Periodic-Zone Model Predictive Control for Artificial Pancreas Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current artificial pancreas systems face challenges in maintaining safe blood glucose levels during sleep due to the risk of nocturnal hypoglycemia, as they are not designed to adapt to the time-dependent nature of human physiology and insulin dynamics, leading to potential life-threatening events.
Innovation Solution
A Periodic-Zone Model Predictive Control (PZMPC) strategy is implemented, which adjusts blood glucose target zones and insulin input constraints based on the time of day, maintaining higher glucose levels at night and reducing insulin delivery to minimize the risk of hypoglycemia during sleep.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed glycemic target zone is used throughout the day, then the controller can maintain consistent control parameters, but the risk of nocturnal hypoglycemia increases because the controller cannot adapt to nighttime physiological changes
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed glycemic target zone to a time-varying target zone that changes throughout the day. The controller dynamically adjusts the target glucose levels based on the time of day, with higher targets during nighttime hours to prevent hypoglycemia. This dynamic adaptation allows the system to respond to circadian rhythms and physiological changes without requiring complex real-time sensing of metabolic state.
Solution Approach 2:
The patent implements periodic action through daily oscillating target zones that follow a circadian pattern. The glycemic target varies periodically throughout the day, with predetermined higher levels during nighttime and lower levels during daytime. This periodic modulation aligns with natural human physiology and eliminates the need for continuous complex calculations while maintaining safety.
2Reliability
If the blood glucose target zone is raised during the night, then the risk of hypoglycemia is reduced, but the manufacturing precision of glucose control deteriorates because the target deviates from the ideal euglycemic range
Solution Approach 1:
The patent applies beforehand cushioning by preemptively raising the glycemic target zone during nighttime before hypoglycemic events can occur. This prophylactic adjustment creates a safety buffer that accounts for the reduced awareness and response capability during sleep. The higher nighttime target acts as a cushion against the natural tendency toward hypoglycemia during rest periods.
Solution Approach 2:
The patent changes the target glucose parameter based on time of day. Instead of maintaining a fixed euglycemic target of 80-140 mg/dL, the system implements time-dependent target parameters that are elevated during nighttime hours. This parameter modification accepts temporary deviation from ideal glucose levels in exchange for preventing life-threatening hypoglycemic events.
3Reliability
If the bound on maximum insulin infusion rate is reduced at night, then controller-induced hypoglycemia is prevented, but the effectiveness of insulin delivery deteriorates because the controller cannot respond aggressively to hyperglycemic events
Solution Approach 1:
The patent changes the maximum insulin infusion rate parameter based on time of day. During nighttime, the maximum infusion rate is reduced to prevent aggressive insulin delivery that could cause hypoglycemia. During daytime, the full infusion capability is restored to handle mealtime hyperglycemia effectively. This temporal parameter modulation balances safety and effectiveness.
Data Source
AI summary
A controller for an artificial pancreas for automated insulin delivery to patients with type 1 diabetes mellitus (T1DM) that enforces safe insulin delivery throughout both day and night, wherein the controller employs zone model predictive control, whereby real-time optimization, based on a model of a human's insulin response, is utilized to regulate blood glucose levels to a safe zone, and time-dependent zones that smoothly modulate the controller correction based on the time of day, wherein the controller strategically strives to maintain an 80-140 mg/dL glucose zone during the day, a 110-220 mg/dL zone at night, and a smooth transition of 2 hour duration in between.