Skip to main content

IBM’s new A.I. warns diabetes patients of dangerous blood sugar levels

Image used with permission by copyright holder

From diagnostic fingernail sensors to the use of its Watson platform to help invent new drugs, IBM has impressively positioned itself at the forefront of medical tech. Today, January 3, it announced the latest project in this field: a new mobile app feature that’s designed to work as an early warning sign for diabetics about the perils of “going low” on their blood sugar levels.

To the uninitiated, keeping these blood sugar levels in check sounds easy: Simply avoid eating food with too much sugar and you’re good to go. However, the reality is that things are more complicated than that. A person living with type 1 diabetes has to make upwards of 180 decisions every single day, all of which can affect their well being. These can range from how long to sleep or exercise to when that person should have a cup of coffee. It’s an enormous amount of data to juggle — and, should consistent wrong decisions be made, the result can be weakness, fainting and, potentially, even death.

Fortunately, where humans struggle to weigh up complex data sets, machine learning tools turn out to be pretty darn great at it. This is what IBM and the world’s largest medical device company Medtronic have attempted with their new predictive tool. Baked into the Sugar.IQ app, IQcast analyzes your data to assess the likelihood of you “going low” within the next several hours.

“Avoiding complications like hypoglycemia is a tremendous burden, but fortunately it is a problem that can help be alleviated by learning from data, which is where A.I. comes in,” Dr. Lisa Latts, deputy chief health officer at IBM Watson Health, told Digital Trends. “Using machine learning models and predictive algorithms, IQcast analyzes multiple signals coming from a user, such as glucose levels, insulin data, food logs, and prior hypoglycemic events to assess whether a user has a low, medium, or high chance of experiencing hypoglycemia within the upcoming four hours.”

The app works in conjunction with Medtronic’s Guardian Connect continuous glucose monitor; adding on top of this a predictive model which gets smarter over time. At the most recent American Diabetes Association Scientific Sessions, the team behind the tool presented data that showed that, when using a continuous glucose monitor in conjunction with Sugar.IQ, users are more likely to achieve an extra 36 minutes per day in a healthy glucose range of 70-180 mg/dL. That includes 30 minutes less time per day in hyperglycemia, and 6 minutes less time per day in hypoglycemia.

The Sugar.IQ app, now with added IQcast feature, is available today to download for iOS in the U.S. In order to function, it must be used in tandem with the Guardian Connect continuous glucose monitor system, built and offered by Medtronic.

Editors' Recommendations

Luke Dormehl
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
IBM’s latest A.I. can help doctors detect how quickly a disease is progressing
ibm probabilistic ai huntingtons diseases doctor seeing patient

While there obviously shouldn't be an outcry for robots to take the jobs of human physicians, that doesn’t mean artificial intelligence can’t play an important role when it comes to helping doctors carry out certain aspects of their jobs -- such as recognizing early symptoms of serious diseases that can go unnoticed. A delay in diagnosis can make it more difficult to implement effective treatments or even find cures.

This is something that IBM wants to help with, courtesy of a new probabilistic A.I. model that can be used to better understand patients’ individual conditions and, crucially, how they are likely to develop. It can give information about a wide range of symptoms -- from shaking hands to mood swings -- and then use this data to help identify the biomarkers of diseases including diabetes, Huntington’s disease, Alzheimer’s, Parkinson’s, and more. The A.I. can help doctors understand more about the progression of these neurological disorders, as well as pinpoint how advanced they are.

Read more
Smell like Skynet: IBM creates an A.I. that will design the perfect perfume
ivm machine learning new perfumes 30581001157 afff68113f k

Using AI to create new fragrances

“I love your perfume.” “Thanks. It’s Skynet.” OK, so that’s not necessarily the most likely conversation for you to have in the near future, but it very well could be. That’s because IBM Research has teamed up with master fragrance maker Symrise to use its artificial intelligence expertise to develop next-gen fragrances, courtesy of IBM’s impressive machine learning technology.

Read more
Meet the game-changing pitching robot that can perfectly mimic any human throw
baseball hitter swings and misses

Who’s your favorite baseball pitcher? Shane McClanahan? Sandy Alcantara? Justin Verlander? Whoever you said, two of the top sports-tech companies in the U.S. -- Rapsodo and Trajekt Sports -- have teamed up to build a robot version of them, and the results are reportedly uncannily accurate.

Okay, so we’re not talking about walking-talking-pitching standalone robots, as great a sci-fi-tinged MLB ad as that would be. However, Rapsodo and Trajekt have combined their considerable powers to throw a slew of different technologies at the problem of building a machine that's able to accurately simulate the pitching style of whichever player you want to practice batting against -- and they may just have pulled it off, too.

Read more