Digital twin tech powers smarter insulin delivery in artificial pancreas trials
Researchers in the US have designed an interactive artificial pancreas system that uses digital twin technology to control type 1 diabetes, showing promising results in early trials.
Designed by scientists at the University of Virginia, the new technology called Adaptive Biobehavioral Control (ABC), significantly improved blood sugar control by adapting more closely to the user’s changing physiological needs.
Boris Kovatchev, PhD, a director of the University of Virginia’s Center for Diabetes Technology and lead author of the study, revealed that the tech optimizes the automated insulin delivery system in the artificial pancreas every two weeks.
It also gives users access to a digital twin simulation, allowing them to safely experiment with different strategies for managing their blood sugar before applying changes in real life.
“Artificial pancreas systems require adjustments by those who use them to adapt to a person’s changing insulin demands,” Kovatchev pointed out.
Smarter insulin delivery
Digital twins, first used during NASA’s Apollo missions in the 1960s, are virtual models that replicate real-world systems or processes, mirroring their features, behavior, and performance in a simulated environment.
But despite having been around for decades, this is the first time the technology has been linked to each user through a personalized cloud-based model, giving people with diabetes a safe way to test and understand how different adjustments might affect their artificial pancreas before making real changes.
A six-month-long clinical study carried out by the research team found that participants using the ABC technology increased their time in a safe blood-sugar range from 72 to 77 percent and lowered their average blood sugar (A1c) from 6.8 to 6.6 percent.
According to Kovatchev, while automated insulin delivery systems like the artificial pancreas have helped users manage type 1 diabetes, the new technology aims to solve two key remaining challenges.
The first challenge, the scientist revealed, is improving daytime blood sugar control, when activities like eating and exercising cause frequent spikes and drops.
The second, on the other hand, is overcoming the plateau most users hit after early progress, typically stabilizing at 70 to 75 percent time in range, a trend researchers believe is due to difficulties in adapting to how the system functions over time.
Shaping the future of medicine
The research team revealed that the ABC technology tackles these challenges through two main strategies, including the use of digital twins, computer simulations that replicate each user’s metabolic system.
In addition to adjusting the artificial pancreas in response to changes in users’ physiology and behavior, the system also offers an interactive simulation tool.
This in turn, allows users to safely test different settings, such as tweaking the overnight insulin delivery rate, before applying them to their actual device.
“Human-machine co-adaptation is critical for conditions like type 1 diabetes where treatment decisions are made both by the artificial pancreas algorithm and the person who wears it,” Kovatchev elaborated.
“Digital-twin technology is very helpful in facilitating this co-adaptation,” he concluded in a press release.
Source: Interesting Engineering
Does Retinol Reverse Signs of Aging? Here’s The Science.
Digital twin tech powers smarter insulin delivery in artificial pancreas trials