Spectroscopic Quantification With SC-SVM Sample Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing spectroscopic quantification methods suffer from false positive identifications due to unknown samples being misclassified or measured incorrectly, leading to inaccurate determinations of component concentrations.
Innovation Solution
Employing a single class support vector machine (SC-SVM) technique to verify that unknown samples belong to the class of materials the quantification model is configured to analyze, using confidence metrics and decision values to filter out false positives, and generating quantification models that exclude incorrect training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional spectroscopic quantification methods are used, then measurement speed and non-destructive analysis are improved, but false positive identifications and measurement accuracy deteriorate
Solution Approach 1:
The patent applies preliminary action by performing outlier detection and sample verification before final quantification. The system pre-processes spectral data using multiple validation techniques (standard deviation analysis, PCA scoring, PLS prediction comparison) to identify and exclude false positives before generating final concentration results, thereby maintaining measurement speed while improving accuracy
Solution Approach 2:
The patent introduces intermediary verification steps between spectral acquisition and final quantification. Multiple intermediate checks including outlier detection algorithms, confidence metric calculations, and cross-validation procedures act as mediators to filter unreliable measurements, resolving the contradiction between fast measurement and accurate results
2Productivity
If spectroscopic analysis is performed on unknown samples without verification, then productivity is improved, but reliability deteriorates due to false positive identifications
Solution Approach 1:
The system applies self-service by enabling automatic verification and self-correction of spectral measurements. The quantification system autonomously performs outlier detection, confidence assessment, and invalid result identification without requiring manual intervention, maintaining high productivity while ensuring reliability through built-in self-validation mechanisms
Solution Approach 2:
The patent implements feedback mechanisms where quantification results are continuously validated against reference data and confidence thresholds. When measurements fall outside expected parameters, the system automatically flags them for review or rejection, creating a feedback loop that maintains reliability without significantly impacting productivity
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A device may receive information identifying results of a spectroscopic measurement performed on an unknown sample. The device may determine a decision boundary for a quantification model based on a configurable parameter, such that a first plurality of training set samples of the quantification model is within the decision boundary and a second plurality of training set samples of the quantification model is not within the decision boundary. The device may determine a distance metric for the spectroscopic measurement performed on the unknown sample relative to the decision boundary. The device may determine a plurality of distance metrics for the second plurality of training set samples of the quantification model relative to the decision boundary. The device may provide information indicating whether the spectroscopic measurement performed on the unknown sample corresponds to the quantification model.