Hi,
I will appreciate if somebody could reply to my message that was posted earlier. I copy-paste the message here again.
We are currently in the process of analyzing some data that was defaced using mri_deface command of freesurfer. This defaced data was then used as an input to recon-all and the volume and thickness was computed. These results were then
compared against the same subject’s recon-all with the face but with and without the deface flag within recon-all. All of the steps were done on version 5.3.0 on CentOS. We have got some differences and we are unable to explain why. Some of the regions have
error greater than 20% when the input was a defaced brain. We would appreciate if you could advise as to how to approach to the same results as we were running the data with the face .
To summarize: (The commands are not precise below but we ensured the commands are run correctly per the command help)
1)
Step 1: mri_deface -i original_T1 -o deface_T1
2)
Step 2: recon-all -i deface_T1
3)
Step 3: recon-all –i original_T1 –deface
4)
Step 4: recon-all –i original_T1
5)
The results from Step 3 is exactly equal to Step 4
6)
The results from Step 2 and Step3/4 have errors >20% in certain regions.
7)
Step 1-4 are done with version 5.3.0
Any response will be greatly appreciated.
Thanks
Regards
Virendra
===================================
Please consider the environment before printing this e-mail