Week 9: Hurdles!
Hi, guys!Â
Welcome to week 9 🙂 For the past weeks, many things have gone smoothly! I collected many data, was able to perform trials, and had lots of fun doing my project; however, there were also a lot of instances when things went downhill. In this blog, I will explain all the negative aspects I encountered during my project so that I can improve upon these factors for future projects + update the “problems” section of my research project.Â
- A little bit of vaguenessÂ
One of the three questions I wanted to answer through my senior project was, “Are these devices worth it in terms of cost and time?” And while I do have an answer, I obtained it by researching scholarly articles and learning about the software with the research team; I wasn’t able to specifically perform trials where I directly compared their speed, etc. I did retain some useful information, but I wish I could have had some data to prove it!
Here’s what I found out: So, as of right now, no. We shouldn’t begin implementing these in hospitals…not yet, anyway. For Mediapipe, especially, it requires us to utilize coding and a lot of data analyzing before we even retrieve data- right now, we’re trying to find the quickest and most accurate devices, and waiting for data doesn’t seem like the smartest idea during urgent situations. The mmwave radar is a lot better with this since it shows us data right away. But the point of using this is to assess triage efficiently; as of now, the radar can only find vitals accurately for only 1 person at a time. This doesn’t seem very useful in waiting rooms filled with patients. Still, both of the devices are accurate, I’d say! Just not as efficient as we want them to be.
2. It’s just a theory
Okay, so as much as I want to be some wizard who can improve hospital policy, it’s not that easy. Another question I’ve been trying to answer was, “Can kinesthetics become the seventh vital sign?”. And right now, I’d probably say it COULD be and it SHOULD be, according to some of the data. But this is just me proposing a theory- kinesthetics isn’t going to just suddenly become a vital sign! I wish I could have collected more evidence and worked deeper with the radar/mediapipe in the long run to really answer this question, but as of now, this is what I know:
Kinesthetics is a great way to evaluate patient assessments. Unlike the other six vital signs, this one measures and tracks peoples’ way of moving, which can be extremely useful for physical therapy patients + any conditions tied closely to orthopedics. Looking at my data for Mediapipe, I definitely think doing a lot of revamping would be helpful, especially since my data for walking (one of the most common movements for gait analysis) was not that accurate. I have hope, though, after seeing the data for limping; those were fairly accurate! These discrepancies are due to Mediapipe’s inability to locate checkpoints specifically. I mentioned this before, but to reiterate, we can choose which points to place the markings (e.g. the ankles, knee, hip, etc). But we can’t place the markers specifically in that area, so Mediapipe could place the marker at the side of the knee rather than the front of it, which heavily altered my data for angles.
Those were the biggest obstacles I faced! There were fewer experiment-based ones, such as not having enough room, facing technical difficulties, etc, but those (for the most part) were able to be resolved at the end! Right now, my research team and I are struggling to get the new radar working; I hope to test the multiple-people detection aspect in the next few weeks, but I am not sure if it’s possible. The device keeps on failing to turn on, which may be another reason why it cannot be part of triage assessments yet. Hopefully things start working out soon!
As always, let me know if you guys have any questions!
– Linda Lee
Comments:
All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.