Hi Maria,
I'm a little confused about how many volumes you have actually
acquired (after pre-steady state, say), 50, or a couple more than
50? If you have more than 50, there is no reason not to include
them in the analysis.
In any case, regardless of your design, it does no good to stimulate
the subject at the end of a run (after you stop acquiring data).
There is also little to gain by throwing away volumes at the end of
a run. They will often help to distinguish between baseline and
task.
- rick