News for hackers. Stuff that matters.

A comment on Toshiba device tests cancer types from a drop of blood. Parent comment

Re:Toshiba device tests cancer types from a drop of blood by jolmg on Monday November 25, 2019 @05:59PM (#21633788)
Does that translate to 5-10% chance of false-negative? Can't that be improved simply by having multiple tests with a few more drops of blood?
[ Parent | on HN ]
Re:Toshiba device tests cancer types from a drop of blood by jobigoud on Monday November 25, 2019 @10:56PM (#21635541)
90% sensitivity = 10% false negatives. When the patient has the condition, it is detected 90/100 of the time.

95% specificity = 5% false positives. When the patient doesn't have the condition, it is correctly not detected 95/100 of the time.

[ Parent | on HN ]