Description:
Summary
Artificial pancreas devices help maintain healthy glucose levels in patients with type 1 diabetes (T1D), but incidences of hypoglycemia are still common after episodes of eating, sleeping and exercise. Oregon Health & Science University researchers have developed an algorithm that features adaptive responses to these factors as well as individual variability, allowing for more personalized and accurate control of type 1 diabetes.
Technology Overview
Fluctuations in glucose and insulin sensitivity occur in a variety of situations, often making it necessary for individuals with T1D to manually administer insulin even when using an automated insulin/glucagon pump. The laboratory of Dr. Peter Jacobs has developed a novel control algorithm for use in artificial pancreas devices, which provides more adaptive responses to common causes of glucose dysregulation, including eating, sleeping and exercising, than previous algorithms. These adaptive algorithms were tested using in silico T1D virtual populations and significantly reduced time spent in hypoglycemia compared to a traditional artificial pancreas. Importantly, this algorithm also includes a feature that allows the control model to adapt over time to match the insulin sensitivity of the patient, providing a personized medicine level approach to T1D glucose regulation. This technology could be utilized in a mobile app, insulin pump or any computerized device and has the potential to reduce the need for manual insulin dosing and improve glucose control for T1D patients.
Publications
Jacobs et al., "Integrating metabolic expenditure information from wearable fitness sensors into an AI-augmented automated insulin delivery system: a randomised clinical trial" The Lancet 5(2023):E607-E617. Link
Resalat et al., “Adaptive tuning of basal and bolus insulin to reduce postprandial hypoglycemia in a hybrid artificial pancreas.” Journal of Process Control 80(2019): 247-254. Link
Resalat et al., “Design of a dual-hormone model predictive control for artificial pancreas with exercise model.” Conf Proc IEEE Eng Med Biol Soc (2016): 2270-2273. Link
Resalat et al., “Evaluation of model complexity in model predictive control within an exercise-enabled artificial pancreas.” IFAC-PapersOnLine 50(2017): 7756-7761. Link
Licensing Opportunity
This technology is available for licensing and collaborative agreements.