Jordan E Pinsker1, Joon Bok Lee2, Eyal Dassau3, Dale E Seborg2, Paige K Bradley1, Ravi Gondhalekar2, Wendy C Bevier1, Lauren Huyett2, Howard C Zisser2, Francis J Doyle4. 1. William Sansum Diabetes Center, Santa Barbara, CA. 2. William Sansum Diabetes Center, Santa Barbara, CA Department of Chemical Engineering, University of California, Santa Barbara, Santa Barbara, CA. 3. William Sansum Diabetes Center, Santa Barbara, CA Department of Chemical Engineering, University of California, Santa Barbara, Santa Barbara, CA John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA. 4. William Sansum Diabetes Center, Santa Barbara, CA Department of Chemical Engineering, University of California, Santa Barbara, Santa Barbara, CA John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA frank_doyle@seas.harvard.edu.
Abstract
OBJECTIVE: To evaluate two widely used control algorithms for an artificial pancreas (AP) under nonideal but comparable clinical conditions. RESEARCH DESIGN AND METHODS: After a pilot safety and feasibility study (n = 10), closed-loop control (CLC) was evaluated in a randomized, crossover trial of 20 additional adults with type 1 diabetes. Personalized model predictive control (MPC) and proportional integral derivative (PID) algorithms were compared in supervised 27.5-h CLC sessions. Challenges included overnight control after a 65-g dinner, response to a 50-g breakfast, and response to an unannounced 65-g lunch. Boluses of announced dinner and breakfast meals were given at mealtime. The primary outcome was time in glucose range 70-180 mg/dL. RESULTS:Mean time in range 70-180 mg/dL was greater for MPC than for PID (74.4 vs. 63.7%, P = 0.020). Mean glucose was also lower for MPC than PID during the entire trial duration (138 vs. 160 mg/dL, P = 0.012) and 5 h after the unannounced 65-g meal (181 vs. 220 mg/dL, P = 0.019). There was no significant difference in time with glucose <70 mg/dL throughout the trial period. CONCLUSIONS: This first comprehensive study to compare MPC and PID control for the AP indicates that MPC performed particularly well, achieving nearly 75% time in the target range, including the unannounced meal. Although both forms of CLC provided safe and effective glucose management, MPC performed as well or better than PID in all metrics.
RCT Entities:
OBJECTIVE: To evaluate two widely used control algorithms for an artificial pancreas (AP) under nonideal but comparable clinical conditions. RESEARCH DESIGN AND METHODS: After a pilot safety and feasibility study (n = 10), closed-loop control (CLC) was evaluated in a randomized, crossover trial of 20 additional adults with type 1 diabetes. Personalized model predictive control (MPC) and proportional integral derivative (PID) algorithms were compared in supervised 27.5-h CLC sessions. Challenges included overnight control after a 65-g dinner, response to a 50-g breakfast, and response to an unannounced 65-g lunch. Boluses of announced dinner and breakfast meals were given at mealtime. The primary outcome was time in glucose range 70-180 mg/dL. RESULTS: Mean time in range 70-180 mg/dL was greater for MPC than for PID (74.4 vs. 63.7%, P = 0.020). Mean glucose was also lower for MPC than PID during the entire trial duration (138 vs. 160 mg/dL, P = 0.012) and 5 h after the unannounced 65-g meal (181 vs. 220 mg/dL, P = 0.019). There was no significant difference in time with glucose <70 mg/dL throughout the trial period. CONCLUSIONS: This first comprehensive study to compare MPC and PID control for the AP indicates that MPC performed particularly well, achieving nearly 75% time in the target range, including the unannounced meal. Although both forms of CLC provided safe and effective glucose management, MPC performed as well or better than PID in all metrics.
Authors: Daniel A Finan; Thomas W McCann; Linda Mackowiak; Eyal Dassau; Stephen D Patek; Boris P Kovatchev; Francis J Doyle; Howard Zisser; Henry Anhalt; Ramakrishna Venugopalan Journal: J Diabetes Sci Technol Date: 2014-01-01
Authors: Lalo Magni; Davide M Raimondo; Luca Bossi; Chiara Dalla Man; Giuseppe De Nicolao; Boris Kovatchev; Claudio Cobelli Journal: J Diabetes Sci Technol Date: 2007-11
Authors: Rebecca A Harvey; Eyal Dassau; Wendy C Bevier; Dale E Seborg; Lois Jovanovič; Francis J Doyle; Howard C Zisser Journal: Diabetes Technol Ther Date: 2014-01-28 Impact factor: 6.118
Authors: Eyal Dassau; Howard Zisser; Rebecca A Harvey; Matthew W Percival; Benyamin Grosman; Wendy Bevier; Eran Atlas; Shahar Miller; Revital Nimri; Lois Jovanovic; Francis J Doyle Journal: Diabetes Care Date: 2012-11-27 Impact factor: 19.112
Authors: Trang T Ly; Marc D Breton; Patrick Keith-Hynes; Daniel De Salvo; Paula Clinton; Kari Benassi; Benton Mize; Daniel Chernavvsky; Jéróme Place; Darrell M Wilson; Boris P Kovatchev; Bruce A Buckingham Journal: Diabetes Care Date: 2014-05-30 Impact factor: 19.112
Authors: Jordan E Pinsker; Alejandro J Laguna Sanz; Joon Bok Lee; Mei Mei Church; Camille Andre; Laura E Lindsey; Francis J Doyle; Eyal Dassau Journal: Diabetes Technol Ther Date: 2018-07 Impact factor: 6.118
Authors: Gregory P Forlenza; Sunil Deshpande; Trang T Ly; Daniel P Howsmon; Faye Cameron; Nihat Baysal; Eric Mauritzen; Tatiana Marcal; Lindsey Towers; B Wayne Bequette; Lauren M Huyett; Jordan E Pinsker; Ravi Gondhalekar; Francis J Doyle; David M Maahs; Bruce A Buckingham; Eyal Dassau Journal: Diabetes Care Date: 2017-06-05 Impact factor: 19.112
Authors: Joon Bok Lee; Eyal Dassau; Ravi Gondhalekar; Dale E Seborg; Jordan E Pinsker; Francis J Doyle Journal: Ind Eng Chem Res Date: 2016-10-27 Impact factor: 3.720