New Clamp-PID Algorithm for Automated Glucose Clamps Improves Clamp Quality.

Area of Science:

• Biomedical Engineering
• Control Systems
• Diabetes Technology

Background:

• Automated glucose clamp experiments maintain blood glucose (BG) levels near a target using algorithms to control glucose infusion rates (GIRs).
• Clamp quality, essential for accurate BG control, relies heavily on the effectiveness of these algorithms.
• The established Biostator (BS) algorithm is a benchmark for comparison in developing new clamp control strategies.

Purpose of the Study:

• To develop and evaluate a novel clamp algorithm based on the proportional-integral-derivative (PID) control approach.
• To compare the clamp quality of the new PID-based algorithm against the existing Biostator algorithm.
• To validate the performance of the PID algorithm through in silico, in vitro, and in vivo studies.

Main Methods:

• In silico optimization of the PID-based algorithm was performed using numerical simulations.
• The optimized Clamp-PID algorithm underwent in vitro experimental testing.
• In vivo validation was conducted in a small clinical study with 5 participants.

Main Results:

• The Clamp-PID algorithm demonstrated superior clamp quality compared to the BS algorithm across all tested environments (in silico, in vitro, in vivo).
• Numerical simulations showed a significant reduction in precision and absolute control deviation (ACD) with Clamp-PID.
• In vitro and clinical studies confirmed these improvements, with notable decreases in precision and ACD for the Clamp-PID algorithm.

Conclusions:

• The novel Clamp-PID algorithm offers improved precision and reduced absolute control deviation (ACD) in automated glucose clamp experiments.
• The findings suggest that the Clamp-PID algorithm represents a significant advancement over the Biostator algorithm for glucose control.
• This improved performance has implications for enhancing the accuracy and reliability of automated glycemic management systems.