Personalized type 1 diabetes decision support using digital twins

Summary
The current technology is the first decision support tool for type 1 diabetes that provides personalized recommendations based on digital twin simulations and accounts for variables such as carbohydrate consumption, exercise, and nighttime risk of hypoglycemia.

Technology Overview
Managing glucose levels is challenging for individuals with type 1 diabetes (T1D) since multiple factors can impact an individual’s insulin sensitivity. Current decision support tools lack personalized recommendations and fail to account for factors like exercise and meal consumption that can alter insulin responsiveness.
The current technology is an algorithm that provides personalized decision support for individuals managing T1D using a “digital twin” trained on the individual’s own glucose and insulin data. This tool has the ability to replay different insulin settings on retrospective data to provide personalized recommendations to changes in insulin therapy.  Features of the digital twin technology for T1D include:

• Determination of optimal amounts of insulin that should be dosed for meals.
• Recommendations for reduced insulin or carbohydrate consumption prior to exercise to avoid dangerous hypoglycemia.
• Anticipation of overnight low glucose levels and recommendations to avoid hypoglycemia and enhance sleep safety.
• Interpretable and explainable recommendations that can be visualized to enhance user understanding and compliance.
• Utility to improve insulin dose recommendations for patients using multiple daily insulin injections.
• Additional utility for patients using automated insulin pumps through recommendation of insulin infusion rates for different times of day.

Publication
Young G, et al. Design and In Silico Evaluation of an Exercise Decision Support System Using Digital Twin Models. J Diabetes Sci Technol. 2024 Mar;18(2):324-334. doi: 10.1177/19322968231223217

Licensing Opportunity
This technology is available for licensing.