Closed-Loop Insulin Infusion Control with PID and IOB Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current insulin infusion systems are limited by the speed of insulin delivery and lack efficient control mechanisms to mimic the natural insulin response of a healthy pancreas, leading to suboptimal glucose regulation and potential complications such as excessive weight gain and atherosclerosis.
Innovation Solution
A processor-implemented method and electronic device that estimate the current insulin on board (IOB) value, calculate an IOB rate, and determine an adjusted insulin infusion rate to regulate insulin delivery, using a PID controller to mimic the natural insulin secretion pattern of a healthy pancreas, ensuring efficient glucose homeostasis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional external infusion pumps are used to deliver insulin, then insulin delivery control is improved, but the system complexity and lack of efficient control mechanisms worsen
Solution Approach 1:
The patent implements a closed-loop control system that continuously monitors glucose levels and adjusts insulin delivery based on real-time feedback. The controller receives glucose level data from sensors, processes this information, and automatically modifies the insulin infusion rate to maintain glucose homeostasis, eliminating the need for complex manual control mechanisms while improving delivery reliability
Solution Approach 2:
The system enables self-regulating insulin delivery through automated control algorithms that mimic natural pancreatic beta-cell function. The controller autonomously calculates and adjusts insulin infusion rates based on glucose level feedback, insulin-on-board compensation, and user profiles without requiring complex external intervention or manual adjustment by the user
2Speed
If fast acting insulin is used in pumps, then insulin delivery speed is improved, but the ability to mimic natural insulin response worsens
Solution Approach 1:
The patent employs dynamic control algorithms that continuously adjust insulin delivery parameters based on real-time glucose levels, rate of glucose change, and insulin-on-board calculations. The system adapts the insulin infusion rate dynamically to match natural pancreatic response patterns, using fast-acting insulin but modulating its delivery to achieve physiological-like glucose control rather than simple rapid delivery
Solution Approach 2:
The system changes multiple delivery parameters simultaneously including infusion rate, bolus size, insulin-on-board compensation factors, and prediction horizon based on current glucose state. By dynamically adjusting these parameters, the system uses fast-acting insulin while mimicking the nuanced temporal patterns of natural insulin secretion, achieving both speed and adaptability
3Reliability
If closed-loop control is implemented, then glucose regulation is improved, but the system complexity and potential complications worsen
Solution Approach 1:
The patent divides the control system into distinct functional modules including glucose level monitoring, insulin-on-board calculation, control algorithm processing, and infusion delivery control. This modular segmentation allows each component to perform its specific function with optimized complexity, reducing overall system complexity while maintaining effective closed-loop glucose regulation through coordinated module interaction
Data Source
AI summary
Processor-implemented methods of controlling an insulin infusion device for a user are provided here. A first method obtains and analyzes calibration factors (and corresponding timestamp data) for a continuous glucose sensor, and regulates entry into a closed-loop operating mode of the infusion device based on the calibration factors and timestamp data. A second method obtains a most recent sensor glucose value and a target glucose setpoint value for the user at the outset of the closed-loop mode. The second method adjusts the closed-loop insulin infusion rate over time, in response to the sensor glucose value and the setpoint value. A third method calculates an upper insulin limit that applies to the insulin infusion rate during the closed-loop mode. The insulin limit is calculated based on a fasting blood glucose value of the user, a total daily insulin value of the user, and fasting insulin delivery data for the user.