AI Imaging Flow Cytometry for T-Cell and B-Cell Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods lack a high-throughput tool for systematically quantifying and characterizing the morphology of the immunological synapse and its correlation to T-cell response, and identifying properties predictive for the efficacy of therapeutic antibodies.

Innovation Solution

A method using single-cell imaging flow cytometry combined with artificial intelligence for preprocessing, feature engineering, and explainable predictive machine learning to classify cells in a mixture of T-cells and B-cells or antigen-presenting cells, based on differential labeling and morphological features.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional flow cytometry or imaging methods are used, then cell analysis can be performed, but high-throughput systematic quantification and characterization of immunological synapse morphology is lacking

Engineering Contradiction:

Improvehigh-throughput analysis capabilityVSAvoidsystematic quantification of synapse morphology

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent combines flow cytometry with imaging capabilities to create imaging flow cytometry (IFC), merging the high-throughput cell analysis capability of flow cytometry with the morphological visualization capability of imaging. This allows simultaneous acquisition of both quantitative cell data and detailed synapse morphology images at high throughput, resolving the contradiction between productivity and measurement precision.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces artificial intelligence algorithms as an intermediary between the raw imaging data and the final quantification results. These AI algorithms automatically analyze and quantify synapse morphology features from the images, enabling systematic high-throughput characterization without manual intervention, thus achieving both high productivity and precise measurement.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If traditional gating strategies are used for data analysis, then simple classification is possible, but robust and accurate analysis of high-throughput imaging data is limited

Engineering Contradiction:

Improveaccuracy of cell classificationVSAvoidcomplexity of analysis pipeline

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional manual gating strategies (mechanical/interactive analysis method) with artificial intelligence-based automated analysis. The AI algorithms automatically perform cell classification and synapse morphology characterization, achieving higher accuracy and robustness while reducing the complexity of the analysis pipeline for users.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Loss of information

If no systematic method is used, then existing tools are available, but identification of properties predictive for antibody efficacy is not achieved

Engineering Contradiction:

Improvepredictive information about antibody efficacyVSAvoidrapid antibody screening capability

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The patent performs preliminary systematic quantification and characterization of synapse morphology using imaging flow cytometry and AI analysis before evaluating antibody efficacy. This preliminary action extracts predictive features from synapse morphology that can be used to screen and rank antibody candidates, enabling rapid identification of effective antibodies while preserving all relevant information.

Inventive Principle:

Principle #10Preliminary action

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

Enables the quantitative prediction of therapeutic antibody efficacy by linking morphological features with function, facilitating rapid antibody screening and functional characterization.

Implementation Method 1

applying at least labelled antibodies binding to F-actin, MHCII and CD3 to the cell mixture

Methodology Applied

Scientific EffectAntibody-antigen binding:

Implementation Method 2

acquiring at least one image of the cell mixture

Methodology Applied

Scientific EffectLight microscopy imaging:

Data Source

PatentUS20250306022A1Method for classifying cells

Publication Date: 2025.10.02 F HOFFMANN LA ROCHE INC

AI summary

Herein is reported a method for classifying cells in a cell mixture, wherein the mixture comprises T-cells and B-cells, comprising the steps of first applying at least labelled antibodies binding to F-actin, MHCII and CD3 to the cell mixture to obtained a labelled cell mixture, wherein the antibodies are each labelled with a dye, wherein the dyes have different (non-overlapping) emission wavelengths, second acquiring at least one image of the cell mixture, and third classifying the cells in the cell mixture to be an isolated cell if the cell is a single cell, is F-Actin positive and is MHCII positive and CD3 negative, or is MHCII negative and CD3 positive, or to be a doublet or multiplet of cells if the cell is an aggregate of two or three cells, is F-Actin positive, MHCII positive and CD3 positive.