Automated Diagnosis and
the Future of Biomedicine
Have you ever wondered what is going on in your doctor’s mind during a consultation? A large part of it is likely to be a process of observation, data gathering and analysis based – more or less accurately – on what you say. I emphasise the ‘more or less’ because many patients are surprised to see how often what they say to doctors is ignored or just plain misrepresented in their medical reports. And if, as Einstein noted in relation to physics, “Theory determines what is observed”, then in the case of the ‘observations’ and ‘data’ gathered by the doctor in listening to you the same applies. Indeed it is often the case, paraphrasing Einstein, that in the case of biomedical practice ‘Theory’ determines not just what is ‘observed’ but even what is heard by doctors. The biomedical practitioner’s chief aim is, like a radio, to distinguish ‘signals’ from ‘noise’ – ‘noise’ being anything the doctor regards as scientifically irrelevant to exercising his diagnostic skills – such as the larger life story behind a patient’s illness, their emotional experience of illness – and of its effect on their life.
Yet what if what doctors are chiefly
doing while listening to you
is merely functioning as a human ‘black box’ between what they
regard as significant ‘input’ and giving what they regard as
all-important ‘output’, i.e. probable diagnoses, suggested
treatments, prescriptions, referrals for further tests or to
specialists? As this ‘black box’, what is really going on in your
doctor’s mind is comparable to them going through a type of
predetermined computer ‘algorithm’ or ‘decision tree’ in
order to arrive at the most likely or statistically probable
‘diagnoses’ from the ‘observations’ and ‘data’ he or she
is collecting from you. In which case, however, the doctor is
effectively functioning as nothing more than a sort of human
computer – and therefore
could just as well be ultimately replaced by
a computer – one fitted out with what is called ‘automated
diagnostic software’. If you think this is a wild, futuristic idea
– then think again. For ‘Computer-Assisted Diagnosis’ is
already a reality – and is even used to get a ‘second opinion’
by the most highly revered of biomedical diagnosticians.
“At last he spoke. “Lymphoma with
secondary hemophagocytic syndrome,” he said. The crowd erupted in
applause. Professionals in every field revere their superstars, and
in medicine the best diagnosticians are held in particularly high
esteem. Dr. Gurpreet Dhaliwal, 39, a self-effacing associate
professor of clinical medicine at the University of California, San
Francisco, is considered one of the most skilful clinical
diagnosticians in practice today… Since medical school, he has been
an insatiable reader of case reports in medical journals, and case
conferences from other hospitals.
To observe him at work is like
watching Steven Spielberg tackle a script or Rory McIlroy a golf
course. He was given new information bit by bit – lab, imaging and
biopsy results. Over the course of the session, he drew on an
encyclopedic familiarity with thousands of syndromes. He deftly
dismissed red herrings while picking up on clues that others might
ignore, gradually homing in on the accurate diagnosis.”
Nevertheless the same New York Times
report asked: “Just how special is Dr Dhaliwal’s talent? More to
the point, what can he do that a computer cannot?”. It also
acknowledged that even the famous Dr Dhaliwal “…occasionally uses
a diagnostic checklist program called Isabel, just to make certain he
hasn’t forgotten something.” It adds that the program has yet to
offer a diagnosis that Dr Dhaliwal missed” – without saying
whether this program could actually have done the same job as Dr
Dhaliwal, and that in even less than the 45-minutes he is given at
medical conferences to display his showman-like skills in diagnosing
“vexingly difficult cases”.
Yet aside from a small number of
diagnostic ‘geniuses’ and ‘gurus’ of the sort represented by
Dr Dhaliwal (as well as in the fictional American TV series ‘Dr
House’ – a doctor with similarly encyclopaedic knowledge of
biomedicine and insatiable interest in the latest medical research),
today there is not a doctor in the world, whether generalist or
specialist, with a mental ‘storage’ space of enough encyclopaedic
capacity or ‘Gigabyte’ equivalents to remember even what he or
she learnt in their medical training – let alone find the
motivation or time to keep up with the latest biomedical research. So
we end up with a system whereby, reaching the limits of his
or her knowledge, your doctor
will refer you to a specialist of one sort or another – even though
the knowledge of specialists too, is not encyclopaedic – and is no
less innately limited by the interest in and time they have in
keeping up with the latest research and developments in their field.
Hence also the often diametrically opposed opinions that find
expression among researchers and/or clinical practitioners in
the same specialty – one
hardly compatible with the smug complacency of doctors or specialists
each of whom generally claim their own personal viewpoints, diagnoses
or clinical judgements to be the most accurate or ‘objective’.
