193
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 19 Jan 2024
193 points (98.0% liked)
Asklemmy
43992 readers
624 users here now
A loosely moderated place to ask open-ended questions
If your post meets the following criteria, it's welcome here!
- Open-ended question
- Not offensive: at this point, we do not have the bandwidth to moderate overtly political discussions. Assume best intent and be excellent to each other.
- Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
- Not ad nauseam inducing: please make sure it is a question that would be new to most members
- An actual topic of discussion
Looking for support?
Looking for a community?
- Lemmyverse: community search
- sub.rehab: maps old subreddits to fediverse options, marks official as such
- !lemmy411@lemmy.ca: a community for finding communities
~Icon~ ~by~ ~@Double_A@discuss.tchncs.de~
founded 5 years ago
MODERATORS
At least I can rest assured of the fact that AI will be next to useless in my intended field. Emergency medicine is an environment where you get a random constellation of symptoms and complaints with very little direction on which are related to the current illness, and which ones are not currently relevant. Also, in the time it would take to get all the info into the AI for a trauma/cardiac/code situation, the patient might be dead or rapidly heading in that direction.
Can’t AI aggregate the data on triage outcomes to prioritize larger scale emergencies… it has to be useful somehow I would think
It wouldn't need to be AI, just sone statistics and "Crdiac arrest? Piority 9. Broken arm? piority 1" decision-making.
I'm a tech wizard not a healer, there probably are factors that define one cardiac arrest even more critical than the other i just do not know them.
There's some things you look for that are difficult to describe to someone who hasn't seen it before. That's part of why experience is so valuable in Emergency Medicine, and it's not uncommon to put your best nurses out in triage. People will do this kinda twitchy/wilting/loss of focus/change in pallor/change in posture right before they go down. I don't have a good way to describe it, and it might be easier to draw even, because it really is a body language thing and the general appearance of the patient that can inform your decision making.
I have thought about trying to plan out a learning algorithm that could spit out suggestions for triage level and preliminary tests based on input data like vital signs, symptoms, and complaints... but I would never implement something like that as anything beyond a tool for the nurses at triage to use. There would have to always be an option to override the algorithm because there's some aspects of patient presentation that are not easily quantifiable. I'd never be able to explain it in a way that one could input it into a computer, but even with my limited experience, I can tell which patients are going to crump on me.