help
help > RE: Flat activity distribution
Jan 25, 2016 06:01 AM | Alfonso Nieto-Castanon - Boston University
RE: Flat activity distribution
Hi Meghan,
Those plots represent the histogram/distribution of voxel-to-voxel connectivity values before and after Denoising. The flat-looking distribution (before denoising) may be perhaps indicative of the presence of a few strong outlier scans, but that seems to be relatively "fixed" after denoising so there is probably not reason for concern. I would suggest to look at whether the "after denoising" distribution for this subject is similar to the corresponding distributions for all other subjects (e.g. click on the 'display all' button below the histogram display to look at these distributions superimposed for all subjects). If that is not the case then yes, there would be some reason for concern, and you might want to either apply a more aggressive form of Denoising (e.g. more conservative outlier removal thresholds) or consider removing this subject from your second-level analyses.
Hope this helps
Alfonso
Originally posted by Meghan Caulfield:
Those plots represent the histogram/distribution of voxel-to-voxel connectivity values before and after Denoising. The flat-looking distribution (before denoising) may be perhaps indicative of the presence of a few strong outlier scans, but that seems to be relatively "fixed" after denoising so there is probably not reason for concern. I would suggest to look at whether the "after denoising" distribution for this subject is similar to the corresponding distributions for all other subjects (e.g. click on the 'display all' button below the histogram display to look at these distributions superimposed for all subjects). If that is not the case then yes, there would be some reason for concern, and you might want to either apply a more aggressive form of Denoising (e.g. more conservative outlier removal thresholds) or consider removing this subject from your second-level analyses.
Hope this helps
Alfonso
Originally posted by Meghan Caulfield:
Hello,
First of all I want to thank you - I use these boards a lot to help troubleshoot analyses and your detailed and helpful responses are appreciated!
I am doing an analysis in acute Stroke and there is a funny activity pattern in one of my participants. I attached a screen shot from the denoising stage. I am wondering if I should be concerned about this distributed pattern of activity? None of the examples available in the user guides look like it, nor does my other participant. Could it be that they fell asleep in the scanner? (its been known to happen in our population).
Thank you for your advice!
First of all I want to thank you - I use these boards a lot to help troubleshoot analyses and your detailed and helpful responses are appreciated!
I am doing an analysis in acute Stroke and there is a funny activity pattern in one of my participants. I attached a screen shot from the denoising stage. I am wondering if I should be concerned about this distributed pattern of activity? None of the examples available in the user guides look like it, nor does my other participant. Could it be that they fell asleep in the scanner? (its been known to happen in our population).
Thank you for your advice!
Threaded View
| Title | Author | Date |
|---|---|---|
| Meghan Caulfield | Jan 22, 2016 | |
| Alfonso Nieto-Castanon | Jan 25, 2016 | |