IT robots take a bite from the apple

April 9, 2009

Automation is the grease that makes workflow glide. Single clicks and macros are lesser elements of this process. The real gains are made under the covers of IT systems by algorithms with preprogrammed agenda. Far more intelligent tools than these will soon be needed to handle the wave of EMRs gathering off the shores of U.S. healthcare.

Automation is the grease that makes workflow glide. Single clicks and macros are lesser elements of this process. The real gains are made under the covers of IT systems by algorithms with preprogrammed agenda. Far more intelligent tools than these will soon be needed to handle the wave of EMRs gathering off the shores of U.S. healthcare.

No one will be satisfied that EMRs are collecting and dispersing huge quantities of data any more than the mayor of Philadelphia would be happy to see hydrants spewing water at the scene of a fire. All that information collected by the hundreds, if not thousands, of EMR systems due to flood the healthcare system in coming years will have to be channeled to specific purposes, things like assessing reimbursement trends, figuring patient outcomes, and identifying fraud and corruption.

Who in their right mind would want to pore over these data to tease out the minutia they contain? How could institutions afford to pay a labor force large enough to crunch those numbers? The answers may lie in a new breed of healthcare IT, one that spins algorithms into artificial intelligence. The first such technology has already arrived: a robot called Adam, a computer system that automates the scientific process.

Created by the Biotechnology and Biological Sciences Research Council, Adam recently became the first machine to independently discover new scientific knowledge. Its creators, scientists at Aberystwyth University and the University of Cambridge, describe Adam's exploits in a paper published April 3 in the journal Science.

Acting autonomously, the robot discovered simple but new knowledge about the genomics of a type of baker's yeast that scientists use to model complex life systems. If done by people, this kind of research is "difficult and irksome," according to Ross King, who led the research at Aberystwyth University, "but [it's] easy for robot scientists."

In its study of baker's yeast, Adam hypothesized certain genes code for specific enzymes that catalyze biochemical reactions. The robot then devised experiments to test these predictions, ran the experiments using laboratory robotics, interpreted the results, and repeated the cycle.

Encouraged by Adam's success, King envisions teams of human and robot scientists working together in laboratories, their artificial intelligence laying the bricks for a foundation that leads to grander discoveries. Adam's first companion, aptly named Eve, is being designed to help scientists look for new drugs to fight tropical diseases.

Could the offspring of Adam and Eve work on mundane processes that one day transform medical practice in this country? With these two robot scientists already plowing new ground, could finding the answer be as simple as switching orchards?