UIC Researchers Create Voice-Enabled Coach to Manage Type 2 Diabetes
More than 30 million Americans had diabetes in 2015, according to the American Diabetes Association. That’s just over 9 percent of the U.S. population.
The vast majority of those with the disease have Type 2 diabetes, which most often occurs in middle-aged adults and older adults, according to the National Institute of Diabetes and Digestive and Kidney Diseases.
While there is no cure for the disease, it can be managed through medication and healthy behaviors such as exercising, eating nutritious meals and quitting smoking.
“A lot of these lifestyle changes can be very stressful,” said Olusola Ajilore, associate professor of psychiatry in the University Illinois at Chicago College of Medicine. “I teach a class on diabetes and emotions for patients with Type 2 diabetes and some of the challenges they bring up are having to change their diet or increasing their physical activity or exercise.”
To help newly diagnosed patients adjust, Ajilore and fellow researchers at UIC created DiaBetty, described as a diabetes coach and educator that’s sensitive and responsive to a patient’s mood. Using voice-enabled technology, DiaBetty is programmed to offer context-specific diabetes education, remind patients to take their medications and provide social support.
DiaBetty is a also a finalist in a contest using Amazon’s voice technology to improve the experience of newly diagnosed Type 2 diabetes patients. As a finalist, researchers received $25,000 to develop their technology and 10,000 in Amazon Web Services credits to cover usage costs. A winner will be announced in October and receive a $125,000 grand prize.
“DiaBetty brings back this mind-body connection,” said Alex Leow, associate professor of psychiatry and bioengineering in the College of Medicine. “People don’t understand medical illness and mental wel-lbeing are interconnected.”
“DiaBetty not only addresses a patient’s physical health but their social and emotional well-being as well,” said Ajilore, who has spent more than a decade studying diabetes and depression. “Patients with Type 2 diabetes are significantly more likely to develop major depression and that can impact their diabetes management, making it more difficult for them to adhere to treatment plans with medications, nutrition or physical activity.”
To help patients adhere to treatment plans, DiaBetty adapts its response based on patients’ moods, which it can detect by analyzing the text of their speech as well as the acoustic features of their speech, including speed, tone, pitch and volume.
If a patient were to say, “I feel terrible,” the word “terrible” would be tagged negatively and indicate a bad mood, explained Ajilore. DiaBetty would then try to boost the patient’s mood, by offering relaxation tips, for example, or sharing an inspirational message from another Type 2 diabetes patient.
“For people to be successful in managing their diabetes it does take a lot of psychological reserve to make all the adjustments they need in their life to remain healthy,” Ajilore said.
“We really hope DiaBetty will be a tool to provide ongoing support ... because most diabetes education comes in a single appointment with a diabetes educator, nutritionist or nurse. In order to be successful with self-management a patient needs ongoing support and management.”
Along with Leow and Ajilore, UIC researchers Faraz Hussain, Jun Ma and Ben Gerber and Melvin McInnis at the University of Michigan helped develop DiaBetty.
Follow Kristen Thometz on Twitter: @kristenthometz
May 23: The BiAffect app, the winner in a contest using Apple’s Research Kit, was developed to predict manic episodes and depressive moods in people with bipolar disorder, based on how they interact with their cellphones.
April 10: The new Office of Chronic Disease Prevention and Control will “streamline” the city’s current efforts and look for new ways to help prevent and improve the management of chronic conditions and diseases.
Jan. 17: Can an app reduce a person’s depression or anxiety? Or predict when someone might have a manic episode? Researchers from two local universities say yes.