Posted in:



1. How do I use the Siemens_interleave parameter?
Possible parameter values: 0, 1, or undefined (blank)
If the session was scanned on a Siemens Trio (not Allegra, not Vision, not Skyra) scanner, set this parameter to 1 (or any nonzero value).
Otherwise: set this parameter to 0 or leave undefined.

When Siemens_interleave=1, the BOLD processing reorders the slices based on the following Siemens Interleave Slice Reordering scheme:

If TOTAL # SLICES = EVEN, then the excitation order when “interleaved” is EVENS first, ODDS second, e.g.
2,4,6,8….. and then 1,3,5,7….

If TOTAL # SLICES = ODD, then the excitation order when “interleaved” is ODDS first, EVENS second, e.g.
1,3,5,7….. and then 2,4,6,8…...

2. What does the skip parameter do?

If skip is set to, say 5, and 192 frames are acquired, deband_4dfp
would skip the first 5 frames and create the "debanded" image
(homogeniety corrected image). ALL frames are debanded.
The pre-steady state frames (1 to skip) are debanded individually. The steady state frames are debanded as a group (one set of debanding constants
applied to all steady state frames). The skip decides how many frames it takes to reach "steady state".

However, for all the subsequent steps
and images generated (r3d, xr3d_atl etc) the 5 frames would be
included. The skip parameter only impacts which frames are selected
for the deband_4dfp step.

3. What should the value of skip be for TRIO scans?

Regarding the optimal number of skipped frames on Trio data,
the issue is a bit complicated. The original reason to skip frames was
so that various computations could avoid being affected by
pre-steady state magnetization effects, which manifest principally
as grossly more signal. Since the Trio does not allow pre-steady state
data to be saved it would seem that "0" is the correct skip.
However, it has recently been realized that there are cognitive
scan start effects in Trio data (probably induced by the scan onset
signal which is obvious to the subject as gradient swtiching noise).
So, even on a Trio, many experimentors are skipping frames for reasons
other than steady state magnetization. In practice, these congitive effects
make no different to generic preprocessing. They would come into play
in resting state fcMRI and theoretically also in task-based fMRI.