Imaging Analyzer Segmentation Using Reusable Clustering Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional imaging analysis methods are inefficient for processing multiple samples, requiring excessive time and not facilitating comparative analysis, and do not effectively utilize segmentation results across samples.
Innovation Solution
An imaging analyzer that includes a clustering execution section for classifying spectral data into clusters, a model storage section for storing clustering models, and a segmentation execution section for creating segmentation images using these models, allowing for rapid segmentation and comparison across multiple samples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If segmentation is executed for each sample using conventional statistical analysis processing, then regions with same components can be indicated to analysts, but the time required for processing becomes very long
Solution Approach 1:
The patent applies preliminary action by performing clustering processing on a reference sample before analyzing test samples. The clustering model obtained from the reference sample is stored and then applied to multiple test samples, eliminating the need to perform time-consuming segmentation processing on each sample individually while maintaining segmentation accuracy.
2Ease of operation
If conventional segmentation processing is performed on each sample independently, then segmentation results can be obtained, but comparison analysis among multiple samples becomes difficult
Solution Approach 1:
The patent implements universality by creating a clustering model from a reference sample that can be universally applied to multiple different test samples. This single model serves multiple functions: it segments the reference sample, segments all test samples, and enables direct comparison of segmentation results across samples, making the system versatile for comparative analysis.
3Reliability
If segmentation results from one sample are not used for analyzing other samples, then each sample can be processed independently, but the segmentation results are not effectively utilized
Solution Approach 1:
The patent applies copying by creating a clustering model from the reference sample that captures the segmentation patterns and characteristics. This model is then copied and applied to multiple test samples, effectively reusing the segmentation knowledge gained from the reference sample to efficiently analyze other samples while maintaining reliability.
Data Source
AI summary
An imaging mass spectrometry unit (1) executes predetermined analysis on each of a plurality of micro areas set in a two-dimensional measurement region on a sample or a three-dimensional measurement region in a sample to acquire spectrum data. A clustering execution section (21) classifies spectrum data for a plurality of measurement points obtained for a reference sample into any of a plurality of clusters. A clustering model information storage section (22) stores a clustering model obtained by clustering processing. A segmentation execution section (23) classifies respective spectral data for a plurality of measurement points obtained for a sample other than a reference sample using a clustering model, and a spatial distribution image creation section (24) creates a segmentation image obtained by partitioning a two-dimensional or three-dimensional image into a plurality of small regions on the basis of a result of the classification.