IBM’s Watson Addresses Errors of Diagnosis

I’m beginning to see a pattern here. Two weeks ago, I wrote about clinical decision support in context of Dr. Larry Weed’s new book. Two weeks before that, I commented about physicians worrying that patients would perceive them as being incompetent if they relied on CDS. Today, I’m back to the same topic.

Deny the obvious all you want, physicians, but clinical decision support is coming, and once it’s here, it’s not going away.

I just got back back from the new IBM Healthcare Innovation Lab in downtown Chicago, the company’s third such center in the U.S. and eighth worldwide. While kickoff included a “healthcare leadership exchange” with such thought leaders as HIMSS CEO Steve Lieber and Allscripts Healthcare Solutions Chief Innovation Officer Stanley Crane, the real star was not a person, but a computer. IBM’s Watson, to be specific.

People stayed after lunch mostly to see a demo of Watson processing healthcare data—and IBM Chief Medical Scientist Dr. Marty Kohn said this was the first audience to see this demo. Make no mistake, IBM is positioning Watson as a clinical decision support tool, particularly for the much-ignored area of diagnostic decision support.

Saying that perhaps 25 percent of all healthcare errors are errors of diagnosis, Kohn noted how getting the diagnosis right can prevent all kinds of unnecessary complications and spending. “Of course, if you’ve made the wrong diagnosis, picking the right course of treatment becomes a challenge,” Kohn said.

And after the diagnosis, Watson can prevent treatment errors by, say, scanning EMR data for patient allergies to recommend against a drug that might cause a harmful interaction, then suggest alternative therapies. Kohn presented the case of a 29-year-old pregnant woman who was diagnosed with Lyme disease. A common treatment is the antibiotic doxycyline, but Kohn noted that it’s contraindicated during pregnancy.

Watson, according to Kohn, draws preliminary conclusions according to presenting symptoms, then scans multiple sources of information to present recommendations. Watson does look at the notoriously incomplete and inaccurate Wikipedia, Kohn said, mostly because that user-edited site covers so many topics, but then verifies information from other sources.

Watson then displays reasons why it believes the diagnosis may be correct so the doctor can make an informed decision. “It won’t let you ignore all the possible diagnoses,” Kohn said. But it won’t actually make the final call. “Watson is going to be in a supportive role rather than actually making decisions.” Kohn added.

What the supercomputer does is process vast amounts of data in a short amount of time., something that even the sharpest human mind could never do. And that’s what clinical decision support is supposed to be all about.

About the author

Neil Versel

Neil Versel


  • Useful tool indeed. I hope Watson will be able to decipher the “notoriously incomplete and inaccurate” information put out by the FDA, Big Pharma, et al and make really useful diagnoses; or perhaps it will succombe to GIGO (Garbage In, Garbage Out).

  • Yes yes yes!

    I’m just reading (long overdue) Jerome Groopman’s superb How Doctors Think. In my view it’s going to be absolute an “must read” for any e-patient who wants to be an effective partner with their physicians. And, importantly, it talks at great length about the difficulties of diagnosis.

    When IBM first announced that Watson was going into health, this is exactly what I hoped: it could be a high-speed research server, roping in all the information in the world – literally! – and serving it up on a platter to the harried physician.

    Impossible? Well, my wife the veterinarian says one place where she works, a system does propose possible diagnoses based on lab results, and it’s “pretty good.”

    Keep an eye on this for us, Neil.

  • I’ll be interested to see this develop for sure. Although, the example that they gave you of checking a drug with a pregnant lady is pretty weak. I would think they’d come up with a better example that the human mind likely wouldn’t have been able to do as easily.

  • How about the medical profession joins the rest of us who recognize that things like Watson make us all smarter. Guys it’s just a tool. Use it. We like to use a phrase in our leadership meetings at work. “let go of the vine, don’t fear the unknown.” This tool has the ability to compare huge data sets in the blink of an eye and report back findings that providers shouldn’t be scared of rather be grateful for in treating patients. It’s not going to hurt you…

  • Peter Dolan,
    I think doctors do need to use such tools. Although, I think there’s something to say about patients accepting doctors who use the tools as well. Patients expectations need to be that doctors aren’t going to know everything and that they’re going to use tools like this to provide the best care possible. Not all patients are this way, just like all doctors aren’t against this tool. Either way, we need better involvement so the tools get even better and improve healthcare.

  • do not be against can work faster than any physician!certainly,they should be under our general control, but we must start using WATSON as soon as possible!

Click here to post a comment