There are two possible explanations that I can think of:
1) It is an instability in the template generation, which was
fixed in the updates for SPM8. If you are still using SPM8b, or
the original SPM8, then I would strongly advise installing the
2) The smears happen in regions where there is no data in the
imported images because this part corresponds to a place that is
outside the FOV of all the images (NaNs in the imported images).
It is therefore not so easy to assign a value to the average, so
the algorithm has to extrapolate from values that do exist. If
this is the case, then there should be little or no effect on
the estimated warps.
> I am new to SPM. I have been using DARTEL for inter-subject
> registration of T1 images. Following the manual's instructions, and
> keeping the settings to the default, I have used Dartel's Create
> Template with images segmented with the New Segment function. In the
> final template (template 6) I get a long grey smear above the brain at
> the top of the template (identical on both grey and white matter
> templates) - it is not on any of the other prior templates (1-5), nor
> is it present when I run Create Template using images segmented with
> the old Segment function. I have attached a picture of the final 'new
> segment' Dartel template to illustrate the problem.
> I can easily solve the problem by using the template generated using
> the images segmented with the old segment function. However, I was
> wondering if anyone knew why this problem arose with the New Segment,
> and whether it was a problem with the program, or whether I needed to
> adjust some of the setting?
> Thank you for your help.
> Lauren Cloutman.
John Ashburner <[log in to unmask]>