Neural Network Model for Physiological Parameter Estimation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing physiological parameter estimation systems face challenges in accurately distinguishing between correctly calculated values and erroneous measurements, particularly due to misanalysis by algorithms, leading to incorrect physiological parameter values.
Innovation Solution
A method and system that generate a trained model using a neural network, where first correctly calculated physiological parameter data is compared with second data from an algorithmically calculated parameter to create training labels, enabling the network to differentiate between correct and incorrect answers, thus improving estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If an algorithm automatically calculates physiological parameter values from waveform data, then productivity and automation are improved, but measurement precision deteriorates due to algorithmic misanalysis
Solution Approach 1:
The system uses a neural network that has been trained with labeled data (correctly calculated values vs. algorithmically calculated values) to provide feedback on the reliability of automatic calculations. The neural network estimates the probability that an automatically calculated physiological parameter is correct, allowing the system to identify and correct algorithmic misanalysis while maintaining automated processing.
Solution Approach 2:
The neural network acts as an intermediary between the automatic calculation algorithm and the final physiological parameter output. It evaluates the algorithm's results and provides a reliability assessment, serving as a mediating layer that filters out incorrect automatic calculations without eliminating the automated calculation process itself.
2Measurement precision
If a neural network is trained to distinguish correct from incorrect physiological parameter calculations, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The neural network is trained in advance using a dataset of labeled examples (correctly calculated physiological parameters and corresponding waveform data). This preliminary training phase allows the network to learn patterns of correct versus incorrect calculations before being deployed for actual physiological parameter estimation, so that during operation, the system can directly apply the trained model without complex real-time analysis.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A method for generating a trained model is applied to an estimation apparatus configured to estimate a probability that a value of a predetermined physiological parameter is correctly calculated based on waveform data acquired from a subject being tested. The method includes: acquiring first data corresponding to a value of a first physiological parameter that has been correctly calculated from first waveform data; inputting second waveform data to an algorithm automatically calculating a value of a second physiological parameter acquired from input waveform data and to output second data; generating third data including a training label indicating whether the value of the second physiological parameter corresponding to the second data is a correct answer or an incorrect answer by comparing the second data with the first data; and training a neural network by using the second waveform data and the third data, to generate a trained model.