddock | May 16, 2011 | |
| Pierre Bellec | May 16, 2011 | |