Adaptive Safety Constraints in Automated Insulin Delivery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional Automated Insulin Delivery (AID) systems face challenges with safety constraints that either limit insulin delivery, leading to hyperglycemia or result in frequent hypoglycemic events due to one-size-fits-all settings based on general population norms, failing to account for individual insulin sensitivity variations.

Innovation Solution

A medicament delivery system that dynamically adjusts safety constraints based on user-specific data, including frequency of hypoglycemic and hyperglycemic events, to tailor insulin delivery and reduce the risk of severe glycemic events by tightening or relaxing constraints as needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If tight safety constraints are employed to prevent hypoglycemia, then hypoglycemic events are reduced, but hyperglycemia increases due to limited insulin delivery and reduced adaptability

Engineering Contradiction:
ImprovesafetyVSAvoidinsulin delivery effectiveness
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements dynamic safety constraints that automatically adjust based on real-time system performance. The control system monitors hypoglycemic and hyperglycemic events and modifies constraint parameters accordingly, transitioning from static to dynamic constraint management. This allows the system to tighten constraints when hypoglycemia risk is high and relax them when safe, optimizing both safety and insulin delivery effectiveness.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent employs feedback mechanisms where system performance data (frequency and severity of glycemic events) is continuously monitored and used to adjust safety constraints. The control system analyzes historical data and modifies constraint parameters based on observed outcomes, creating a closed-loop system that adapts to individual user needs and improves both safety and effectiveness over time.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If conventional safety constraints based on general population norms are used, then a standardized safety level is maintained, but individual user needs are not met resulting in either hypoglycemia or hyperglycemia

Engineering Contradiction:
Improveindividual customizationVSAvoidsafety
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies local quality by customizing safety constraints to match individual user characteristics and response patterns. Instead of applying uniform population-based constraints, the system adapts constraint parameters to each user's specific insulin sensitivity, glucose response, and safety preferences. This localized customization ensures optimal safety and effectiveness for each individual while maintaining overall system reliability.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS20260020817A1Dynamic and customized adjustment of safety constraints for a medicament delivery system
Publication Date: 2026.01.22 INSULET CORP
  • US20260020817A1 patent drawing
  • US20260020817A1 patent drawing
  • US20260020817A1 patent drawing

AI summary

Exemplary embodiments may customize the safety constraints of a medicament delivery system for a user. The customization may occur on an ongoing basis so as to dynamically adapt to the changing medicament needs of the user and thus provide better therapeutic results for the user. The safety constraints may be adjusted at periodic intervals, such as daily, hourly, or even per operational cycle of the control system of the medicament delivery system. The adjustments may be based on recent performance of the control system in providing desired therapeutic results. The safety constraint(s) that are adjusted may be those that were invoked more than the other safety constraints in a recent interval.