Dynamic Patient-Specific Insulin Therapy Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current clinical approaches for managing diabetes, particularly in diabetic patients, fail to account for patient-specific factors such as physiological variability, metabolic differences, and the effects of stress, exercise, and meals, leading to suboptimal insulin delivery and glucose control.

Innovation Solution

A computerized system for developing patient-specific therapies using dynamic modeling of patient physiology, which includes physiological and metabolic models, and mathematical analysis to determine insulin delivery and dosage, integrating data from glucose measurements, patient activities, and other physiological parameters to provide personalized insulin therapy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional therapy approaches are used for glucose control, then basic glucose monitoring is provided, but patient-specific factors such as physiological variability, metabolic differences, and effects of stress, exercise, sickness, and meals are not accounted for

Engineering Contradiction:
Improveaccounting for patient-specific factorsVSAvoidcomplexity of therapy approach
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system employs dynamic modeling of patient physiology that continuously adapts to changing conditions. The model updates insulin sensitivity, carbohydrate-to-insulin ratio, and other parameters in response to real-time glucose measurements, meal intake, physical activity, and stress indicators, allowing the therapy to dynamically adjust to patient-specific factors rather than using static protocols

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes key therapeutic parameters including insulin delivery rate, timing, and dosage based on patient-specific physiological states. By monitoring multiple parameters (glucose levels, meal composition, activity level, stress markers) and adjusting insulin parameters accordingly, the system adapts therapy to individual patient needs while managing complexity through automated parameter optimization

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If a series of measurements is used to understand system dynamics, then more comprehensive data is obtained, but it becomes difficult to translate data into actionable information

Engineering Contradiction:
Improvecompleteness of physiological dataVSAvoiddifficulty of translating data to actionable information
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system continuously monitors glucose measurements and other physiological parameters, compares them against the dynamic model predictions, and uses this feedback to adjust insulin delivery recommendations. This closed-loop feedback mechanism transforms raw measurement data into actionable therapy adjustments by automatically identifying deviations from expected physiological behavior and generating appropriate responses

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The dynamic physiological model acts as an intermediary between raw measurement data and therapeutic decisions. The model processes multiple measurements (glucose, meals, activity, stress) and translates them into meaningful predictions about patient physiology, making the complex data actionable by providing interpretable recommendations for insulin delivery

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of manufacture

If population-based approaches are used for drug dosing, then general guidelines are established, but patient-specific variability in pharmacokinetics and pharmacodynamics is not captured

Engineering Contradiction:
Improvesimplicity of dosing guidelineVSAvoidprecision of patient-specific insulin response
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system performs preliminary characterization of patient-specific physiological parameters during an initial period when the patient uses the system. By measuring glucose responses to meals, exercise, and stress during this learning phase, the system builds a personalized dynamic model that captures individual variability before providing long-term therapy recommendations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system enables patients to self-monitor glucose levels, report meal intake, activity, and stress conditions, and automatically generates personalized therapy recommendations. This self-service approach allows the system to capture patient-specific data without requiring manual input from healthcare providers, improving measurement precision while maintaining ease of use

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2562664B1System for determining an insulin delivery and communicating a dose in automated pancreas software
Publication Date: 2020.11.25 ROCHE DIABETES CARE GMBH
  • EP2562664B1 patent drawingFigure 1
  • EP2562664B1 patent drawingFigure 2
  • EP2562664B1 patent drawingFigure 3

AI summary

A system for developing patient-specific therapies based on dynamic modeling of patient-specific physiology and method thereof are disclosed. The system includes software modules configured to provide access via a computer to one or more data collection protocols defining at least a type of patient-specific data to be collected and a manner in which the patient-specific data is to be collected, and to information from which one or more patient-specific models, configured to simulate one or more aspects of the patient's physiology, is developed. Another software module of the system is configured to provide access via the computer to one or more software tools that apply patient-specific data, collected according to the one or more data collection protocols, to the one or more patient specific models to determine therefrom one or more patient-specific therapies.