On-Demand Whole Slide Image Registration Using ROI Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Whole slide imaging poses challenges during the visual examination of digitized slides, necessitating improved techniques for registration and analysis.
Innovation Solution
An apparatus and method for on-demand registration of whole slide images using a processor to receive digital images with pyramid structures, identify regions of interest, determine a transformation matrix, and register corresponding regions across different magnification levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If whole slide images are scanned at high resolution to maintain analysis quality, then image quality is improved, but storage requirements and computation time increase
Solution Approach 1:
The patent divides the whole slide image into multiple pyramid levels with different resolutions. Instead of processing the entire high-resolution image, the system segments the image into a hierarchy of resolutions, allowing computation to occur at lower resolution levels first and progressively refine to higher resolutions only where needed, thereby reducing overall computation time while maintaining analysis quality.
Solution Approach 2:
The patent applies different resolution qualities to different regions of the image based on their importance. Regions of interest are processed at higher resolutions while less critical areas use lower resolutions. This local quality approach maintains analysis quality for critical regions while reducing computation time and storage for the entire image.
2Measurement precision
If whole slide images are scanned at high resolution to maintain analysis quality, then image quality is improved, but storage requirements increase
Solution Approach 1:
The patent segments the image data into a pyramid structure with multiple resolution levels. Instead of storing one large high-resolution image, the system stores multiple smaller images at different resolutions. This segmentation reduces total storage requirements while maintaining the ability to access high-resolution data when needed for analysis.
Solution Approach 2:
The patent stores different quality levels in different regions of the pyramid structure. Full high-resolution data is stored only for regions that require detailed analysis, while other regions are stored at lower resolutions. This local quality approach optimizes storage requirements while preserving image quality where it matters most.
3Measurement precision
If registration is performed at high magnification level to ensure accuracy, then registration precision is improved, but computation time increases
Solution Approach 1:
The patent performs preliminary registration actions at lower magnification levels before finalizing registration at high magnification. The coarse registration is established first using lower-resolution pyramid levels, which reduces the complexity of the subsequent high-resolution registration step. This preliminary action ensures that the final high-magnification registration starts from a pre-aligned state, improving precision while reducing computation time.
Solution Approach 2:
The patent adds the dimension of magnification level to the registration process by utilizing the pyramid structure. Instead of performing registration in a single dimension at high resolution, the system performs registration across multiple dimensions (magnification levels), progressing from low to high magnification. This dimensional approach breaks down the complex high-resolution registration into manageable steps.
Data Source
AI summary
A method for on-demand registration of whole slide images (WSIs), comprising receiving at least two digital images, a first digital image of a reference slide and a second digital image on a candidate slide, each at a first magnification level, identifying a first region of interest (ROI) on the first digital image at a target magnification level, registering a portion of the second digital image to a portion of the first digital slide image at the first magnification level to derive a transformation matrix, applying the transformation matrix to the first ROI to identify a second ROI on the second digital image at the first magnification level, mapping the second ROI to a corresponding second ROI on the second digital image at the target magnification level, and registering the corresponding second ROI to the first ROI at the target magnification level.