Diagnosis of Illnesses - Black Magic
When a person feels ill and talks to his physician, he relies on the judgement of a professional. But quite often physicians disagree. Why is there any disagreement at all in a scientific field?
The problem is that human body is so complex that no one really knows how it all works. For many illnesses, even science hits the wall of complexity and reaches its limits. A solid diagnosis requires juggling hundreds of variables and the sequence of their interactions. This is way too complex for human brains to handle.
Computers shine in dealing with complexity and analyze millions of possibilities in seconds and deliver a much more precise diagnosis. But most doctors do not use computers as they do not trust them. The result is computers are rarely used except to look up data. Doctors are a conservative lot and do not want computers to replace them.
The cost of false positive diagnoses have become so prohibitive that insurance companies are very interested in automated diagnosis using supercomputers.
Increasingly, the computerized diagnosis will start replacing human doctors and will also suggest remedies. It is impossible for humans to compete with machines over the long run.
A number of projects are underway now. The most exciting one is IBM's Watson machine being populated by data from Columbia University Medical diagnosis database. It is a learning machine that incorporates all new information in real time and continuously refine its diagnosis.
Humans are suddenly confronting a new reality of smart machines that can far exceed their creators in physical as well as intellectual abilities. We already have come to trust the flying machines flown by computers. We will soon do so in healthcare. Human will still be the masters, but will be increasingly dependent on machines.
Automation is not about trying to simply duplicate human skills, but to exceed it by a wide margin and be a lot more trustworthy than humans.
KK
