Regression and Rerouting
We are now in the ninth week of the internship, and as the title suggests, we had to go back and redo some of the projects we completed to ensure that they met the expectations of possible investors. We still had work to do, but by this point, it was anticipated that we would have a fully functional prototype that was suitable for public displays. The week began on a busy note thanks to this and presentations on Tuesday that were expected to have a good turnout of various persons.
On Monday, we began with a group meeting to discuss the tasks we needed to complete and how to assign them. We faced some challenges on Monday because we hadn't finished the IOT portion of our projects and the laser cutter wasn't working, so we couldn't create a laser-cut enclosure for them. This was a significant setback because the most of our projects were already functional and we were only left with the device case to finish.
The situation with the casings was the same on Tuesday, but the neonatal monitor mobile app was ready to be connected to our device, but the arduino we were using had some connectivity issues with our firebase backend. Therefore, I began experimenting with utilizing an esp32, which works great with firebase, however the temperature readings were wrong. Chikumbusto Walani, one of our teaching assistants, and I spent a lot of time researching this problem before determining that the esp32 had an analog to digital conversion (ADC) manufacturing error.
On Wednesday, I tried to 3D print a casing for both the neonatal and the vein finder but still had the same issues as before with the 3D printers. We tried in several attempts but the filament in the printers often got stuck which required that we had to restart the whole process of 3D printing. This went on for the neonatal monitor which resulted into us some where around 12 midnight with no success while for the vein finder the casing finished off the following day.
We finished fabricating the vein finder's casing on Thursday. I attempted to apply some patches for the esp32's ADC issue that I found online while the casing was being finished. At first, the changes appeared to be successful as the readings began to approach the desired outcomes, but by the end of the day, the fixes had failed since they still had a significant margin of error, which is not ideal for medical technologies. The neonatal monitor's 3D case needed to be sent to a MUST design studio technician so they could try to print it for us at around 3 o'clock that day. However, this did not work out well because their 3D printers also experienced issues, so we gave up.
We worked with very little motivation on Friday due to the failure of our prior attempts to create our gadgets, and as a result, we only presented the laparoscopic training kit, leaving the other devices unpowered or incomplete. Will Moyo, a few MUST interns, our teaching assistants, and the studio's technicians were present for the presentation.
Week nine was by far the most frustrating of all the weeks, but we were applauded for having functional systems, so we closed the week with some desire to get things done, and I was fired up to make sure that the upcoming presentations were superior to what we delivered this week.