Sep 18 2019
The ARTORG Center for Biomedical Engineering Research of the University of Bern has received a grant from JDRF, the leading international sponsor for type 1 diabetes research.
Prof. Dr. Stavroula Mougiakakou, ARTORG Center, University of Bern. (Image credit: ©ARTORG Center, University of Bern.)
Thanks to this prestigious research grant, a research team headed by Stavroula Mougiakakou will explore a large, real-world dataset to create sophisticated algorithms for automated insulin delivery that can predict seriously high or low blood sugar levels. The goal is to enhance and personalize insulin treatment.
Diabetic patients have to maintain their blood sugar levels at a standard range at all times. Currently, scientifically confirmed automated insulin delivery (AID) systems are available as tools to improve self-management. These systems allow diabetic patients to more effectively control their condition to avoid hyperglycemic (high-glucose) and hypoglycemic (low-glucose) and events.
But these tools still have certain limitations, as the algorithms used in these systems do not react satisfactorily to variables impacting the blood sugar variations in individuals, such as physical activity or food intake.
Bernese Research Group Succeeds with Advanced Algorithms
The laboratory Artificial Intelligence in Health and Nutrition of the ARTORG Center for Biomedical Engineering Research of the University of Bern suggests the use of big data and deep and reinforcement learning technologies (machine learning tools) to enhance the prediction accuracy of AID algorithms. The Artificial Intelligence (AI) algorithms will be taught to predict seriously high or low blood sugar levels in real life circumstances.
If we can predict future blood glucose levels, we can provide early warnings and thus improve each patient’s safety.
Dr Stavroula Mougiakakou, Professor and Project’s Principal Investigator, ARTORG Center, University of Bern
Prof. Dr. Mougiakakou also heads the laboratory at the ARTORG Center.
Prof. Mougiakakou’s group is one of only eight laboratories to receive the coveted research grants, which were awarded through a request for applications (RFA) by the US-based diabetes research foundation JDRF. The grant of approximately 144,000 USD most significantly offers access to big data, comprising diabetes-specific patients’ data from thousands of insulin pumps ad glucose monitors.
The de-identified data was collated by Tidepool—a nonprofit organization dedicated to making diabetes data more accessible, meaningful, and actionable for clinicians, diabetic patients, and scientists—through the Tidepool Big Data Donation Project.
We are honored and proud that JDRF recognizes the potential of and our expertise in applications of AI in diabetes. This grant gives us a unique opportunity to access big diabetes-related data and use it synergetically with advanced AI algorithms to uncover patterns and trends that bring us closer to more precise and personalized insulin treatment.
Dr Stavroula Mougiakakou, Professor and Project’s Chief Investigator, ARTORG Center, University of Bern
In the late 1990s, Prof. Mougiakakou presented the use of AI in insulin treatment optimization.
Machine Learning to Gain Diabetes Insights from Big Data
The data JDRF and Tidepool will offer access to has been de-identified and unified in purposeful ways for the use of clinicians and scientists.
“This data is a big step forward for our research,” stated Qingnan Sun, PhD student at the ARTORG laboratory working on the JDRF funded project. “The data access will help us to refine the algorithms that are used in AID systems, making it possible to warn a person at least half an hour before they develop hypo- or hyperglycemia.”
Personalizing Blood Sugar Predictions
“First the AI algorithms will analyze glucose data to detect for each person, how age, bodily fitness, insulin treatment, number of years with the disease, as well as daily routines influence his or her glucose control, ” explained Prof. Mougiakakou.
Subsequently the model uses these findings to predict hypo-or hyperglycemic events early enough so that the person with diabetes can react and prevent their onset. It is important to mention that the model will continue to learn individual’s pattern and habits while in use.
Dr. Stavroula Mougiakakou, Professor and Project’s Chief Investigator, ARTORG Center, University of Bern