Neural Network ESR Correction via Aggregation Curve
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ESR measurement methods, such as the Westergren method, are slow and require large blood volumes, making them inefficient for rapid testing, and the erythrocyte aggregation method lacks accuracy in reflecting natural sedimentation rates, leading to inconsistencies with Westergren method results.
Innovation Solution
An apparatus and system that uses an ESR detection device to obtain an erythrocyte aggregation curve and a blood cell analyzer to acquire a histogram or scattergram, which are then input into a neural network model to calculate an accurate ESR measurement result, ensuring consistency with Westergren method results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the Westergren method is used to measure ESR, then measurement accuracy is improved, but measurement time increases to one hour
Solution Approach 1:
The patent applies preliminary action by performing multiple measurements at different time points (0.5h, 1h, 1.5h, 2h, 3h) and using the sedimentation distance at 0.5h combined with a correction factor to predict the final ESR value. This allows the measurement process to be completed in half the traditional time while maintaining accuracy through computational correction.
2Measurement precision
If the Westergren method is used to measure ESR, then measurement accuracy is improved, but blood volume requirement increases to about 1 mL
Solution Approach 1:
The patent uses preliminary measurements at 0.5h combined with a correction factor algorithm to predict the final ESR value, allowing accurate results with reduced blood volume (100-200 μL) instead of requiring the full 1 mL needed for traditional Westergren method completion.
3Productivity
If the erythrocyte aggregation method is used to measure ESR, then measurement speed is improved to about 20 s, but measurement accuracy deteriorates with significant discrepancy from Westergren method results
Solution Approach 1:
The patent applies feedback by using the sedimentation distance measured at 0.5h as input to a correction algorithm that calculates the final ESR value. This feedback mechanism allows the system to use early measurement data to predict the final result, achieving both rapid measurement and high accuracy.
Solution Approach 2:
The patent replaces the traditional mechanical waiting period for complete sedimentation with a computational correction system. Instead of mechanically allowing 1 hour for sedimentation, the system uses optical detection at 0.5h combined with algorithmic correction to achieve the same accuracy, substituting computational processing for temporal waiting.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system enables rapid and accurate ESR measurement with improved consistency compared to the Westergren method, using a neural network model to process data from the erythrocyte aggregation curve and blood cell analysis for precise results.
Implementation Method 1
measuring a change in a scattering rate/transmissivity of blood cells to light during the formation of rouleaux red blood cells
Implementation Method 2
obtain an erythrocyte aggregation curve of light intensity transmitted through the blood sample to be tested or of light scattered by the blood sample to be tested as a function of time
Implementation Method 3
Erythrocyte sedimentation rate (ESR for short) refers to a rate at which red blood cells in anticoagulated blood in vitro naturally sediment under specified conditions
Implementation Method 4
red blood cells in anticoagulated blood in vitro naturally sediment under specified conditions
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Embodiments of the disclosure relate to a system, an apparatus and a method for measuring erythrocyte sedimentation rate (ESR), and a computer-readable storage medium. The method includes: acquiring an erythrocyte aggregation curve of a blood sample to be tested; acquiring a blood cell histogram and/or a blood cell scattergram of the blood sample to be tested, the blood cell histogram and/or the blood cell scattergram including at least a histogram and/or a scattergram related to red blood cells; and inputting the erythrocyte aggregation curve and the blood cell histogram and/or the blood cell scattergram into a neural network model, to calculate a first ESR measurement result by using the neural network model. Therefore, the erythrocyte aggregation curve obtained by using an erythrocyte aggregation method is corrected by means of the neural network model by using the blood cell histogram and/or the blood cell scattergram obtained by a hematology analyzer, so that a more accurate ESR can be obtained rapidly.