IBM and Medtronic look to combine powerful analytics and cognitive computing to develop a host of personalized diabetes management technologies.

Kristopher Sturgis

At CES, Medtronic shed light on its partnership with IBM's newly formed Watson Health unit, which is seeking to bring innovative diabetes technologies to patients.

The goal is to develop a Watson-enabled app that enables a Medtronic device to continuously gather and analyze data in real-time. The app and device will work in tandem to enable patients to pool the data with their own electronic medical records, insurance information, and even population health data to develop a network of information that can optimize patient care through advanced analytic models.

"Since we announced our partnership with Medtronic last April, we have been working aggressively to advance research around how cognitive computing could help people with diabetes," says Kathy McGroddy, vice president of Partnerships & Solutions at IBM Watson Health.

Medtronic became one of the first developers to create a device that can send blood sugar data to a smartphone app. The next step was coming up with new ways to use that data to enable patients to better manage the disease. IBM hopes to help bridge that gap through the development of personalized care plans that can provide real-time decision support in an effort to improve care efficiency and outcomes.

"This app will apply the power of Watson to real-time data from Medtronic's insulin pumps and glucose monitors, as well as other available data points, to help people with diabetes manage their health, and control their blood sugar," McGroddy says. "IBM and Medtronic are also planning to develop a variety of other diabetes solutions that leverage Watson technology, such as solutions for healthcare systems and advancements in artificial pancreas technology."

So far in pilot research, IBM and Medtronic applied cognitive analytics to 600 anonymous patient cases using data from Medtronic insulin pumps and glucose monitors. The study found that the technology was able to predict hypoglycemia (low blood sugar) up to three hours in advance of onset. Presumably this kind of early detection would enable virtually any patient with diabetes to take appropriate action and prevent any serious potential health threats.

"There is truly an explosion of data that has never been harnessed to deliver the types of insights we are creating," McGroddy says. "We are trailblazing solutions that we believe could transform how people with diabetes manage their health."

According to the International Diabetes Federation (IDF), more than 387 million people worldwide live with diabetes, and managing the disease has proven to be both difficult and costly for patients and healthcare providers. Additionally, the IDF estimates that global spending to manage and treat complications from diabetes totaled at least $612 billion in 2014.

As for the immediate future, McGroddy says that a first version of the app is expected to roll out this summer, and will focus on hypoglycemia. IBM intends to build on its initial version with new features and capabilities in an effort to continue to improve user experience and overall patient care.

"We see so much potential for Watson to help doctors, patients, and researchers make sense of all of this disparate health data -- but we at IBM know we can't do it alone," Mcgroddy says. "That's why partnerships like the one we have with Medtronic are at the heart of IBM Watson Health. We can't tackle diabetes alone, but together with Medtronic and other partners, we think we can."