Microscope Slide Coordinate Registration via Cell Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated imaging systems face challenges in accurately relocating cells of interest from one imaging station to another due to differences in coordinate systems and inaccuracies in slide positioning, which can result in cells being displaced from the image, especially when switching between low and high magnification views.
Innovation Solution
The system determines a coordinate transformation between stage coordinates of two imaging stations using cells on a biological sample, selecting reference images that are adequately covered and spaced apart to calculate a mathematical transformation that converts coordinates from one station to another, allowing precise relocation of cells of interest without the need for pre-printed fiducial marks on microscope slides.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-printed fiducial marks are used on microscope slides to establish coordinate systems, then coordinate transformation between imaging stations is enabled, but slide manufacturing costs increase and the process becomes more complex
Solution Approach 1:
The slide serves its own coordinate registration function through the biological sample itself. The sample's natural features (cells, tissue structures) act as the fiducial reference, eliminating the need for externally added fiducial marks. This self-service approach reduces manufacturing complexity and cost while maintaining coordinate transformation accuracy.
Solution Approach 2:
The system creates a digital copy of the sample's coordinate information by imaging the sample at multiple stations and using image correlation algorithms to establish coordinate transformations. This digital copying approach replaces physical fiducial marks, simplifying the manufacturing process while preserving measurement precision.
2Measurement precision
If pre-printed fiducial marks are used on microscope slides, then coordinate transformation can be performed, but the time required to locate and detect these marks increases
Solution Approach 1:
The sample itself provides the reference information needed for coordinate transformation. By using the sample's inherent features rather than external fiducial marks, the system eliminates the time-consuming search and detection process associated with locating fiducial marks, while still achieving accurate coordinate registration.
Solution Approach 2:
The system performs preliminary imaging of the sample to capture its coordinate information before transformation is needed. This preliminary action stores the sample's spatial configuration in advance, allowing rapid coordinate transformation without real-time fiducial mark detection.
3Measurement precision
If fiducial marks are searched at high magnification, then precise coordinate registration can be achieved, but the field of view becomes smaller making mark location more difficult
Solution Approach 1:
Instead of searching for fiducial marks at high magnification (which reduces field of view), the system inverts the approach by using low-magnification images to establish coordinate relationships. The sample features visible at low magnification serve as reference points, and their coordinates are transformed to high-magnification space, eliminating the field-of-view limitation.
Solution Approach 2:
The system transitions from searching in the spatial dimension (scanning for marks in the field of view) to using the image data dimension. By correlating image data from different magnifications and stations, the system establishes coordinate transformations without being constrained by the limited field of view at high magnification.
4Ease of operation
If stage coordinates are used directly between imaging stations, then relocation is simple, but positioning errors occur due to stage inaccuracies and slide loading variations
Solution Approach 1:
The system introduces an intermediary coordinate system based on the sample itself rather than directly using stage coordinates. The sample's features serve as a mediator that links different imaging stations, allowing simple stage movement while achieving precise positioning through sample-referenced coordinate transformation.
Solution Approach 2:
The system uses feedback from actual sample imaging to correct stage coordinate errors. By comparing the observed sample positions with expected positions and using this feedback to refine coordinate transformations, the system achieves precise cell relocation despite initial stage positioning inaccuracies.
Data Source
AI summary
Systems, methods and computer program products for mapping coordinates of various imaging stations are described. In some implementations, cells (e.g., red blood cells) in a biological specimen can be used for determining the mapping information between the imaging stations. The use of cells allows a target image (e.g., an image of a sub-region of cells in the biological specimen) taken by one imaging station to be pattern-matched to a reference image (e.g., an image showing a larger region of cells in the biological specimen that also includes the sub-region) taken by another imaging station. Once the target image is matched to the reference image, point by point correspondence (and therefore coordinates) between the target image and the reference image can be established for computing the coordinate transformation to map the imaging stations.