History of AFNI updates  

|
May 27, 2009 06:00PM
Hi Andrew,

That motion is tiny, but looks fine (i.e. a great subject! :). It
looks like 7 valid runs of maybe 144 TRs per run (based on motion
between runs).

In your case, 720 is a multiple of the reps (TRs per run) and is
therefore from motion between runs. There is no need to censor such
motion. However yes, you could feed 720 in via -CENSORTR. While
720 is a 'global' TR, your -stim_times do not need to be global,
that would be a separate consideration.

---

It looks like 1dFlagMotion should be executed once per run.

Note that 1d_tool.py has a -derivative option, as well as -set_nruns.
So you could use something like example 7 (1d_tool.py -help) to help
look for large motion.

---

We do not generally use 3dFourier (as you can tell since afni_proc.py
does not even include the ability). The polynomial baseline in
3dDeconvolve serves as a low-pass filter, while tracking degrees of
freedom.

- rick

Subject Author Posted

Censoring Motion Points

Andrew Jahn May 27, 2009 05:16PM

Re: Censoring Motion Points

rick reynolds May 27, 2009 06:00PM

Re: Censoring Motion Points

Andrew Jahn May 28, 2009 04:50PM

Re: Censoring Motion Points

rick reynolds May 28, 2009 06:14PM