Dear David,
As there are several programs specialized on animal data it wouldn't be surprising to find some more atlases or templates, plus those "hidden" ones which you can only receive by asking people, for which you have to stumble across some papers in the first place. The same holds for mouse MRI, see e.g. the Suppl. 2 in Ma et al. (2014, PLOS ONE, https://dx.doi.org/10.1371/journal.pone.0086576 ). Even if they haven't been used in combination with SPM, maybe not even constructed for image registration purpose, they might be worth a try. It would be just important to make sure about the spatial resolution, instead of adjusting the voxel size accordingly some of the images are upscaled so that1 mm in image space corresponds to maybe 0.25 mm in world space.
As an anecdote, some years ago I worked a little with rat data acquired on a clinical scanner combined with a custom-built coil system. In the T2 and EPI data we had quite some signal drop with increasing distance from the coil elements due to B1 inhomogenities. When trying to normalize onto a template with uniform signal distribution this failed all the time, but segmentation-based normalisation with another template including tissue probabiliy maps did work well (which can be explained by the bias-correction during segmentation that accounts for intensity inhomogenities, which we could have applied to our data as well to then go with the other template). Another difference was the cropped olfactory bulb in one of the templates. If you're especially interested in this region then that template would probably be suboptimal. It might not necessarily be that straightforward to see the pros and cons of a specific template though.
Best
Helmut
|