News for hackers. Stuff that matters.
Re:AI can detect coronavirus from CT scans in twenty seconds by chrisa on Friday March 13, 2020 @08:06PM (#22571948)
As the first comment there points out, a 96% accuracy doesn't mean much without knowing the false positive and false negative rates.

I have no doubt that this is good work in the right direction - but the headline alone doesn't make it a home run.

Re:AI can detect coronavirus from CT scans in twenty seconds by mike_n on Friday March 13, 2020 @08:34PM (#22572057)
Do human doctors have a better rate? And at what cost/throughput?
Re:AI can detect coronavirus from CT scans in twenty seconds by jupp0r on Friday March 13, 2020 @08:38PM (#22572078)
This is the base rate fallacy.

    return “healthy”;
would have a 99.9% accuracy right now.
Re:AI can detect coronavirus from CT scans in twenty seconds by yellow_lead on Friday March 13, 2020 @09:16PM (#22572332)
I think you have a good idea for an AI startup here.
Re:AI can detect coronavirus from CT scans in twenty seconds by smacktoward on Friday March 13, 2020 @09:44PM (#22572472)
For bonus points, have it return one out of every 1000 or so requests with "not hot dog."
Re:AI can detect coronavirus from CT scans in twenty seconds by droidist2 on Friday March 13, 2020 @08:45PM (#22572109)
Ahh, you need to use ROC AUC or something.
Re:AI can detect coronavirus from CT scans in twenty seconds by contravariant on Friday March 13, 2020 @08:59PM (#22572206)
ROC AUC is somewhat meaningless when basically any amount of false positives is going to massively decrease the utility of the test.
Re:AI can detect coronavirus from CT scans in twenty seconds by vikramkr on Friday March 13, 2020 @08:20PM (#22571997)
I suppose in a case where like half the population had the virus those might factor less. And if the errors skew towards false positives, it could be a quick prescreen to get an enriched pool of candidates for manual checking. Definitely not a panacea yet though