A digital twin can help people with type 1 diabetes to better control their condition using artificial pancreas technology, shows research led by the University of Virginia.
Automated insulin delivery (AID) systems combined with continuous glucose sensors, known as an artificial pancreas, are becoming more commonly used by people with type 1 diabetes.
The artificial pancreas technology has the potential to significantly improve blood glucose control in people with this condition. However, using the currently available systems seems to be difficult for people to achieve more than 75% of their time in their target blood glucose range.
“The reason for this saturation effect is generally unclear, but an informed speculation is that patients using AID systems do not adapt well to the system’s actions,” write lead author Boris Kovatchev, PhD, director of the University of Virginia Center for Diabetes Technology, and colleagues in npj Digital Medicine.
“This stems from the fact that the information feedback provided by AID software is typically limited to summary statistics… and, occasionally, advice based on artificial intelligence methods. This information is generally passive and does not provide the user with specific instructions on how to optimize their AID system parameters.”
In this study, Kovatchev and colleagues have generated ‘digital twin’ copies of 72 people with type 1 diabetes using artificial pancreas technology along with software to allow them to trial different scenarios using the digital twin. For example, what would happen if they performed a specific activity and adjusted their insulin in a particular way while doing so?
“Artificial pancreas systems require adjustments by those who use them to adapt to a person’s changing insulin demands,” said Kovatchev in a press statement. “This is the first study that maps each person to their ‘digital twin’ in the cloud and enables people with diabetes to experiment with their own data to learn how their artificial pancreas system would react to changes, in a safe simulation environment, before adjusting their system.”
Having access to the digital twin model and software helped improve the amount of time the people in the study spent in their target blood glucose range from an average of 72% to 77%. Average hemoglobin A1c (average blood-sugar level) also went down from 6.8% to 6.6% in the group.
“Emerging ‘digital twin’ technologies, which program certain characteristics, such as a person’s glucose-insulin metabolism, into a computer application, not only enable rapid therapy parameter optimization but also offer educational support to make diabetes management more accessible,” note the authors.
The researchers say they plan to explore broader applicability of the concept of human-machine interaction for the management of diabetes, the usability of the software they created in different AID platforms and age groups, as well as long-term sustainability of the observed benefits.
