[M T]

curious of any of the medical experts on this board have thoughts on this testing approach

https://www.globaledentity.com/identifyi...dividuals/

Not a medical expert, but I'll pipe up.
It may be valid, but it seems wrapped in snake oil.  The usual B.S. trying to convince you how good they are, rather than just proving it.  Where are there experimental reports published?

Take this:

Our testing as of 17 Aug 2020 involved 100,000 individuals, with an accuracy of 99% in identifying sick-symptomatic, infected-asymptomatic, and healthy individuals during our identity screening for UID™️. Any infected-asymptomatic or sick individuals were then verified with a clinical test if they scanned positive for SARS-CoV-2 (Covid-19) in the pre-screening using our NIR system.

Did you notice that they could get that result by testing 100K people that aren't aren't aware of symptoms, detecting 1000 as positive by their tests, and sending them for clinical tests, and declaring everyone else healthy (without testing them)?  Actually, from the way they said it, they could be wrong on all 1000, and still been accurate 99%.   Maybe that's not what they did, but my scenario fits what they say.

It would be nice if this worked to detect COVID and not just having a low PO2 (which could be asthma, allergies, sinus infection, ...).   But it looks like a company that wants to be bought up by a bigger player.

=====

On this topic, I'm curious about this claim out of MIT that their algorithm can detect COVID-19 from recorded (by smart phone) sounds of a forced cough. (published open journal paper).  98.5% sensitivity; 94.2% specificity.  "100% asymptomatic detection rate" 88% accuracy on all subjects; 5732 subjects.   I haven't read the paper so I can't comment on the size of their training and test sets, or anything else.

In the same open journal is another paper claiming to detect COVID-19 from voice from Tel Aviv.  Only 88 subjects.  Not as good in the results.


For all of this (including circulation), are they detecting COVID-19, or just the body's reaction to any illness?  I would suspect the latter.   Sorta like a different kind of measure than temperature, BP, PO2, ...

[M T]

curious of any of the medical experts on this board have thoughts on this testing approach

https://www.globaledentity.com/identifyi...dividuals/

Not a medical expert, but I'll pipe up.
It may be valid, but it seems wrapped in snake oil.  The usual B.S. trying to convince you how good they are, rather than just proving it.  Where are there experimental reports published?

Take this:

Our testing as of 17 Aug 2020 involved 100,000 individuals, with an accuracy of 99% in identifying sick-symptomatic, infected-asymptomatic, and healthy individuals during our identity screening for UID™️. Any infected-asymptomatic or sick individuals were then verified with a clinical test if they scanned positive for SARS-CoV-2 (Covid-19) in the pre-screening using our NIR system.

Did you notice that they could get that result by testing 100K people that aren't aren't aware of symptoms, detecting 1000 as positive by their tests, and sending them for clinical tests, and declaring everyone else healthy (without testing them)?  Actually, from the way they said it, they could be wrong on all 1000, and still been accurate 99%.   Maybe that's not what they did, but my scenario fits what they say.

It would be nice if this worked to detect COVID and not just having a low PO2 (which could be asthma, allergies, sinus infection, ...).   But it looks like a company that wants to be bought up by a bigger player.

=====

On this topic, I'm curious about this claim out of MIT that their algorithm can detect COVID-19 from recorded (by smart phone) sounds of a forced cough. (published open journal paper).  98.5% sensitivity; 94.2% specificity.  "100% asymptomatic detection rate" 88% accuracy on all subjects; 5732 subjects.   I haven't read the paper so I can't comment on the size of their training and test sets, or anything else.

In the same open journal is another paper claiming to detect COVID-19 from voice from Tel Aviv.  Only 88 subjects.  Not as good in the results.


For all of this (including circulation), are they detecting COVID-19, or just the body's reaction to any illness?  I would suspect the latter.   Sorta like a different kind of measure than temperature, BP, PO2, ...