While I’m at it, I thought I might also include my new, finalized abstract:

The focus of this project was to develop the methodology for reconstructing and analyzing dog brains from MRI scans for the comparison of the volumes of different subcortical structures, so that such differences may be correlated with personality trait differences between dog breeds. Existing MRI atlases have insufficient detail or slice thinness to produce accurate reconstructions, and histological atlases cannot be easily reconstructed.  Recent personality research has identified regions in the human brain that correlate with the Big Five personality traits.  As dogs have the widest variation in terms of size and personality traits, they were selected as models for evaluating the neuroanatomical correlates of personality.  A temperament survey was administered to over 3500 dogs of different breeds, and the data were analyzed to produce breed scores for traits measured by question groupings.  Breed comparisons that exhibited the greatest contrast on these traits were isolated. MRI scan parameters for a fixed dog brain were chosen to accommodate the changes that occur with fixation, and high quality scans of this reference brain were conducted.  These scans were then segmented and reconstructed using Amira™ into 3-dimensional representations of the brain surface and volume as reference material.  Other dog brains were reconstructed from multi-axis clinical MRI scans and morphed to fit the reference brain, so that the scaling parameters may be obtained.  The parameters will be used to rescale the segmented volumes of subcortical structures.  These standardized volumes will then be compared and the differences will be correlated with the obtained personality differences.

Hello all,

My work here at Princeton is just about done for the summer.  I got some really high quality MRI scans at UPenn, and I have been able to do 3D reconstructions of that brain and some clinical brains.  The process for the volumetric analysis is pretty much set, and all of the personality data has been processed, so we now know what breed comparisons we are looking for (apparently while Border Collies are very trainable, Beagles are basically impossible to train).  Just finished up my poster for this project, which outlines the whole process, so that next year the y can actually get all of the scans they need and get some results.  Still have another week here, and I will most likely be putting together a guide on how to do what I have done.  Good luck wrapping up all of your projects!

After our first scanning session, the MRI coil is broken again!  We will have to wait until August until it is fixed.  This, of course, was unacceptable, so I started to look for other options.  The new scanner here at Princeton is also a 3T scanner, but is mostly meant for humans.  They have a flexible extremity coil that *may* give us decent scans, if I can get my hands on some PVC pipe to protect the brain.  We won’t be able to use this scanner until the 21st at the earliest.  I have also been in correspondence with UPenn to see if we could get time on their higher power scanner (9.4T).  They agreed, but the mass of red tape at both institutions has greatly hindered progress on that front.  Just today, however, I made something of a breakthrough and should hopefully have scan time scheduled within the next week (fingers crossed).  The data analysis also hit some road blocks in terms of statistical problems, but I think that we have almost resolved those issues.  I just hope that it all works out.  With a little luck and persistence, I may eventually have something to show for my work.  For now, I will leave you with what we got from our first scanning session.

See a full scan here!