I tried BIS 3.01 and 2.61.  Both have SPECT Processing Tool to obtain subtraction image of interictal - ictal.  I loaded the difference_normalized image of the sample case but the subtraction seems misaligned with only 2 1/2 of the axial slices got data and the rest of the slices look like garbage.  The ISAS results from the same datatree look good comparing to the coronal slices from the Yale web site.

What did I miss?

If I want to trouble shoot which Tcl script I might start with?

Rgds,
Marco
lsk203@ha.org.hk

If the ISAS results look good, I don't think the issue are the registrations. My guess is that the "space" image was set incorrectly or the overly threshold for the difference image need to be lowered.
 
If you put the difference image into a viewer in native space, how does that look? It should have data on nearly every slice.
 
If that looks ok, try a lower threshold when creating the overlay. Maybe something like 10, just to see if it helps. By the default the low threshold on the overlay tool is a bit high for the difference image.
 
Dustin