The sad fact that comes to light here is
that it is precisely because
medical practice and the functioning of the medical mind takes data
accumulated in the form of impersonal biomedical ‘science’,
‘research ’, ‘evidence’ and ‘observations’ as its
diagnostic foundation, there is nothing that sophisticated diagnostic
software with ever-larger databases of biomedical knowledge could not
in fact do better
than any human doctor – generalist or specialist. Given also that
biomedicine effectively treats the human body as no more than a
biological machine, much like a car, that may be in need of repair,
it comes as no surprise to read the following arguments for a new
‘Automated Medical Diagnosis System’ – the aim of which is to
eliminate “human bias” of any sort – even in using current
forms of ‘Computer Assisted Diagnosis’:
“Cars can be plugged in at the
mechanics for electronic diagnosis, customer issues logged in
enterprise support systems receive immediate potential solutions to
their issue prior to a customer service representative looking at it,
and computers send error reports when an application crashes. In
industries across the world automated diagnostics becomes more and
more prevalent leveraging continually advancing algorithms that
become increasingly intelligent in identifying solutions to known
problems. Yet in the health care industry Doctors have outdated and
limited access to potential solutions… Enter
symptom, disease type, test name or code
requests one physician diagnosis database. As with any human search
that begins with keywords chosen by the user, bias inherently
influences the results. If a Doctor has an assumed diagnosis, they
will immediately begin searching for further evidence that their
assumption can be validated. And if it isn’t, then they will have
missed other potential diagnoses. Additionally, if the Doctor begins
searching by symptoms, while these may be accurate, the order or
weight given to any one symptom will give a bias toward related
diagnosis when in fact, there may be a symptom not given any credit
and thus not included in the search. Regardless of whether you
consider today’s databases or the older process of researching in
books, the results are always influenced by the bias of the
researchers’ initial assumptions. What is needed instead is an
approach that minimizes human bias and considers all relevant and
irrelevant data in determining a diagnosis. Computer software does
this well.”
“With an automated medical diagnosis
system, Doctors could be presented with multiple potential diagnoses
based on all of the patient’s current and past details. Such a
system could be designed for automated medical diagnosis that is
based on probability, utility and decision theory.”
“Essentially, the computer software
could be fed human observations of symptoms, test results, and any
machine data collected such as blood pressure, heart rate, oxygen
levels, etc. The software would then compare these observations with
a database of potential diseases and external agents (e.g. viruses,
bacteria) to determine the most probable diagnosis. These results
would then be presented back to the doctor along with a probability
rating indicating which ones are likely most relevant or accurate.
Each diagnosis could also then be presented with additional direction
to the doctor to further explore for additional symptoms and/or order
an additional test. These additional observations and/or test results
would again then be fed into the system where it could re-evaluate
the probable diagnoses cancelling out some while raising the
probability of others. In addition to immediate interactions with the
software, Intensive Care Unit’s machine observation data (e.g.,
oxygen levels, heart monitors) could be constantly fed into the
system to allow the software to be looking for patterns that match
other known diagnosis that would never be able to be caught by a
human as it would take too much time to evaluate the data.”
What we are presented with here is indeed
a futuristic image – albeit a dystopic one. Already hospitals and
large clinical surgeries with multiple doctors have become, as Dr
David Zigmond notes, more like airports than the so-called
‘old-fashioned’ surgeries run by family doctors – who knew
their patients and their lives intimately. In contrast, in our
clinical ‘airports’ you check in, go through a gate to see to see
a doctor and check out with a prescription or referral for a further
test of some sort or a consultation with a specialist.
For decades now, employment in
manufacturing industries such as car-making has fallen through the
introduction of robots that are pre-programmed to do the job required
of them more precisely than any human worker could. Now however, we
are confronted with the prospect of human beings, all seen as
anonymous body-machines to be processed in automated high-tech
hospital ‘repair stations’, and, as in some futuristic science
fiction movie – conveyed on a factory-like medical conveyor belt.
Stop one on the belt: patients’ bodies are CCTV’d and perhaps
even scanned in different ways while what they look like and say is
digitally recorded to be searched for diagnostic signs, keywords and
patterns. Stop 2: they are sent on a second conveyor line where blood
tests or further scans are conducted automatically. Stop 3: they are
either discharged from this fully automated hi-tech hospital ‘garage’
– where they are given any necessary drugs, a dose of radiotherapy,
are either then conveyed to a stationary ‘cubicle’ for further
processing – or else to an operating theatre ‘manned’ entirely
by robots. And all this with the whole technological process merely
supervised by a handful of technicians and actual human doctors.
Finally, Stop 4: either a crematorium or a bill to be paid – as in
coming to the exit barrier of a car park.
Not only is this science fantasy prospect
a conceivable one – it is also an entirely rational
one in the framework of
biomedicine. Which only goes to show how totally irrational
medical ‘rationality’ can be. For what this picture essentially
lacks is what is most essential, not just to healing (as opposed to
mere ‘medicine ’) but to human health as
such – namely the care and
care-giving of other human
beings – not to mention
truly human
insight, observation, empathy, as well as human life experience and
life understanding. On the other hand, it is no
less conceivable that the
development and refinement of ‘automated medical diagnosis’ could
serve precisely to free
doctors of the future from their current role as human computers,
allowing them instead to devote most of their time and awareness to
their human role as healer, care-giver or ‘life doctor’ – there
to explore the patient’s experience of illness and its meaning
in the context of their lives and relationships – something no
computer software or database will ever be able to do. In this way
biomedical knowledge would find its true place as a tool subordinate
to the art of
healing – rather than becoming a substitute for that art. And since
the terminology of biomedical science itself is as rich in metaphor
as in ‘fact’, more precise and accurate biomedical diagnoses
would also allow for more accurate and penetrating analyses of their
own symbolic dimensions
of meaning.
Until we reach this point however, next
time you see a doctor bear in mind the question posed at the start –
what is actually going on in the doctor’s mind and what is he or
she actually there
for? For it will make a world of difference to your health whether
the doctor is only there for a specific ‘what’ (for example, to
go through a quasi-automated mental-diagnostic process and then refer
you on for further medical processing) or whether the doctor is there
for a very specific human being or ‘who’ – for
you.
Reference:
Zigmond, David If
you want good personal Healthcare see a Vet
No comments:
Post a Comment