Cell Clustering With Traceable Fluorescence Data Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge of clustering multidimensional data from flow cytometers results in information loss due to dimension compression, making it difficult to verify the validity of clustering results against the original measurement data.

Innovation Solution

An information processing apparatus that clusters dimensionally-compressed data while allowing verification by associating it with the original measurement data, using fluorescence separation and machine learning to derive expression levels of fluorescent substances, and providing traceability through image displays.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If dimension compression is performed on multidimensional flow cytometer data, then clustering performance is improved, but information loss occurs making verification difficult

Engineering Contradiction:
Improveclustering performanceVSAvoidmeasurement data information
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent applies preliminary action by performing dimension compression on flow cytometer data before clustering analysis. This pre-processing step reduces the computational complexity and improves clustering performance by eliminating redundant dimensions while preserving the essential information needed for accurate cell population identification.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements copying by creating and storing a correspondence relationship between the compressed dimension data and the original measurement data. This copy allows users to trace back and verify clustering results against the original high-dimensional data, eliminating information loss concerns while maintaining clustering efficiency.

Inventive Principle:
Principle #26Copying

2Adaptability or versatility

If the number of fluorescences measured at one time is increased, then measurement capability is improved, but data analysis complexity increases due to combinatorial explosion

Engineering Contradiction:
Improvemeasurement capabilityVSAvoiddata analysis complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies parameter changes by transforming the high-dimensional measurement data into a compressed dimensional space. This parameter transformation reduces the complexity of data analysis while preserving the essential characteristics needed for clustering, effectively managing the combinatorial explosion issue.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements dimensionality change by compressing the high-dimensional flow cytometer data into a lower-dimensional space suitable for clustering analysis. This dimensionality reduction maintains the essential information while making the data manageable for computational analysis, thereby reducing analysis complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Productivity

If dimension compression is applied to enable clustering, then processing efficiency is improved, but traceability to original data is reduced

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidtraceability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies feedback by establishing a correspondence relationship that allows traceability from compressed data back to original measurement data. This feedback mechanism ensures that while processing efficiency is improved through compression, users can still verify the accuracy and validity of clustering results by referencing the original high-dimensional data.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3881224B1Information processing apparatus, information processing method, and program
Publication Date: 2026.04.15 SONY GROUP CORP
  • EP3881224B1 patent drawingFigure 1
  • EP3881224B1 patent drawingFigure 2
  • EP3881224B1 patent drawingFigure 3A

AI summary

There is provided an information processing apparatus including: an information storage unit that stores a result of sensing light from cells, namely first data, and a result of separating the first data into a plurality of fluorescences, namely second data, in association with each other; a clustering unit that clusters the cells into a plurality of clusters on a basis of the second data; and an output unit that outputs a clustering result from the clustering unit. The output unit additionally outputs at least one or more of the first data and the second data about the cells included in a cluster selected by a user from among the plurality of clusters.