Co-Expression ROI Mapping for Reproducible Immunoscore Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional immunoscore computation methods for multiplex assays are subjective and biased due to manual selection of fields of view (FOVs) or regions of interest (ROIs), leading to non-reproducible results.
Innovation Solution
A computer-implemented method for co-expression analysis that automatically identifies co-localized regions of interest (ROIs) by computing heat maps, applying thresholds and filters, and registering markers in a common coordinate system, enabling objective and reproducible immunoscore computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual selection of fields of view (FOVs) or regions of interest (ROIs) is used in traditional immunoscore computation, then the process allows expert reader judgment and flexibility, but the results become subjective and biased, leading to non-reproducible outcomes
Solution Approach 1:
The patent replaces the manual mechanical process of expert readers selecting FOVs and counting cells with an automated computer-based image analysis system. The system automatically identifies ROIs, segments cells, and computes immunoscores, eliminating human subjectivity while maintaining the ability to process and evaluate tissue sample images. This substitution directly resolves the contradiction by removing the source of variability (human judgment) while preserving the analytical capability.
2Reliability
If automated algorithms are used for ROI identification and cell counting, then objectivity and reproducibility are improved, but the complexity of the analysis system increases
Solution Approach 1:
The patent divides the complex task of immunoscore computation into distinct modular segments: (1) automatic ROI identification based on image features, (2) cell segmentation within ROIs, (3) marker detection and classification, and (4) score computation. Each module performs a specific function independently, which reduces overall system complexity while maintaining objectivity. The segmentation allows the system to handle complexity in manageable, reusable components.
3Loss of information
If multiple stains are used to detect different biomarkers in multiplex assays, then rich diagnostic information is generated, but the analysis becomes more complex and time-consuming
Solution Approach 1:
The patent implements continuous automated processing that handles multiple stains and markers simultaneously without interruption. The system processes all color channels and detects multiple markers in parallel through automated image analysis, eliminating the need for sequential manual examination of each stain. This continuous automated action preserves all diagnostic information from the multiplex assay while dramatically reducing the time required compared to traditional sequential analysis methods.
Data Source
Figure 1A~1B
Figure 2A
Figure 2B
AI summary
Described herein are methods for co-expression analysis of multiple markers in a tissue sample comprising: computing a heat map of marker expression for each of a plurality of single marker channel images, wherein each of the plurality of single marker channel images comprise a single marker; identifying one or more candidate regions of interest in each heat map of marker expression; computing overlay masks comprising the identified one or more candidate regions of interest from each heat map of marker expression; determining one or more co-localized regions of interest from the overlay masks; mapping the one or more co-localized regions of interest to a same coordinate position in each of the plurality of single marker channel images; and estimating a number of cells in at least one of the determined one or more co-localized regions of interest in each of the plurality of single marker channel images.