Good question! I'm trying to work out the same thing so I can
automate the process. That probably won't happen soon enough
to help you, though.
You can get a measure of the autocorrelation (first order) in your
data using the 3dTstat -DW option, which generates the Durbin-
Watson statistic.
As far as correcting it, somebody else could help here, or else
I'll get back to you next week.
rich