sudok1 - Fotolia
BOSTON -- From Medtronic's point of view, the key to effectively helping patients with diabetes better control their chronic disease does not lie solely in the medical devices the company manufactures.
That notion forms the basis of a partnership between Medtronic, based in Minneapolis, and IBM Watson Health that started in April.
What will ultimately increase the effectiveness of the involved technologies is "the convergence of devices with data and analytics, and the software to power that," said Jeff Ruiz, general manager of San Antonio operations at Medtronic. "This is why the partnership with Watson made sense for us."
Ruiz spoke this week at the Global Pediatric Innovation Summit and Awards in Boston, an event Medtronic and IBM sponsored.
IBM Watson Health connects the data dots
IBM Watson Health is a business unit within IBM that aims to provide insights to providers, insurers and patients using Watson's cognitive computing power.
Watson is able to make connections based off data and information faster than any person, said Deb DiSanzo, general manager for IBM Watson Health in Boston. Whether Watson is looking at a group of studies that have been done or at a database, the software gives a list of possible care pathways and makes a suggestion about which one it thinks is the best course to take.
But IBM Watson Health offers more than cognitive computing technology; it offers an ecosystem of partnering companies, DiSanzo said. Some of the Watson partners include heavyweights such as Epic Systems, CVS Health and Apple.
Ultimately, these partnerships open up more possibilities to deliver quality care, ranging from how Watson can assist Epic to improve the interaction between patient and physician, to how Watson can help a nurse at a CVS Minute Clinic take better care of a patient they may not have ever seen before, and, therefore, may not know the patient's history, DiSanzo said. Not only that, but Ruiz pointed out that these insights derived from Watson need to be delivered to patients in ways they can access it, for example, via smartwatches or smartphones.
"We have to evolve the capability and the way we're building the solutions in an agile way to be able to get it to the right place at the right time in a form that the patient will want to use," Ruiz said.
This approach is part of Medtronic's work with IBM Watson Health. "What we did with Watson right out of the gate is we did a study taking all of the data that we have," Ruiz said.
Ruiz is referring to the data of 195 million patients in Medtronic's database that the company has accumulated over the years. Ruiz explained that Watson was able to sift through the data and cluster the patients into six groups, thereby reducing variability.
"[We] now predict hypoglycemic events with up to 90% accuracy three hours in advance," Ruiz said. Something a device or person could never do -- at least, not as fast as Watson. Now, Medtronic is able to get those insights into the hands of doctors and patients faster than before.
"Now, not only would you get notified when it's happening, you would get notified in advance, so it never has to happen," Ruiz said.
Patient self-care is a must with diabetes
Ruiz also added that when it comes to diabetes, 90% of the care falls on the patients' shoulders in between doctor visits, which means being able to deliver such helpful information to the patients in an effective manner is crucial to improving quality care.
"It's really what happens in between that makes the difference in terms of whether or not [the patient will] be able to control their blood sugars and avoid some of the horrific complications associated with diabetes," Ruiz said.
Further, using all this data is also a step in the direction of precision medicine, he said.
"As we leverage that database, I think it does then begin taking it down to precision medicine," Ruiz said. "[Diabetes is] individualized, and every patient is different."
Definition of cognitive computing
Startup deploys Internet of Things sensors to help diabetics
Medtronic links devices to remote patient monitoring