[LINK] Robots in health care could lead to a doctorless, hospital

[Chris Johnson]

If we were able to choose between a robotic medical care system with 95%
correctness (based on its past history) and a human system with 90%
(likewise), what rational person would not choose the robotic?
Anjali Jaiprakash, Jonathan Roberts, Ross Crawford in The Conversation
wrote:
> Most rational people would opt for the course of action that is more
> likely to save their child.

In our use of financial systems we are always being told "past
performance is not a predictor of future performance". That caveat
aside, what is the basis of humans making choices between health
systems? We are none of us purely rational, particularly when it comes
to our own health and the health of others we care for. We need to
trust. Unless we develop trust in the system, we will not use it for
matters that have deeply attached emotional roots in us. A system that
gives the right answer 95% of the time, but gives a crazy answer 1 time
in 20, without providing any means of asking it for an explanation-
cannot be trusted. We have more trust in a lower performing human-based
system, because it at least has the possibility of providing us a
rationale, and can be held to a post hoc examination of its reasoning -
because there is an ultimate chain of responsibility back through other
scrutinising people in the medical system, and back to political
oversight and the courts, if necessary. The threat of legal action does
not govern all decisions, but it does provide a control system that
effects quality control. Only if the quality of the whole system can be
improved should we incorporate mechanical aids or changes to procedures.
(A case in contention at present is the argument about universal
screening for prostate cancer: appears rational, turns out to do more
harm than not screening, within our current system).
A decision making system that does not provide a rationale - is not
designed to be scrutable, in Judy Kay's terms - is less useful and less
trustworthy than one with less accuracy but which provides explanations
that the user or user's expert supporters can understand, and can use to
make meta-judgements to go with the decision or not.
As Janet Hawtin says, using decision making systems in support of
experts both enhances the value of the systems, increases the leverage
of the combination of mechanised knowledge and human experience and
judgement, and provides psychological satisfaction for the patient,
their loved ones, and the people who work in medicine. (now let's extend
this idea to providing education...)

Janet Hawtin wrote:

> What happens if we relook at those professions with the intention to enable
> the kind of augmented capacity to be effective in their work that
> technology can offer, so that future generations will inherit tools which
> help them to be challenged and wise and have respected expertise.
> 
> I understand that our current economic model encourages us to break down
> the role of humanity until it is cheap or redundant, I think we need a
> different economic model which is designed for a healthy thriving ecology
> and a sentient and fully engaged community where everyone is able to
> contribute to their best capacity.
> 
> It is possible that technologies can learn. It is more interesting to find
> out how people with face to face expertise can use technology which learns
> along with them rather than finding yet another way to make humanity
> obsolete in the interests of industrialising for individual financial
> advantage. We do not need replication we need to understand why we are
> still using logic/value which is breaking the ecology, how it works and how
> to change ourselves to fit?