Pixel Classifier Segments Blurred Tissue Areas for Cancer Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cancer scoring methods are inaccurate due to the inclusion of blurred areas in digital images of stained tissue, which are difficult for human pathologists and computer-assisted systems to identify and exclude, leading to inconsistent and unreliable diagnostic results.
Innovation Solution
A method involving the training of a pixel classifier to distinguish between unblurred and blurred areas in digital images of stained tissue by artificially blurring a learning tile and comparing pixel values, allowing for the segmentation of images and exclusion of blurred regions from analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual marking of blurred areas by pathologists is used, then large blurred areas can be excluded, but small blurred areas caused by microdroplets cannot be identified
Solution Approach 1:
The patent replaces the manual mechanical marking process with an automated computer-based image processing system that uses algorithms to detect and segment blurred areas, thereby substituting human manual operation with automated mechanical/computational processes
Solution Approach 2:
The system creates a processed copy of the original digital image where blurred areas are segmented and highlighted, allowing the pathologist to review the automated results and make final decisions based on this copied representation rather than working directly with the original complex image
2Productivity
If computer-assisted image analysis is performed on all areas of the image, then productivity increases, but accuracy decreases due to inclusion of blurred areas
Solution Approach 1:
The patent segments the digital image into distinct regions: blurred areas and unblurred areas. This segmentation allows the system to apply different processing rules to different parts of the image, performing automated analysis only on valid unblurred regions while excluding blurred regions that would compromise accuracy
Solution Approach 2:
The system applies different quality standards and processing approaches to different local regions of the image. Unblurred areas undergo full automated analysis for productivity, while blurred areas are excluded from analysis to maintain reliability, thus allowing local optimization of both speed and accuracy
Data Source
AI summary
A method for identifying blurred areas in digital images of stained tissue involves artificially blurring a learning tile and then training a pixel classifier to correctly classify each pixel as belonging either to the learning tile or to a blurred copy. A learning tile is first selected from a digital image of stained tissue. The learning tile is copied and blurred by applying a filter to each pixel. The pixel classifier is trained to correctly classify each pixel as belonging either to the learning tile or to the blurred, copied learning tile. The pixel classifier then classifies each pixel of the entire digital image as most likely resembling either the learning tile or the blurred learning tile. The digital image is segmented into blurred and unblurred areas based on the pixel classification. The blurred areas and the unblurred areas of the digital image are identified on a graphical user interface.