Hypoglycemia Prediction Algorithm for Exercise Risk

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Patients with type 1 diabetes face challenges in predicting and managing exercise-induced hypoglycemia, which is a significant barrier to regular physical activity due to the risk of hypoglycemic events.

Innovation Solution

A system and method that utilize a hypoglycemia prediction algorithm and a classifier algorithm to determine a hypoglycemia risk signal and state, respectively, based on blood glucose signals, insulin on board ratios, and exercise parameters, enabling actionable alerts and adjustments to mitigate hypoglycemic risk.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If patients with type 1 diabetes engage in regular physical activity, then quality of life and cardiovascular health improve, but the risk of hypoglycemia increases

Engineering Contradiction:

Improvecardiovascular healthVSAvoidhypoglycemia risk

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary assessment of hypoglycemia risk before exercise begins by analyzing baseline blood glucose levels, insulin on board, and exercise parameters. This allows preventive actions to be taken before the harmful effect occurs

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors blood glucose levels during and after exercise, providing real-time feedback to detect trends toward hypoglycemia. This enables dynamic adjustment of insulin delivery or carbohydrate recommendations to counteract the harmful effect while maintaining exercise benefits

Inventive Principle:

Principle #23Feedback

2Object-affected harmful factors

If patients reduce insulin doses or increase carbohydrate intake to prevent hypoglycemia, then hypoglycemia risk decreases, but glycemic control deteriorates

Engineering Contradiction:

Improvehypoglycemia riskVSAvoidglycemic control

Core Design Contradiction:

Object-affected harmful factorsVSReliability

Solution Approach 1:

The system dynamically adjusts insulin delivery parameters or carbohydrate recommendations based on real-time analysis of blood glucose trends, exercise intensity, and insulin on board. This allows precise modification of treatment parameters to prevent hypoglycemia without compromising overall glycemic control

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system applies partial adjustments to insulin or carbohydrate intake based on the calculated hypoglycemia risk level, rather than blanket reductions or increases. This ensures glycemic control is maintained while providing sufficient protection against hypoglycemia

Inventive Principle:

Principle #16Partial or excessive action

3Object-affected harmful factors

If a predictive hypoglycemia model is implemented, then hypoglycemia risk is reduced, but system complexity increases

Engineering Contradiction:

Improvehypoglycemia riskVSAvoidprediction system complexity

Core Design Contradiction:

Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The predictive model is segmented into distinct functional modules: data acquisition from multiple sources, exercise parameter input, hypoglycemia risk calculation using specific formulas, and actionable recommendation generation. This modular structure reduces overall system complexity while maintaining predictive accuracy

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12336847B2Method, system and computer readable medium for predictive hypoglycemia detection for mild to moderate exercise

Publication Date: 2025.06.24 UNIV OF VIRGINIA PATENT FOUND

AI summary

A system for generating a hypoglycemia risk signal associated with exercise-induced Hypoglycemia. The system can include a processor configured to obtain a blood glucose signal (BGstart), a ratio of absolute insulin on board over total daily insulin signal (IOBabs/TDI), and an initial glycemic slope signal (S0); generate a hypoglycemia risk signal based on a hypoglycemia prediction algorithm that determines the probability of a user being hypoglycemic during or after exercise based on the obtained BGstart, IOBabs/TDI and S0.