Color me skeptical. Also color me concerned about exacerbating overdiagnosis and overtreatment.https://venturebeat.com/2019/05/07/mit-csails-ai-can-predict-the-onset-of-breast-cancer-5-years-in-advance/ …
-
-
Replying to @gorskon
We'll have to wait and see. But yes, the potential for substantially more biopsies is real. And therefore more false positives (and equivocals) leading to unnecessary surgeries.
1 reply 0 retweets 0 likes -
This Tweet is unavailable.
-
Replying to @redpillgreen @ether_doc
Overdiagnosis is not a "false positive."
2 replies 1 retweet 3 likes -
This Tweet is unavailable.
-
Replying to @redpillgreen @ether_doc
In the case of overdiagnosis, that definitely remains to be seen. Finding lesions earlier tends to exacerbate, not improve, overdiagnosis; which could very well more than negate any improvements in accuracy/prediction. Also, the outcome data generally available is just too crude.
1 reply 1 retweet 2 likes -
This Tweet is unavailable.
That AI could do better is a HUGE assumption with respect to mammography given the limitations of the technique, which is strictly anatomical (and then in a very limited sense) and reveals nothing about biology. Biology, not anatomy, will be key to reducing overdiagnosis.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.