Week 1: June 12 - June 16

Monday marked my first day at the CVFD Lab. Walking into the Wisconsin Institute for Medical Research, I was very excited but also nervous. How could a high school student like me hope to keep up with work which undergraduate and graduate students and PhDs were doing? Would I be able to understand anything they were saying? Would I mess up even the most menial tasks that they would have for me? In the midst of this self-doubt circulating through my mind, my mentor, Rafael Medero, a graduate student in the lab, greeted me and showed me to their lab in the second floor of WIMR. Along the way, he introduced me to the other lab members including the head of the lab, Dr. Alejandro Roldán-Alzate. As I walked into the lab, the self-doubts I had had before quieted as I glanced at their 3D printers and their wide range of objects and setups to simulate various parts of the cardiovascular system. Rafael showed me one printer (a formlabs) which prints objects by curing resin with a laser, far from what I had experienced in any of my science classes. He then showed me a 3D-printed TCPC connected to tubing which he said they had run fluid through to collect data on fluid dynamics. Afterwards, Rafael brought me to their lab in the Mechanical Engineering Building, where he showed me the experiment he was running for a conference he has next week. Through primarily observing this experiment this week in order to get a better feel for the CVFD Lab's research, my excitement increasingly overshadowed my nervousness for working in a lab at the UW this summer.

Particle Image Velocimetry

Rafael's experiment revolved around using particle image velocimetry (PIV) analysis of a fluid being pumped through a carotid bifurcation model. I was able to closely observe the process he went through to set up the experiment and collect data. Before starting, he had to align a laser beneath the point where the model would be and calibrate high speed cameras. The laser made particles seeded in the fluid visible and the high speed cameras took bursts of photos while the fluid (a solution of glycerol and water) flowed through the model, a silicone mold with tubing connected to it. The resulting series of photos were saved in a program that could produce an image of velocity vectors based on the movement of the particles over a period of time.

Casting

This past week, I also had the opportunity to observe and help two undergraduates, Luis and Armando, working under Rafael, undergo a casting process to make more silicone models for PIV analysis. The particular models they were working on were two arteries with a stenosis—one with a 50 percent reduction and one with an 80 percent reduction. They began by 3D printing molds. They then poured a metal with a low melting point into the molds and left them in baths of cool water to harden. Once hardened, they removed the 3D-printed mold, leaving the solid metal shaped in an artery with a stenosis. Eventually, they will attach the metal to a box, pour in silicone, and then melt out the metal once the silicone has cured, leaving the silicone mold; however, each metal piece that has been produced so far has not been viable for this next step as they have either had bubbles or broken.

Calibration Mold

As mentioned previously, before taking data for a PIV, the high speed cameras must first be calibrated. To do this, the high speed cameras must be aligned and properly focused. To simulate the images for his particular experiment during calibration, Rafael has been using a calibration plate with white dots on it provided by the company that made the high speed cameras and sliding it into a silicone mold that was made by a undergraduate student last year with the water-glycerol solution. The reason for using such a solution is not only because it is quite viscous, but also because it is composed in such a way that the index of refraction of the solution is the same as the silicone mold; thus, light will not bend as it passes from the silicone mold to the solution, making the silicone mold transparent and allowing the high speed cameras to view the particles as they pass through. However, due to its accumulation of fingerprints, the calibration mold is becoming less and less transparent, making it harder to calibrate the cameras before the experiment. The lab has actually had this problem with many of its silicone molds, but are still in the process of finding solutions. Until then, Rafael asked me to make a new calibration mold for him to use in the meantime.

To do this, I will undergo the casting process, but will only need to 3D an object to make the mold around as this mold in particular does not necessitate using the low melting point metal mentioned above. I began by designing the mold and then constructing the object to be 3D printed in a program called SolidWorks on Thursday. On Friday, I glued together the box to hold the 3D-printed object where the silicone will be poured in and left it to dry over this weekend.

3D Printing

After I finished constructing the object in SolidWorks, I converted it to an STL (the file format for 3D printing) and opened it in a software to prepare it for printing. Because of the functionality of FormLabs printers (which I used), the object ends up getting printed upside-down. If the object was merely printed as it is, some uncured resin would get stuck on top of the model, which could lead to it being misshapen or breaking. As a result, the general method is the print the model at an angle and attach it to a base with support beams. That way, the model will be supported and uncured resin will drip off of the model as it prints. Nonetheless, this method makes the object take a longer time to print (mine was quite small and took around 10 hours). Once finished, I took the model out of the printer and soaked it in an alcohol solution to remove excess resin and smooth out its surfaces. After the support beams were removed, I sandpapered any rough areas or sides that were longer than I wanted.

Report abuse