MMPPC: A Probabilistic Closed-loop Artificial Pancreas to Handle Unannounced Meals and Exercise — MMPPC  

Type 1 Diabetes Clinical Trial

Official title: MMPPC: A Probabilistic Closed-loop Artificial Pancreas to Handle Unannounced

Status Recruiting

Clinical Trial Summary

To determine the safety, feasibility and efficacy of the Multiple Model Probabilistic Predictive Control / Diabetes Assistant (MMPPC/DiAS) system in full day and night closed-loop control in adolescents and adults with type 1 diabetes. This will be addressed in two parts. There will be an initial 36 hour inpatient study with hourly plasma glucose monitoring and scheduled exercise and meals. If the system is shown to be safe, feasible and effective, the study will proceed to supervised hotel studies with remote monitoring to test system use over 3 days/2 nights.

Clinical Trial Description

To further develop and refine an artificial pancreas system using the MMPPC framework that the investigators have developed. The MMPPC controller has been tested in a number of in silico studies as well as a small cohort of clinical studies. The controller is unique compared to other closed-loop systems in that it does not require meal announcement or a priming dose of insulin prior to a meal and is therefore a fully closed-loop system. Initial results were encouraging and the investigators achieved acceptable glucose control during the day and night.

With this protocol, the investigators will further test the algorithm, as well how well the algorithm operates within an automated, portable closed-loop device, the DIAS in clinical studies. The DIAS system is able to use glucose sensor input and communicate to an insulin pump. The algorithm is incorporated within an Android phone, which determines insulin delivery. The studies will involve an initial inpatient phase, with frequent plasma glucose monitoring and proceed to supervised, remotely-monitored studies in a hotel setting. ;

Study Design

Allocation: Non-Randomized, Endpoint Classification: Safety/Efficacy Study, Intervention Model: Parallel Assignment, Masking: Open Label, Primary Purpose: Treatment

Related Conditions & MeSH terms

• Type 1 Diabetes

• NCT number: NCT02480309
• Study type: Interventional
• Source: University of Colorado, Denver
• Contact: David Maahs, MD
• Phone: 303-724-6706
• Email: maahs.david@ucdenver.edu
• Status: Recruiting
• Phase: N/A
• Start date: June 2015
• Completion date: March 2017