It’s a futuristic idea to be sure: Harnessing the intelligence of IBM Watson, the Jeopardy-winning supercomputer, to support cardiac-care recovery, reduce hospital readmissions, and save healthcare costs.

But that’s what the Colorado-based healthcare nonprofit Centura Health and consumer enterprise platform vendor Welltok are doing right now with CaféWell Concierge, currently in pilot with consumers who are transitioning back to everyday life after experiencing a heart condition.

See all of our HIMSS16 previews

“Cognitive computing mimics the way humans think by combining artificial intelligence and machine-learning algorithms,” said Jeff Margolis, CEO of Welltok. “IBM Watson understands natural language in context; it can determine the intent of a phrase or question and provide a pertinent, useful response.”

Margolis, along with Pam Nicholson, senior vice president of strategy for Centura Health, will present “Applying Cognitive Computing to Population Health” at HIMSS16.

Nicholson said that the technology the companies are piloting learns from interactions to provide personalized recommendations over time. For example, if a consumer has opportunities to join a team activity challenge and schedule a one-on-one coaching session, and consistently chooses the team activity, the app would recognize that this person favors social activities and recommend a support group at the local community center.

[Also: IBM Watson picked to help tackle heart disease]

Among the other ways that patients interact with CaféWell: Finding options for cardiac rehabilitation exercises and activities; researching new heart-healthy recipes and dishes at local restaurants, and identifying educational resources and videos on living with heart conditions.

In these ways, cognitive computing is broadening the scope of healthcare delivery “so that it can happen outside what we normally think of as the healthcare setting,'” Nicholson said. “The contrast with traditional health care, where we only get to interact with the consumer when they step inside our four walls and temporarily become a patient, is profound.”

As the real-world pilot users continue to train the application’s “brain,” as Margolis put it, these innovators see ample opportunities for cognitive computing to have an impact in the four areas of health: healthy behaviors, genetics, medical interventions, and environment.

[Like Healthcare IT News on Facebook]

“In five years, we believe that predictive analytics will further evolve and enable providers and consumers to make better health decisions while allowing for care to be highly personalized,” Nicholson said. “This will require us to continue to build solutions that incorporate timely and new data sources, offer reliable, consistent predictive models, deliver convenient and immediate personalized health and lifestyle recommendations, and learn intuitively and rapidly.”

The session "Applying Cognitive Computing to Population Health,” is slated to take place March 3, 2016 from 1 to 2 p.m. in Rock of Ages Theater at the Sands Expo Convention Center.

Twitter: @HenryPowderly

This story is part of our ongoing coverage of the HIMSS16 conference. Follow our live blog for real-time updates, and visit Destination HIMSS16 for a full rundown of our reporting from the show. For a selection of some of the best social media posts of the show, visit our Trending at #HIMSS16 hub.