The lives of people with Type 1 diabetes could be significantly enhanced through algorithms that connect glucose monitors and insulin pumps to automatically regulate blood glucose to healthy levels, in the same fashion that cruise control in an automobile regulates speed.

A new project funded by JDRF, the leading global organization funding Type 1 diabetes research, and led by Wayne Bequette, professor of chemical and biological engineering at Rensselaer Polytechnic Institute, aims to use artificial intelligence and big data techniques to analyze information gathered from thousands of continuous glucose monitors and insulin pumps. Researchers will use that information to improve algorithms that control these critical devices.

People with Type 1 diabetes must test their blood sugar often to decide how much insulin they should inject using a needle or insulin pump. Until fairly recently, blood sugar could only be tested by performing a finger stick to obtain a blood sample that would be analyzed by a glucose meter — a process most people only do 5 to 6 times a day.

Now, continuous glucose monitors can give people a better idea of whether their blood sugar is trending high or low by providing blood sugar estimates every five minutes, without frequent finger sticks.

The data analysis in this research will enable engineers to improve models that predict the effect of insulin and meals on glucose levels, yielding better control of blood sugar levels.

Typically, Bequette said, researchers only have access to limited data during clinical trials. This project will give him and his team a broader, more accurate snapshot of a person’s daily life as they analyze de-identified data collected by Tidepool — a nonprofit that makes diabetes data more accessible, meaningful, and actionable for people with diabetes, clinicians, and researchers — through the Tidepool Big Data Donation Project.

The research will also uncover how often sensor, insulin pump, and infusion set faults occur.

Bequette and his team have developed algorithms that can detect whether or not the signals from the glucose monitoring device can be trusted, to provide a check and balance for those irregularities. They could improve those processes further if they knew how often anomalies are happening, and could further break that data down by age group. “If we look at hundreds of people we can say, ‘Oh, certain problems occur more often in this age group, this type of population, or with this particular type of sensor,’” Bequette said. “If, for example, you find that it’s more likely that people 8 to 12 years old have these types of irregularities, then you can account for that in your algorithm, and provide more personalized control while reducing burden.”

For nearly two decades, Bequette has been developing and improving algorithms to enhance the lives of people with Type 1 diabetes. For example, he has developed and tested a closed-loop artificial pancreas for individuals with Type 1 diabetes. The system automatically adjusts an insulin infusion pump based on signals from a continuous glucose monitor.

Bequette first became familiar with Type 1 diabetes in 1977, when his sister was diagnosed. “At that time there was limited knowledge about the disease,” said Bequette, “so my mother quickly became the ‘decision support system’ that my sister needed. There were no blood glucose meters, no rapid-acting insulin, no insulin pumps, and syringes were not as easy to use as today’s insulin pens. The state-of-technology has advanced significantly in the past 40 years.”

“Data driven decisions, combined with technology, help improve quality of life and reduce co-morbidities associated with diabetes,” said Deepak Vashishth, director of the Center for Biotechnology and Interdisciplinary Studies (CBIS) at Rensselaer.

Source: Rensselaer Polytechnic Institute