open-discussion > PET analysis
Showing 1-3 of 3 posts
Display:
Results per page:
Jan 8, 2016  07:01 PM | Larry Olson - Emory University
PET analysis
I have developed some remarkably useful methods for analyzing PET, and I would like to supplement them with either SPM or using the BIS multisubject tools. 
1) I have a normal dataset. Is anyone aware of a sharable normal data set of attenuation corrected PETs and corresponding T1s?
2) I am not sure from the manual if the methods used by BIS are different than or the same as SPM?
3) (Biggest issue:) It is not clear from the manual which tools to use and how to use them to compare a single patient subject against a normal group. I can create the group average, T-map and STD files. But if I try to run a comparison against a "multisubject group2" consisting of a single image, the anaysis doesn't do anything (crashes or empty result). I guess the dual multisubject tools are designed for groups of more than one case, yet I know it is possible to somehow do this one subject against normal controls analysis- after all that is what is done with the ISAS method.

I will be appreciative of any helpful guidance...

Thanks in advance.
Jan 11, 2016  06:01 PM | Xenophon Papademetris
PET analysis
For item 3:

Use the command line tool bis_ttest (or bis_ttest.tcl) with an optional -dogui 1 flag. This will take 2 4D datasets (1 frame/subject) and compute the t-test between the groups. The multisubject tool can be used to generate the 4D image for group 1 and the second group will just be the single subject if you set the correct option (singlegroup)
Jan 26, 2016  12:01 AM | Larry Olson - Emory University
RE: PET analysis
Thank you!

Works perfectly.