News for hackers. Stuff that matters.

A comment on Half of women will get a false positive 3D mammogram, study finds. Parent comment

Re:Half of women will get a false positive 3D mammogram, stu... by thaumasiotes on Friday March 25, 2022 @01:25PM (#30804894)
But tests are not generally described in terms of accuracy. They're described in terms of sensitivity and specificity, because those quantities are both "accuracy" but they're not closely related to each other.

A test that is always negative has 0% sensitivity and 100% specificity. Nobody thinks that sounds good.

[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by cortesoft on Friday March 25, 2022 @02:35PM (#30805621)
That is how tests are described in scientific circles, but not with the general public. An overall `accuracy` number is usually given.
[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by amelius on Friday March 25, 2022 @07:08PM (#30807862)
Perhaps multiply sensitivity by specificity then, to get a single number that the public can use?
[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by thaumasiotes on Friday March 25, 2022 @08:07PM (#30808149)
That's not a number the public can use. What would it be supposed to mean?
[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by amelius on Friday March 25, 2022 @10:23PM (#30808754)
A fuzzy AND of sensitivity and specificity. Another option is to take the minimum (part of the Zadeh operators).

https://en.wikipedia.org/wiki/Fuzzy_logic

[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by cortesoft on Saturday March 26, 2022 @12:43AM (#30809290)
The point isn’t that you can’t create a single number that can represent the effectiveness, just that you can’t get a single number that the public will understand. People have enough trouble with simple percentages.
[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by lordnacho on Friday March 25, 2022 @04:41PM (#30806838)
You can combine them in an F1 score. Seems like a reasonable way to get a single number, though there might be some drawbacks I haven't thought of.
[ Parent | on HN ]
Re:Half of women will get a false positive 3D mammogram, stu... by thaumasiotes on Friday March 25, 2022 @08:09PM (#30808158)
It looks like an F1 score is calculated from precision and recall. Sensitivity and recall are the same thing, but precision and specificity are different things.

Precision is not a good way to describe a test because, as I mentioned elsewhere, it relies on information about the population being tested.

[ Parent | on HN ]