Re:DeepMind's Mustafa Suleyman says general AI is still a lo...
by dpandey on Monday December 05, 2016 @10:47AM
(#13106132)
Your last thought is key. Medical diagnosis demands extremely high precision as well as 'judgement' because one little deviation can be highly correlated with something serious. And there can be several of these deviations in different places. Sometimes a doctor notices something is off while looking at something else. It's not easy to model out all these while keeping up with the changing knowledge in the field and still avoiding false negatives.
Re:DeepMind's Mustafa Suleyman says general AI is still a lo...
by ska on Monday December 05, 2016 @11:47AM
(#13106726)
Agreed - I think the potential value is quite high in unusual root causes. It's going to be very hard to make a system better at diagnosing 'flu' than a trained clinician. However, "flu-like-symptoms" plus a couple of subtle things in the history may point at a relatively obscure syndrome or whatever, which can be very difficult for a clinician to find.
Another thing to consider in any work like this, you really need to thank about what your false positives mean, in a practical sense. If your numbers are big enough (e.g. screening) and your positives result in a procedure (e.g. biopsy), eventually someone is going to die based on a false positive. This may still be the better option, of course, but you really need to demonstrate that, and think about what your sensitivity and specificity tuning needs.
There are a lot of barriers to presenting it as you imagine, though: some liability, some workflow, and some just the deep lack of appropriate datasets. There is also some adoption issues - clinicians are rarely going to be happy with a black box approach but it can be difficult to explain/expose the details of such analysis in a way that clinicians will accept.
A last thought: this is actually very hard to do well and dangerous to make mistakes with - so your rather cavalier claim that there would be immediate and wide spread improvement of health care should be tempered a bit.