One out of every 11 adults has diabetes. That works out to 415 million people worldwide. 12 cents out of every healthcare dollar spent goes toward diabetes and related complications, for a total of $673 billion every year. If we can reduce those costs by just one percent, that would result in nearly $7 billion in savings every year. This explains why many of the best scientific minds are working hard on this problem. And now they have a new mechanical colleague who is joining the fight: IBM’s Watson.

IBM Watson Health has announced a partnership with Medtronic, working together to develop an app to help prevent, identify, and manage diabetes. Using IBM’s cognitive computing platform to analyze anonymized data from 10,000 patients, the partnership has looked for patterns and trends in insulin use, continuous glucose monitor results, and nutritional information. The result is a hypoglycemia prediction model that has demonstrated 75 to 86 percent accuracy within a two- to four-hour window, based on preliminary testing.

We’ve seen Watson’s abilities applied to other healthcare issues, from reading medical images to helping seniors. This new project tackles a major problem with worldwide implications and the potential to have a major impact on the healthcare budgets of every country.