Tissue Section Alignment via Object Correspondences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating multiple biomarkers across tissue samples are limited by the difficulty of combining multiple stains without cross-reactivity and distinguishing colors, restricting the number of stains that can be used on a single slide, which hampers the gathering of information while increasing costs and complexity.
Innovation Solution
A method for aligning digital images of tissue sections using identified tissue objects and extracted image analysis features to determine correspondences, allowing for the alignment of images regardless of morphological changes and deformations, enabling the combination of biomarker information from multiple stains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple stains are combined on a single slide to evaluate multiple biomarkers, then the amount of information gathered increases, but the difficulty of combining stains without cross-reactivity and distinguishing colors increases, limiting the number of stains that can be used
Solution Approach 1:
The invention divides the evaluation of multiple biomarkers across separate tissue sections rather than attempting to combine multiple stains on a single section. Each section is stained with a different biomarker, and digital image analysis aligns the sections to create a composite view, effectively segmenting the multi-biomarker evaluation process across spatial dimensions.
Solution Approach 2:
The invention transitions from evaluating multiple biomarkers on a single 2D plane (which causes color confusion and cross-reactivity) to evaluating them across multiple 2D sections that are then aligned in 3D space through digital image registration, adding a dimensional solution to a 2D problem.
2Adaptability or versatility
If elastic registration methods are used to align whole-slide images of consecutive tissue sections, then morphological changes and deformations are handled, but computational cost increases and precision is insufficient for the use case
Solution Approach 1:
The invention introduces tissue objects (cells, structures) as intermediary reference points between two tissue sections. These objects serve as anchors for alignment, allowing the system to handle deformations while achieving precise registration by matching corresponding objects across sections rather than relying on pixel-level elastic warping.
3Ease of manufacture
If the number of stains used in an assay is reduced, then the assay becomes less expensive and easier to use in the clinic, but the amount of information gathered from the assay decreases
Solution Approach 1:
The invention merges information from multiple separately-stained tissue sections through digital image alignment and integration. By aligning sections stained with different biomarkers and combining them into a composite analysis, the system achieves multi-biomarker information gathering equivalent to using multiple stains on one section, while actually using simpler staining protocols on individual sections.
Data Source
AI summary
The present invention concerns identifying matching tissue objects in two sections of a tissue block, as imaged using a microscope, for the purpose of mapping the biomarker-specific staining in one section onto the other section. The invention is useful because, in many workflows, it is not possible to add a sufficient number of different biomarker-specific stains to a single slide. By staining instead multiple slides, and mapping the staining data obtained across the slides, one obtains a data set that is similar to what one would be able to obtain by staining a single slide with all those stains. The invention first identifies a set of obviously correct matches, then propagates from those matches, using a priority queue driven process, to optimally match up all fibers in the two sections. The matching is based on the shape and neighborhood configuration of each tissue object.