How IBM is fighting diabetes with data

bethbohnert    November 28, 2017

When our daughter was diagnosed with Type 1 diabetes (T1D) at the age of 17, our family was plunged into an unfamiliar and frightening world.

Suddenly, we were learning about blood sugars, insulin-to-carbohydrate impact and hypoglycemia. Hardest of all was to watch her come to terms with having a life-threatening disease.

To manage her diabetes, my daughter must measure her blood sugar levels throughout the day by pricking her finger and using a glucometer to test the resulting blood drop. Then she needs to calculate the amount of insulin she needs, based on her activity and carbohydrate intake, and inject herself with insulin four times a day. Sometimes, her blood sugar levels fall too low, putting her at risk of insulin shock, which can be fatal if not treated quickly.

Of course, our family isn’t alone with this struggle. It’s estimated that more than 300,000 Canadians have T1D and, around the world, the number of children under 14 with T1D is growing by 3% each year. No one knows what causes T1D and, as yet, there is no cure.

But thanks to a supercomputer named Watson, research into TD1 has taken a giant step towards the day when we may be able to predict and even prevent TD1.

New technology is making it somewhat easier for people with T1D to manage the disease. My daughter can now quickly measure her blood sugars herself or get her HbA1c results at her endocrinologist appointment without having to go to a lab. And many diabetics now use continuous glucose monitors combined with insulin pumps allow them to receive regular, automatic doses of insulin.

But recent partnerships between IBM, JDRF (the leading global organization funding T1D research), as well as medical device manufacturers could move us far beyond where we are now.

These organizations hope to harness the power of IBM’s Watson, a cognitive computing system, to analyze and interpret masses of data from thousands of individuals with T1D.

The challenge, as Jessica Dunne, JDRF’s Director of Research and Discovery notes, is that “we have supported multiple, long-term studies tracking disease progression in different groups of people around the world. But to unlock its full potential, we need to view the data holistically. Unfortunately, the data sets are independent, having been collected in different ways, at different times, in different locations, by different people.” (Dunne, 2017)

Like a giant brain (it’s the size of 10 refrigerators) Watson is able to observe, interpret, evaluate information and decide on a course of action, at speeds and volumes that are impossible for humans.

With these skills, researchers believe that Watson will be able to help us

  1. Predict who’s at risk of developing T1D
    Watson will analyze data JDRF has collected through its global research projects. It will look at a range of variables including genetic information, family history and more to find common features and quantify the risk for TD1. This will allow JDRF to identify risk factors and develop models for predicting who will get the disease.
  1. Make diabetes more easily manageable for those who already have T1D
    Medical device companies have developed continuous glucose monitors, insulin monitors and apps which are used by people with T1D. Watson is able to analyze the data from these devices to predict when low blood sugar levels will drop as early as three hours in advance, allowing the patient to adjust their insulin carbohydrate intake or activity to compensate. Kim Kemble gave a fanciful but enlightening illustration of what this might look like in a recent blog post:

Me: “Watson, I’m going to mow the lawn.”

Watson: (understanding how I’ve reacted to this type of strenuous exercise in the South Florida heat, what my current blood glucose and active insulin levels are, what I’ve eaten and probably so much more). That’s great, Kimberlee. I’ve set a temporary basal for the next hour and a half – and I really think you should eat 30 grams of carbohydrates before you get started. And don’t forget to hydrate.”

Me: “You got it.”

  1. Help doctors provide personalized care to people with T1D No two people develop and experience TD1 in exactly the same way. Watson can analyze genetic information to help diagnose or treat the disease on a more individualized basis, a practice known as precision medicine.One of the major goals for diabetes treatment is the development of an artificial pancreas, which will free patients from the need to test their blood or calculate their insulin units. Watson can help medical device manufacturers move toward realizing this goal by identifying algorithms that will enable the artificial pancreas to anticipate fluctuations in blood sugar levels and respond by automatically adjusting insulin or glucagon dosages to keep blood sugars stable.
  1. Use data to find a cure for T1D
    In later phases of the collaboration between JDRF and IBM, researchers hope to analyze even more complex data that will unlock the secrets of TD1.As Derek Rapp, JDRF President and CEO, says “JDRF supports researchers all over the world, but never before have we been able to analyze their data comprehensively, in a way that can tell us why some children who are at risk get T1D and others do not. IBM’s analysis of the existing data could open the door to understanding the risk factors of T1D in a whole new way, and to one day finding a way to prevent T1D altogether.” (IBM, 2017)

Lessons for Others

Here’s what IBM and Watson can teach us about the future of big data:

Cognitive computing systems can process medical information and other data much faster and much greater quantities than humans can. If Watson can help researchers build diagnostic and treatment tools, this could eventually free doctors to concentrate on the relationship-building side of medicine, a benefit that could also be applied in other industries.

Watson’s data-crunching capabilities can, as Rapp notes, help us understand data “in a whole new way” but at this point humans still need to prepare and input the data in order for the system to develop its body of knowledge. But even with this limitation, cognitive computing is a huge step forward in addressing our biggest challenges, whether they’re in healthcare or another sector.

Organization: IBM
Industry: Computing
Name of Organization Contact: -

Authored by: Beth Bohnert

If you have concerns as to the accuracy of anything posted on this site, please send your concerns to Peter Carr, Program Director, Social Media for Business Performance.


Dunne, J. (2017, August 18). JDRF and IBM piece together the puzzle of type 1 diabetes. [Blog post]. Retrieved from

Lagasse, J. (2017, August 21). IBM, JDRF partnership using machine learning methods to tackle Type 1 diabetes. [Blog post]. Retrieved from

Kemble, K. (2016, September 13). Data and more data: how IoT could impact the lives those with diabetes. [Blog post]. Retrieved from

Seeking Alpha. (2015, April 15). IBM collaboration is Medtronic’s fifth diabetes deal in a month. [Web page]. Retrieved from

IBM. (2017, August 18). JDRF and IBM collaborate to research risk factors for Type 1 Diabetes in children. [Media Release]. Retrieved from