Glycemic State Prediction Using Hidden Markov Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Patients with diabetes face challenges in maintaining stable glucose levels, leading to dangerous hypoglycemic and hyperglycemic states due to frequent insulin injections, which can result in loss of consciousness and death, and existing continuous glucose monitors do not effectively predict future glycemic states to prevent these conditions.
Innovation Solution
A method using continuous glucose monitors to measure glucose concentrations over time, applying multiple mathematical models to predict future glycemic states by deriving glucose prediction equations and employing a Hidden Markov Model to calculate probabilities of entering hypoglycemic, euglycemic, or hyperglycemic states, with a merit index to assess confidence in predictions and trigger alarms when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If frequent insulin injections are used to regulate glucose concentration, then glucose regulation effectiveness is improved, but the likelihood of entering hypoglycemic state increases
Solution Approach 1:
The system performs preliminary action by predicting future glycemic states before they occur. It analyzes historical glucose data and insulin administration patterns to forecast upcoming hypoglycemic or hyperglycemic events, enabling users to take preventive measures before dangerous glucose levels are reached, thus reducing hypoglycemic risk while maintaining effective glucose regulation
Solution Approach 2:
The system implements feedback by continuously monitoring glucose concentrations, comparing actual readings against predicted values, and providing alerts when deviations indicate approaching harmful states. This closed-loop feedback enables dynamic adjustment of insulin dosing to maintain glucose regulation effectiveness while preventing hypoglycemic events
2Productivity
If continuous glucose monitoring is used to frequently measure glucose concentration, then glucose regulation opportunity is improved, but the complexity of the system increases
Solution Approach 1:
The system applies self-service by automatically performing complex predictive analytics and pattern recognition tasks without requiring user intervention. The algorithm autonomously processes continuous glucose data, identifies trends, predicts future states, and generates alerts, thereby enabling frequent monitoring and regulation opportunities while keeping the user interface simple and the perceived system complexity low
Data Source
AI summary
A method for predicting a user's future glycemic state includes measuring a user's glucose concentration at intervals over a time duration, thereby generating a plurality of glucose concentrations as a function of time. First and second glucose prediction equations that are fits to the plurality of glucose concentrations based on first and second non-identical mathematical models, respectively, are then derived. The method also includes calculating first and second predicted glucose concentrations at a future time using the first and second glucose prediction equations, respectively. Thereafter, an average predicted glucose concentration and a merit index are calculated based on the first and second predicted glucose calculations. The plurality of glucose concentrations as a function of time, the merit index and average predicted glucose concentration are input into a trained model (for example, a Hidden Markov Model) that outputs a set of glucose concentration probabilities. The user's future glycemic state is then predicted based on the set of glucose concentration probabilities.