What: “Watson was gradually given information about a fictional patient with an eye problem. As more clues were unveiled — blurred vision, family history of arthritis, Connecticut residence — Watson’s suggested diagnoses evolved from uveitis to Behcet’s disease to Lyme disease. It gave the final diagnosis a 73 percent confidence rating.”
How: “The medical version of Watson was trained with medical textbooks and journals, electronic health records, and sample questions from medical students. However, unlike Jeopardy!, where Watson was required to give a single correct answer, researchers are using Watson’s computing power to receive a set of symptoms and offer several possible diagnoses, ranked in order of the computer’s confidence. ”
Why: “Physicians are unlikely to blindly accept a diagnosis from a computer, so Watson’s purpose is to reduce physician’s mistakes by handling information overload, offering multiple options, and helping them not become too attached to a single diagnosis.”
From medGadget here.