Re using First to assess pallidal and caudate percentage volume change (PVC)
I have a few different groups, but the interest in assessing same-subject (within subject) pallidal and caudate volume atrophy; volume loss between baseline and at 24 months. For each subject, there are 2 pairs of baseline T1s and 2 pairs of month 24 T1s. Each scan pair was taken only seconds apart and while the subject remained in the scanner. As such, and just considering clinically normal controls, it should be possible to confirm consistency or reproducibility of within-subject volume loss assessment by extent of difference between same subject paired scans (Steve Smith et al. 2007 did this for the whole brain comparing among BSI, Siena and Sienax).
To elaborate, if a single baseline T1 is labeled Vaa (one T1) and another single same subject T1 is labeled Vbb, and the same subject month 24 scan pair are each labeled Vaf and Vbf, consistency in percentage volume change could be assessed as follows:
(1) derive segmentation-based volume using FRIST
(2) find the structure percentage volume change for each subject’s baseline (Via, Vbb) and month 24 (Vaf, Vbf) (same-subject) scan pair
scan pair (a)100* (Vaf - Vaa/Vaa)
scan pair (b) 100* (Vbf -Vbb/Vbb)
(3) Find the consistency of volume change from the difference between same-subject scan pairs
PVC difference= (scan pair a) - (scan pair b), where closer to zero is more consistent
Also, does the below naming convention look OK for First: all group base names are the same and have the same number of characters; each group will be run separately; all FIRST processing run within on directory
baseline, base names
conaa01 conaa02 conaa03…
conbb01 conbb02 conbb03…
month 24 base names
conaf01 conaf02 conaf03…
conbf01 conbf02 conbf03…
Any feedback would be much appreciated
Best,
Lance
|