Yeah, I’m still figuring it out, but this. An ICU is the most complex place in a hospital and probably the *hardest* for improvised emergency clinics to replicate. To reduce load on hospitals, maybe we *first* scale up the *other* stuff.https://twitter.com/gallabytes/status/1238744523548667905 …
-
Show this thread
-
Replying to @s_r_constantin
if we know we're going to need ICU capacity for one very specific kind of intensive-care situation (the COVID19 respiratory one) I wonder how much faster we could scale up a sort of one-trick-pony ICU capacity to deal with just that. answer maybe "not much" or "a lot"
2 replies 0 retweets 1 like -
Replying to @EconElan @s_r_constantin
Maybe one place where ML could help (but would never get regulatory approval in time) is predictive analytics on which patients are likely to crash soon (and rough time frame) along with a queueing algorithm of sorts to help minimize situations with urgent tasks>ppl to do them.
1 reply 0 retweets 1 like -
Replying to @EconElan
Maybe, but I’m skeptical. I think there are already known, human-visible risk factors for case severity. Iirc leukopenia is a big one.
1 reply 0 retweets 1 like
Sorry, my mistake, I meant lymphopenia. https://www.medrxiv.org/content/medrxiv/early/2020/03/03/2020.03.01.20029074.full.pdf …
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.