IBM and Roche use AI to forecast blood sugar levels

IBM and Roche are teaming up on an AI solution to a challenge faced by millions worldwide: the relentless daily grind of diabetes management. Their new brainchild, the Accu-Chek SmartGuide Predict app, provides AI-powered glucose forecasting capabilities to users. 

The app doesn’t just track where your glucose levels are—it tells you where they’re heading. Imagine having a weather forecast, but for your blood sugar. That’s essentially what IBM and Roche are creating.

AI-powered diabetes management

The app works alongside Roche’s continuous glucose monitoring sensor, crunching the numbers in real-time to offer predictive insights that can help users stay ahead of potentially dangerous blood sugar swings.

What caught my eye were the three standout features that address very specific worries diabetics face. The “Glucose Predict” function visualises where your glucose might be heading over the next two hours—giving you that crucial window to make adjustments before things go south.

For those who live with the anxiety of hypoglycaemia (when blood sugar plummets to dangerous levels), the “Low Glucose Predict” feature acts like an early warning system, flagging potential lows up to half an hour before they might occur. That’s enough time to take corrective action.

Perhaps most reassuring is the “Night Low Predict” feature, which estimates your risk of overnight hypoglycaemia—often the most frightening prospect for diabetes patients. Before tucking in for the night, the AI-powered diabetes management app gives you a heads-up about whether you might need that bedtime snack. This feature should bring peace of mind to countless households.

“By harnessing the power of AI-enabled predictive technology, Roche’s Accu-Chek SmartGuide Predict App can help empower people with diabetes to take proactive measures to manage their disease,” says Moritz Hartmann, Head of Roche Information Solutions.

How AI is speeding up diabetes research

It’s not just patients benefiting from this partnership. The companies have developed a rather clever research tool using IBM’s watsonx AI platform that’s transforming how clinical study data gets analysed.

Anyone who’s been involved in clinical research knows the mind-numbing tedium of manual data analysis. IBM and Roche’s tool does the heavy lifting—digitising, translating, and categorising all that anonymised clinical data, then connecting the dots between glucose monitoring data and participants’ daily activities.

The result? Researchers can spot meaningful patterns and correlations in a fraction of the time it would normally take. This behind-the-scenes innovation might do more to advance diabetes care and management in the long run than the app itself.

What makes this collaboration particularly interesting is how it brings together two different worlds. You’ve got IBM’s computing prowess and AI know-how pairing up with Roche’s decades of healthcare and diabetes expertise.

”Our long-standing partnership with IBM underscores the potential of cross-industry innovation in addressing unmet healthcare needs and bringing significant advancements to patients faster,” says Hartmann.

“Using cutting-edge technology such as AI and machine learning helps us to accelerate time to market and to improve therapy outcomes at the same time.”

Christian Keller, General Manager of IBM Switzerland, added: “The collaboration with Roche underlines the potential of AI when it’s implemented with a clear goal—assisting patients in managing their diabetes.

“With our technology and consulting expertise we can offer a trusted, customised, and secure technical environment that is essential to enable innovation in healthcare.”

What this means for the future of healthcare tech

Having covered healthcare tech for years, I’ve seen plenty of promising innovations fizzle out. However, this IBM-Roche partnership feels promising—perhaps because it’s addressing such a specific, well-defined problem with a thoughtful, targeted application of AI.

For the estimated 590 million people (or 1 in 9 of the adult population) worldwide living with diabetes, the shift from reactive to predictive management could be gamechanging. It’s not about replacing human judgment, but enhancing it with timely, actionable insights.

The app’s currently only available in Switzerland, which seems a sensible approach—test, refine, and perfect before wider deployment. Healthcare professionals will be keeping tabs on this Swiss rollout to see if it delivers on its promise.

If successful, this collaboration could serve as a blueprint for how tech giants and pharma companies might work together on other chronic conditions. Imagine similar predictive approaches for heart disease, asthma, or Parkinson’s.

For now, though, the focus is squarely on using AI to improve diabetes management and helping people sleep a little easier at night—quite literally, in the case of that clever nocturnal prediction feature. And honestly, that’s a worthwhile enough goal on its own.

(Photo by Alexander Grey)

See also: DeepSeek’s latest AI model a ‘big step backwards’ for free speech

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.