Automated Insulin Delivery During Missing or Inaccurate CGM Readings

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Ambulatory insulin pumps that rely on continuous glucose monitoring (CGM) data for automated insulin delivery often terminate therapy when CGM readings are inaccurate or unavailable, leading to potential safety issues and disruptions in treatment.

Innovation Solution

The system incorporates improved algorithms that enable continuous insulin delivery by employing noise filters, threshold comparisons, and physiological models to estimate glucose values, allowing the system to continue delivering insulin safely during periods of inaccurate or missing CGM data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the system terminates automated insulin delivery when CGM readings are inaccurate or unavailable, then patient safety is maintained by avoiding incorrect dosing decisions, but treatment continuity is disrupted and potential safety issues arise from therapy interruption

Engineering Contradiction:

Improvepatient safetyVSAvoidtreatment continuity

Core Design Contradiction:

ReliabilityVSDuration of action of stationary object

Solution Approach 1:

The system performs preliminary actions by implementing predictive algorithms that forecast future glucose levels based on historical data trends before actual hypoglycemic events occur. This allows the system to proactively adjust insulin delivery to prevent low glucose events, maintaining treatment continuity while ensuring safety during CGM data gaps.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system employs disposable fallback strategies using simple rule-based algorithms that can operate independently when complex predictive models cannot function due to insufficient or inaccurate CGM data. These simplified algorithms serve as temporary replacements, maintaining basic treatment continuity during periods when reliable predictive modeling is not possible.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Productivity

If the system uses CGM readings to calculate automated insulin doses, then insulin delivery is optimized based on real-time glucose data, but the system becomes vulnerable to incorrect dosing when CGM data is inaccurate

Engineering Contradiction:

Improveinsulin delivery optimizationVSAvoiddosing accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system implements feedback mechanisms that continuously monitor the accuracy and reliability of CGM readings. When glucose values fall outside expected physiological ranges or show unrealistic patterns, the system detects these anomalies and adjusts its dosing calculations accordingly, either by weighting the data differently, using alternative algorithms, or temporarily suspending automated dosing until reliable data is restored.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system applies partial action by using only the reliable portion of CGM data for dosing calculations while excluding suspicious or inaccurate readings. Alternatively, it may apply excessive action by implementing multiple layers of validation and cross-checking glucose data against physiological norms, historical patterns, and pump delivery records to ensure dosing accuracy even when some data points are questionable.

Inventive Principle:

Principle #16Partial or excessive action

3Reliability

If the system stops automated delivery when CGM readings are not received, then dosing decisions are not made based on potentially erroneous data, but the system lacks the capability to handle temporary signal loss or communication failures

Engineering Contradiction:

Improvedosing decision accuracyVSAvoidsignal loss handling

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system prepares cushioning measures in advance by implementing fallback algorithms and alternative data sources that can be activated when CGM signal is lost. These pre-prepared contingency plans include using historical glucose patterns, interstitial fluid glucose estimates from alternative sensors, or rule-based dosing algorithms that can operate with limited data, ensuring the system can handle temporary communication failures without compromising patient safety or treatment continuity.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS12564677B2Systems and methods for automated insulin delivery response to inaccurate or missed glucose values

Publication Date: 2026.03.03 TANDEM DIABETES CARE INC

AI summary

Disclosed herein are systems and methods incorporating an ambulatory infusion pump and a CGM. These systems that can include software and related methods to provide improved automated insulin delivery algorithms that enable the algorithms to safely continue delivering insulin for some time periods of missing or known inaccurate glucose values.