Share
clock Created with Sketch.
22/06/17
13:37
Share
Originally posted by Martin Gifford
↑
"And it still beats the current diagnosis of using X-ray......."
I'm not sure about that. At the start of the ann, it says RAP results were compared to clinical diagnosis based on many tests, including x-ray:
"high level of agreement between ResAppDx and the clinical diagnosis (based on clinical presentation, auscultation, imaging and laboratory tests )."
So 70 / 81% (cough / cough plus symptoms) for pneumonia means that RAP is 19-30% worse than the gold standard, which seems to include x-ray (imaging). But we have this later on in the ann, we have this working definition of pneumonia:
"4. Pneumonia is clinically diagnosed viral, bacterial or atypical pneumonia which may or may not have be confirmed by chest x-ray."
That, to me, totally confuses the issue. It would be better if they gave two clear sets of results:
1) Resapp vs doctor without further tests,
2) Resapp vs thorough clinical testing (gold standard).
Expand
That article I posted specifies x-ray is around 50% (which from memory was mention previously here). But if they use the process of elimination by detecting the other "high accurate" diagnosis first, it will be 90%+ accurate would it?