Clinicians and students at Cleveland Clinic are going to be getting some assistance from a Jeopardy champion. Today at the Cleveland Clinic Medical Innovation Summit, IBM Research demonstrated WatsonPaths and Watson EMR Assistant, two new cognitive computing technologies for its Watson computer aimed at assisting healthcare professionals.

WatsonPaths is a diagnostic assistance tool that examines scenarios from medical cases to draw conclusions toward diagnosis. However, rather than simply providing answers, IBM says the system goes further and conveys the reasoning behind its conclusions. The system pulls evidence from reference materials, guidelines, and medical journals and provides students and clinicians with several hypotheses rated on likelihood and certainty. WatsonPaths is also capable of learning over time based on user feedback and interaction with medical experts to grow smarter in its conclusions.

The second techology, Watson EMR Assistant, is designed to help physicians make more informed decisions by processing data from electronic medical records (EMRs) to reveal insights. A single patient's EMR can contain a massive quantity of data that can be difficult to glean much deep insight from. The older the patient is and the more extensive his or her medical history, the more data there is to cull and sort through.

Since Watson is capable of understanding natural language, it can process EMRs more fluidly and understand their content to get a better understanding of a patient's medical history and alert physicians of any potential problems or areas of concern. In a blog post, Michael Barborak, manager of unstructured language engineering at IBM Research, says “It’s not human versus machine, but human plus machine taking on challenges together and achieving more than either could do on its own.”

IBM plans to make WatsonPaths first available to faculty and students at Cleveland Clinic Lerner College of Medicine of Case Western Reserve University to implement into its curriculum for medical students and in clinical lab simulations. In an interview conducted by IBM research, Dr. Neil Mehta, associate professor of medicine and director of education technology at Cleveland Clinic Lerner College of Medicine, says, “When given a case scenario, WatsonPaths, just like our students, would calculate any number of hypotheses—ranking the most-likely solutions. We can then see where it was correct, and where it wasn’t.” He hopes that seeing the differences in WatsonPath's hypotheses versus those of the students will allow for a synergy between the students and the machine. “The students can review the hypotheses generated by WatsonPaths and learn from the exercise. At the same time they could provide feedback to WatsonPaths and help improve its algorithms,” Mehta says.

Mehta hopes that using WatsonPaths to train future doctors will both make them more comfortable with utilizing technologies like this in their decision making and also give patients more confidence that their diagnoses are accurate and have been well thought out.

The current version of WatsonPaths is only a first step. Barborak says plans are underway to improve the system's interactivity. “This is just the start of Watson’s interactions with humans. In the future, you can expect cognitive systems to have written and verbal conversations with people—even debates—all aimed at penetrating complexity so we can make better decisions.”
 

IBM Research has released a video, below, explaining more about WatsonPaths.

  -Chris Wiltz, Associate Editor, MD+DI
[email protected]