Neural Network Analyte Prediction Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting analyte levels, such as glucose in blood, are inadequate for continuous monitoring and managing diabetes, as they fail to provide accurate and timely predictions, leading to risks of hypoglycemia and hyperglycemia.
Innovation Solution
A method and system utilizing a trained neural network with gradually connected layers to predict analyte levels by receiving time-ordered series of monitored levels and dose levels, allowing for interpolation and updating of data to provide accurate future predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction methods are used for analyte levels, then the system is simpler to implement, but the prediction accuracy and reliability are insufficient
Solution Approach 1:
The patent replaces traditional mechanical/mathematical prediction methods with an artificial neural network system. The neural network procedure processes time-ordered series of analyte levels and drug dosage levels to generate predictions, substituting conventional algorithms with a learned model that adapts to individual patient patterns, thereby improving accuracy while managing complexity through automated training and deployment.
Solution Approach 2:
The patent introduces a neural network procedure as an intermediary between raw monitoring data and prediction outputs. This intermediary layer processes and interprets complex temporal patterns in analyte levels and drug dosages, transforming raw data into clinically actionable predictions while isolating the complexity of the prediction algorithm from the user interface and clinical decision-making process.
2Measurement precision
If monitoring frequency is increased to improve prediction accuracy, then more data is available for analysis, but the loss of time and energy increases
Solution Approach 1:
The patent performs preliminary actions by continuously collecting and storing time-ordered series of analyte levels and drug dosage levels in advance of prediction needs. The system maintains historical data repositories that are pre-processed and ready for neural network analysis, eliminating the need for real-time data collection during critical prediction moments and reducing time loss during urgent clinical decisions.
Solution Approach 2:
The patent implements continuous monitoring and continuous prediction capabilities, where the neural network procedure operates continuously on incoming data streams. This continuous action ensures that predictions are always available and up-to-date without requiring intermittent batch processing, thereby minimizing time loss while maintaining high measurement precision through constant data flow and analysis.
Data Source
AI summary
A method of predicting an analyte level comprises receiving a time-ordered series of levels of the analyte, monitored over a time-period; feeding a trained neural network procedure with the monitored levels; and displaying, based on an output received from the procedure, a predicted level of the analyte in a future time. The procedure can comprise a plurality of layers, wherein for at least one pair of layers, a number of inter-layer connections within the pair is higher for later monitored levels than for earlier monitored levels.