Written by 10:30 am AI, Discussions

### Leveraging AI for Implementing Automated Insulin Delivery in Young Children

UVA’s PEDAP-AI trial now recruiting for AI-assisted insulin pump settings in kids with type 1 diabe…

A brief trial is seeking 30 children aged 2-5 whose families are interested in exploring automated insulin delivery. The investigation will assess whether adjustments to insulin pump settings using artificial intelligence are comparable in effectiveness to manual adjustments. Clinical Trials Identifier: NCT06017089

Trial Title: The Pediatric Artificial Pancreas Automated Initialization Trial (PEDAP-AI)

Diabetes Type: Children (ages 2-5) with type 1 diabetes

Trial Sponsor: Dr. Marc Breton at the University of Virginia

Research Focus of the Trial

The primary objective of this research is to evaluate the safety and efficacy of utilizing an AI-based assistant to regulate insulin pump settings for young children commencing automated insulin delivery (AID).

The trial aims to recruit 30 children diagnosed with type 1 diabetes who are currently undergoing insulin therapy through multiple daily injections. Participants will utilize Tandem’s t:slim X2 Control-IQ technology alongside a Dexcom G6 or G7 continuous glucose monitor (CGM) over an eight-week period.

The study will analyze parameters such as time spent within the target glucose range, time below the target range (indicating hypoglycemia), and time above the target range (indicating hyperglycemia).

Significance of the Trial

In recent times, there has been a growing acknowledgment of the benefits of AID systems in assisting children and caregivers in managing type 1 diabetes.

Research indicates that the advantages of AID systems extend to young children with type 1 diabetes as well. For example, a study conducted in 2023 revealed that the Control-IQ technology resulted in a 3-hour increase in time within the target range and a 0.5% reduction in A1C levels among children aged 2-6 with type 1 diabetes.

Similarly, another study demonstrated the safety and efficacy of Medtronic’s MiniMed 780G system in children aged 2-6, showcasing notable enhancements in time within the target range without any instances of diabetic ketoacidosis (DKA) or hypoglycemic events.

While Tandem’s Control-IQ is authorized for use in children aged 6 to 13 and Medtronic’s MiniMed 780G system is approved for ages 7 and above, neither system has received clearance for toddlers or very young children. Presently, the Omnipod 5 is the sole AID system approved for children aged 2 and older.

With the evolution of AI technology, researchers are actively exploring how AI can optimize insulin therapy. For instance, a recent study demonstrated that an AI chatbot facilitated the initiation of basal insulin in individuals with type 2 diabetes, resulting in improved glycemic control.

The initiation of AID typically involves numerous adjustments to insulin pump settings in the initial weeks. In this trial, the AI-based assistant may modify settings such as:

  • Basal rate: The continuous rate at which insulin is delivered in small, incremental doses.
  • Insulin-to-carb ratio: The ratio of rapid-acting insulin required per fixed amount of carbohydrates.
  • Insulin sensitivity factor: Also known as a correction factor, indicating the insulin quantity needed to normalize blood glucose levels.
  • Target blood glucose: Determined by individual treatment objectives.
  • Insulin duration: Referring to the duration of insulin activity post boluses.

By fine-tuning various pump settings, an AI assistant could potentially alleviate the burden on endocrinologists and simplify the process of transitioning to an AID system for children and their caregivers. A successful outcome in this preliminary study may pave the way for larger trials involving a broader group of children with type 1 diabetes.

Expression of Interest

Your child may qualify for participation if they meet the following criteria:

  • Aged between 2-5 years and diagnosed with type 1 diabetes for a minimum of three months
  • Weigh at least 20 pounds
  • Administer a minimum of five units of insulin daily
  • Utilize a Dexcom CGM
  • No history of diabetic ketoacidosis (DKA) in the past six months
  • No occurrences of hypoglycemia in the preceding three months

Children currently using an insulin pump are ineligible to participate.

Recruitment for the trial is presently active in California, Colorado, and Virginia. Detailed inclusion/exclusion criteria can be accessed here.


For further details: Reach out to Dr. Marc Breton at [email protected] or via phone at 434-982-6484.

About the authors

April Hopcroft joined diaTribe in 2023 as a Staff Writer following her tenure as a co-leader of the Diabetes Therapy team at Close Concerns. She completed her studies at Smith College in 2021, majoring in…
Read the full bio »

Visited 2 times, 1 visit(s) today
Tags: , Last modified: February 21, 2024
Close Search Window
Close