danust2012 at gmail.com
Wed Oct 14 21:24:02 UTC 2020
On Wed, Oct 14, 2020 at 9:02 PM William Flynn Wallace via extropy-chat
<extropy-chat at lists.extropy.org> wrote:
> Whether cats can identify the people who are dying, or whether dogs can
> diagnose the virus (they can, I think I read), we ought to be doing far more of this.
Isn't it already being done?
> At this point - correct me if I am wrong - animal's noses are more sensitive than
> any instrument medical people have.
I don't think it's correct to say animal noses -- with the exception
of a certain moth I mentioned earlier -- are more sensitive than
medical instruments per se. It'd be more that particular instruments
are made to detect certain things and unless they're designed for this
(or the analyzer is looking for it), you simply miss out. Whereas an
animal might react to smelling something different period and you
might not understand what it's detected or how exactly it's done so.
So with the dogs, a machine could probably be designed to outsmell
them, but you'd have to know what you're looking for. (Think of a mass
spectrometer or a gas chromatographer. I'm sure they can both
outperform animals in specific areas, but you have to know what you're
looking for or do the analysis broadly enough to detect all the
instrument could. As data analysis has become ever faster and cheaper,
this gets easier.)
> When was the last time a physician smelled you? Intentionally, that is, for
> diagnostic purposes. It might be that some smelling is done by lab people
> but I dunno. I suspect that the medical community does not want to turn
> over anything to lower animals, being the highly conservative people that
> they are (who hate second opinions as it challenges their dominance and
> ability). Since we already have evidence that an AI can diagnose better
> than any physician, why are we seeing more use of the AIs? Ego, I think.
I'm not going to argue that health professionals aren't conservative,
but the use of dogs to detect cancer seems well established:
I'm more worried about seeing the cat reports as being false positives
and about overplaying the hand here.
> I do agree that using the most sensitive noses, those of bears, might be
> costly, but I think that they do dance better than dogs do. And when was
> the last time you saw a dog on a unicycle?
Again, the best 'nose' in the animal kingdom seems to belong to a
moth. Granted, moths probably can't be trained much, though one might
be able to use some form of conditioning to get them to detect a
diagnostic molecule for COVID-19 -- provided you can cook up a batch
of that stuff. Maybe a little GMO-ing could work here too. But if you
know what to look for and it's easy to program a gas chormonatograph
to look for it, then you can just roll that out rather than spend
years in lab trying to coax moths to do it for you. (Not to mention,
it's probably hard to keep moths around for whenever you need them --
whereas any decent lab is likely to have a chromatograph.)
Sample my Kindle books via:
More information about the extropy-chat