One goal of this blog is to share some of our tips and tricks for CT scanning, 3D visualization, and presentation. The inspiration for this post came as I was scooting over to the OU MicroCT Scanning facility on my Vespa with a scan burrito tucked in my pocket. What, you may ask, is a “scan burrito?” A scan burrito is what we call the assembled packet of dead animal that we shove into our microCT scanner. The dead animal du jour was the fleshed-out head of a hatchling false gharial, Tomistoma schlegelii, (USNM 84247), which is an unusual and enigmatic species of crocodilian that today clings tenuously to life in rivers of Malaysia and Sumatra.
In praise of the scan burrito
The problem with CT scanning animal specimens fixed in formaldehyde and stored in alcohol is that you need to remove them from the fluid for scanning, but you don’t want them to dry out. If you wrap them in, say, gauze soaked in alcohol, then they won’t dry out (which is good), but the scanner’s x-rays will “read” all that soaked gauze as being the same kind of substance as the animal (since animals are mostly made up of watery fluids). That means that your lovely specimen will seem to be inseparably clothed in a nasty gauze coat (which is bad). To get around this problem, we wrap the specimen in multiple layers of thin foam sheeting such as the kind that comes with electronics (in this case, the iPad upon which I write this post). We then, wrap the alcohol-soaked gauze around the foam sheeting, and then tuck the whole thing snugly in a plastic bag which we then roll up neatly. It looks like…well, a burrito. The x-rays cook right though the foam sheeting, and so the sheeting, plus the small amounts of air between the sheets, provides a separator that allows us later to easily digitally extract the animal from the surrounding burrito.
Separating self from self
Sometimes you even need to separate different parts of the dead animal from itself with bits of foam so that they’re more easily discriminated in the CT scan dataset. Again, we need planes of air (or foam, which is basically air) around each part. We’ll then use those air planes (sorry) to later cleanly pull things apart digitally in the computer. For the false gharial, we really needed this extra step, because, years ago, the lower jaw part of the head had been dissected free from the top part of the head. We wanted to scan both so that we could assemble them later digitally, so I wrapped them separately with foam sheeting. In the image above, you can see that I pointed the two parts in opposite directions so they’d fit a little better in the scanner (we can position them as we please once they’re in the computer later). I also put a single scan slice in the image that shows the components of the scan burrito in place.
A wee gator embryo was among our first scan burritos
I’m not sure who came up with the term “scan burrito.” Wasn’t me. I blame Ridgely. The term itself is relatively recent, but scan burritos have been staples of the WitmerLab for years. One of our first was for an alligator embryo that we scanned when we first got the OU microCT scanner. In this case, I wanted the limbs and tail to be as free from the body as possible, rather than inseparably plastered to it. I had to make tiny little tubes of foam for the tiny little gator limbs and wrapped the tiny tail and head in foam, too. It came out well, and Ryan Ridgely did his typically superb job of the 3D visualization. I posted movies on the OUµCT Gallery website years ago, but I just now posted them to the WitmerLab YouTube channel so that I can embed them here. (Wasn’t much science on YouTube back then, mostly just skateboarding bulldogs.) Here they are:
And so goes the tale of the scan burrito…indeed it would have to be the world’s worst-tasting burrito. Although, given the 140-proof alcohol it’s soaked in, the meal might become more palatable as it progressed.
— Larry Witmer