Stress Hyperglycemia Simulation Using Segmented Virtual Patient Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current mathematical models for glucose-insulin physiology in ICU settings are inadequate for simulating stress hyperglycemia, as they are not based on the broader principles of stress physiology and require empirical adaptation for each patient, limiting the number and variability of virtual ICU patients that can be created for simulation.

Innovation Solution

A method is developed to derive time-varying hyperglycemic stresses from real ICU patients and apply them to non-critically ill virtual patients using models of normal glucose-insulin physiology, expanding the range of models and increasing the variability of in silico ICU patients, enabling sensitivity analysis and improved insulin infusion therapy protocols.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing models of normal glucose-insulin physiology are empirically adapted for each ICU patient, then a single virtual ICU patient can be created, but the number and variability of in silico ICU patients is severely limited

Engineering Contradiction:

Improveaccuracy of virtual patient representationVSAvoidnumber of virtual patients that can be created

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The model is segmented into two distinct components: (1) a universal stress hyperglycemia model that captures the pathophysiological mechanisms common to all ICU patients, and (2) patient-specific parameters that can be individually configured. This segmentation allows the same validated model structure to be reused across multiple virtual patients with different characteristics, thereby increasing the number of virtual patients that can be created without sacrificing accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The stress hyperglycemia model is designed as a universal framework that can be applied to any ICU patient population. By formulating a generalizable model that captures the essential pathophysiology of stress hyperglycemia, the system enables creation of multiple virtual patients from a single model structure, eliminating the need to empirically adapt a complete model for each individual patient while maintaining physiological accuracy.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Measurement precision

If existing models are modified to fit real ICU patients on a patient-by-patient basis, then accuracy for that specific patient is improved, but the complexity and difficulty of model creation increases

Engineering Contradiction:

Improveaccuracy of glucose-insulin modelingVSAvoidcomplexity of model adaptation process

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The modeling process is divided into two stages: first, a universal stress hyperglycemia model is developed and validated once; second, patient-specific parameters are configured using standardized procedures. This segmentation reduces the complexity of model creation by eliminating the need to perform complete empirical adaptations for each patient, while still maintaining accuracy through the use of patient-specific parameter values.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The stress hyperglycemia model is pre-developed and validated in advance as a universal framework. This preliminary action establishes a ready-to-use model structure that can be quickly instantiated for different virtual patients by simply configuring patient-specific parameters, rather than performing time-consuming empirical adaptations each time a new virtual patient is needed.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If empirical adaptation is performed for each patient, then the model fits that specific patient, but the ease of creation of multiple in silico patients is severely limited

Engineering Contradiction:

Improvemodel fit to patient dataVSAvoidease of creation of virtual patients

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

A universal stress hyperglycemia model is created that can serve multiple virtual patients simultaneously. This universal model maintains reliability by accurately representing the pathophysiology of stress hyperglycemia, while also greatly improving ease of creation since the same validated model can be instantiated for numerous virtual patients by simply varying patient-specific parameters rather than performing empirical adaptation each time.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

Instead of changing the model structure through empirical adaptation for each patient, the system maintains a fixed, validated stress hyperglycemia model and varies only the patient-specific parameters. This approach preserves model reliability while dramatically simplifying the creation process, as parameter configuration is much easier than structural model adaptation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10438700B2Computer simulation for testing and monitoring of treatment strategies for stress hyperglycemia

Publication Date: 2019.10.08 UNIV OF VIRGINIA PATENT FOUND

AI summary

Time-varying hyperglycemic stresses are derived from actual ICU patients and applied to non-critically ill virtual patients, using any model of normal glucose-insulin physiology that fulfills certain requirements, in order to model and simulate stress hyperglycemia. Other aspects provide: 1) a methodology to perform sensitivity analyses of the parameters of ICU insulin infusion therapy protocols and to improve the protocols; and 2) a training system for clinicians about the course and management of stress hyperglycemia in the ICU or other facility.