3 ms·
Sensitivity/Specificity of 0.65/0.76 are laughably horrible for a diagnostic. I note they didn't give a ROC curve or d'.
by hprotagonist 9y ago
Sensitivity/Specificity of 0.65/0.76 are laughably horrible for a diagnostic.
I note they didn't give a ROC curve or d'.
- killjoywashere 9y agoYeah, considering one presumes this is a screening test, the sensitivity is a catastrophe: a negative result means you have no idea if you have cancer or not. You spent money that could have been better spent on just about anything else (composting the dollars, for instance).
- ska 9y agoThere is a reason mammography is the standard for screening, not ultrasound. Honestly the more interesting case is probably this + mammo, as there is research suggesting sensitivity increases with US + mammo, over mammo alone.
- hprotagonist 9y agoI don't see why a screen you cannot trust should be a precursor to a mammogram. It just makes things worse.
- ska 9y agoWho said anything about a precursor?
- hprotagonist 9y ago"untrustworthy adjunct", then.
- ska 9y agoWell, like I said there is some research suggesting classsification on mammo + US can perform better than mammo alone. Hardly an unusual result in ML work. So it's worth looking at.
- hprotagonist 9y agoah, i think i composed your meaning incorrectly. If training their NN on mammogram data and ultrasound data gets them a better ROC, that's good. What they have now is worse than useless, though.