This is the second part of the lecture on artificial intelligence in medicine,
in which we're going to focus on the scaling of
applications to cover a large part of medicine and its deployment.
So, now having seen that we have the capability to model
many different aspects of human expertise and knowledge in a practical way,
it can be used to support clinical practice.
Now, we obviously want to work to evolve,
why can't we do this for all of medicine?
This is the issue of scalability that we referred to earlier.
Many people in the AI medical fields have
aspired to deliver services across a wide range of medicine.
A famous quote by Robert Greenes,
an American researcher in the field,
in his book, Clinical Decision Support, The Road Ahead,
painted a picture of a future in which a large proportion of
medical knowledge and practice was captured in
a computerized form and stored in a repository-
a bit like the original Oxford system of medicine dream.
But, he adds, ''if we could do that with tools for
delivering patient-specific advice at the time of need,
point-of-care, imagine what implications that would have on our ability to
deliver high-quality decision making throughout medicine. ''
That has been a vision that has driven many people in the field.
But, we have to remember thats medicine is probably one of the most complex,
if not the most complex,
and large-scale professional fields of any.
If we just look at the range of skills that a practitioner has,
it's simple things like being able to answer questions,
to recognize when there's a dangerous situation,
to reason and interpret about data.
It is also to be able to solve problems,
including some problems that perhaps haven't been seen in quite this form before;
to make decisions dealing with uncertainty,
to keep an eye on patients
to make sure that they are progressing
and detecting when something might have gone wrong.
To make sure things are done in the right order,
to plan and care in an efficient way,
and to ensure that the plans are carried out.
While at the same time on top of all of these things,
we have to keep in good communication with our colleagues and
coordinate all the different services that may be provided.
Remember, going back to breast cancer,
the people involved in looking after somebody with most cancers will be surgeons,
pathologists, radiologists, medical oncologists, clinical oncologists, and nurses.
This is a huge range of
skills that are needed for them to do their work in a coordinated way.