Second scorpion MRI

On Friday (August 3) I was contacted by Rohini Vidya Shankar (a PhD student who works at the MR at Saint Josephs hospital). She asked if I could bring in another scorpion to the lab to image. I could, and I brought the scorpion in as requested. It went the same as previously (see MR imaging) except this time we did not have anyone who knew how to run the anesthetics so I had to kill the scorpion while it was awake. We also used a slightly different technique when putting the scorpion in the 50mL vial. Originally we simply inserted the scorpion, filled the vial with the perfluorinated fluid, and screwed on the cap, but this always resulted in a small bubble (which can create an artifact by distorting the magnetic signal). This time however, instead of using only the vial cap, we stretched several layers of parafilm (like super plastic wrap) over the top of the vial and pushed the bubbles out. We then screwed the cap over the parafilm, eliminating the bubble and creating a tightly sealed container.

Here are some pictures of the lab:

Video shoot #2

Our second attempt at capturing high speed video of the stinging motions went far better than our first attempt. We managed to capture relatively high consistency in the stinging motions in only seven ten-second runs with few hiccups. We did encounter one problem, but also fixed a former one.

First, the new problem. There was what looked like a piece of dust on the lens, but was discovered to actually be on the CCD. This was fixed by cleaning the CCD with compressed air.

Second, the fixed former problem. Originally, we put the scorpion on ice to make them docile, but this resulted in a long warm-up period. This time we managed to secure the scorpion without cooling it first. This was slightly more difficult than before, but eliminated the warm-up period, thus allowing us to start shooting earlier.

I uploaded the best of the videos to YouTube:

Here are some pictures:

Mirror box v2 test

After having redesigned the trigger mechanism in the mirror box (see Revised trigger mechanism) John and I tested it on one of the two remaining scorpions.

It performed better than expected, getting us a sting 100% of the time. Because of my confidence in the new trigger, I have scheduled another video shoot for this Monday (July 30). Here are some pictures of the test:


Revised trigger mechanism

After the almost complete failure of the original trigger mechanism, some revisions were in order (see Video shoot #1 for the issues).

As mentioned in Revised trigger mechanism design and scorpion habitat, the scorpions responded violently to being brushed or poked with a paintbrush (I assume because of the large number of contact points from the bristles). Therefore I decided to scratch the foam and pushrod design and switch to a paintbrush on the end of a swingarm. Here are the designs and pictures:

Because the electronics were unchanged, all I had to change in the code was the behavior of the servo in the Arduino IDE, and the commands in the Processing IDE. Here are is all the new code:

Version 1.8 (click to view, right click and “Save Link As..” to download) of my Arduino software takes commands via serial as usual, but instead of only being able to write the received angle to the servo, it can now receive “activate” and “retract”, which cause the servo to vibrate and stop respectively.

Version 1.8 (click to view, right click and “Save Link As..” to download) of my Processing code still allows for angles to be assigned to various buttons, but now also has two new buttons: Activate and Retract. Here’s a screenshot of the GUI (some buttons are currently unused):

MRI imaging

Today John and I took the scorpions to St. Joseph’s hospital for the MR (magnetic resonance) imaging.

After considering various ways to kill the scorpion we had selected to image (subject cannot move during imaging), I decided to have the scorpion anesthetized so that it would be easier to handle, then made an incision across its back, right behind the eyes to sever the spinal nerve. We then proceeded to place the dead scorpion in a 50mL vial filled with water and insert it into the MR machine.

Because of the tendency of the scorpion tail to bend, we had some difficulties getting a clear image. We decided that, as we were primarily concerned with getting images of the structure of the tail, to remove it an image it alone (because the tail is not very wide we could put it in a much smaller vial, thus keeping it straight). This improved our images but lead us to discover that we were not getting a strong enough signal from the internal structures of the tail. This was due to the fact that an MR machine sees hydrogen atom nuclei (protons), and as our subject was suspended in water, we were getting a large signal from the water surrounding the tail, thus masking the weaker signal originating from the tail itself (akin to taking a picture of something with the sun behind it). To fix this we switched the water for a perfluorinated fluid (a hydrocarbon-like molecule where all hydrogen atoms have been replaced with fluorine) atoms that does not have a magnetic signal.

After all of this we were finally able to get a clear image. The high resolution scanning was started with an estimated time until completion of six and a half hours.