AI Model for Real-Time HbA1c Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Diabetic patients face challenges in managing their condition between hospital visits due to the difficulty in monitoring and predicting their hemoglobin A1c (HbA1c) levels, which is a crucial indicator of blood sugar control over several months, as they lack real-time access to this information and struggle with self-management without regular HbA1c testing.
Innovation Solution
A method and system that trains an artificial intelligence model using patient information, including exercise and bioinformation, to estimate HbA1c levels, allowing for real-time monitoring and management through a patient terminal, which includes a database unit and neural network modeling unit to generate an AI model for predicting HbA1c levels and providing personalized therapeutic interventions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HbA1c testing is performed only in hospitals every two to three months, then measurement accuracy is ensured, but real-time monitoring capability is lost
Solution Approach 1:
The patent creates a virtual copy of the HbA1c measurement function by training an AI model to predict HbA1c levels based on readily available data (blood glucose readings, exercise information, bioinformation). This AI model serves as a copy that replicates the predictive capability of hospital-based HbA1c testing, enabling real-time estimation without requiring actual hospital visits for measurement.
Solution Approach 2:
The patent introduces an AI model as an intermediary between blood glucose monitoring and HbA1c assessment. Instead of directly measuring HbA1c (which requires hospital equipment), the AI model acts as a mediator that translates common patient data into HbA1c level predictions, bridging the gap between home monitoring capabilities and clinical assessment needs.
2Ease of operation
If patients use patient notebooks or smartphone applications to manage blood sugar, then ease of operation is improved, but ability to monitor HbA1c levels is lost
Solution Approach 1:
The patent merges blood glucose monitoring functionality with HbA1c prediction functionality into a single integrated system. The AI model incorporates multiple data sources (blood glucose readings, exercise information, bioinformation) and combines them to provide both immediate glucose tracking and predictive HbA1c estimation, allowing patients to access both types of information through one application rather than separate tools.
Solution Approach 2:
The patent transforms a simple blood sugar management application into a multi-functional tool that not only tracks blood glucose levels but also predicts HbA1c levels, provides treatment recommendations, and monitors patient compliance. This universal approach allows the same system to perform multiple functions that previously required separate tools or hospital visits.
3Measurement precision
If AI model uses multiple data sources including exercise information and bioinformation, then prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The patent employs a universal AI model architecture that can process multiple types of input data (blood glucose readings, exercise information, bioinformation) through a single integrated system. This multi-functional model handles diverse data sources without requiring separate processing systems, thereby improving prediction accuracy while limiting the increase in overall system complexity.
Data Source
AI summary
A method of training an artificial intelligence model for estimating a hemoglobin A1c (HbA1c) level includes collecting patient information including exercise information and bioinformation of a patient, collecting an actual HbA1c level of the patient, converting the collected patient information and actual HbA1c level into a single standardized data structure format, and training an artificial intelligence model using the converted patient information and actual HbA1c level to generate an artificial intelligence model for estimating an HbA1c level. The bioinformation includes at least one of a blood sugar level, a blood pressure, a heart rate, and a menstrual cycle, and the exercise information is generated on the basis of patient life log data acquired by a patient terminal.