Morphology-Based Duct Tissue Feature Extraction from Local Cell Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing duct tissue feature extraction technologies rely heavily on the overall quantity of nuclei, which is difficult to count accurately and results in low accuracy in reflecting the feature of duct tissue.
Innovation Solution
A method and apparatus for obtaining duct tissue features using computer vision, which involves obtaining an image of duct tissue, determining feature obtaining regions adapted to duct morphology, extracting cell features from these regions, and integrating these features to obtain the duct tissue feature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the overall quantity of nuclei is used as the main feature for duct tissue extraction, then the extraction process is simple, but the accuracy of reflecting duct tissue features is low
Solution Approach 1:
The patent segments the duct tissue image into multiple sub-regions and further divides cells into different types (epithelial cells, luminal cells, myoepithelial cells). Instead of treating the entire tissue as one unit, the system extracts features from each segment separately and combines them, thereby improving measurement precision while maintaining manageable complexity through systematic processing
Solution Approach 2:
The patent applies different feature extraction methods to different local regions and cell types within the duct tissue. Each cell type (epithelial, luminal, myoepithelial) has specific morphological features extracted locally, rather than applying a uniform extraction method across the entire tissue sample. This localized approach significantly improves the accuracy of duct tissue feature representation
2Ease of operation
If the overall quantity of nuclei is counted to represent duct tissue features, then the measurement process is straightforward, but the difficulty of accurate counting is high due to large quantities
Solution Approach 1:
The patent segments the large quantity of nuclei into smaller manageable units by first dividing the duct tissue into sub-regions and then further segmenting by cell type. This hierarchical segmentation transforms the overwhelming task of counting all nuclei into a series of smaller, more accurate local counting operations, thereby improving measurement precision
Solution Approach 2:
The patent introduces cell type classification as an intermediary step between image acquisition and feature extraction. By classifying nuclei into different cell types (epithelial, luminal, myoepithelial) based on morphological features, the system creates intermediate categories that make the counting and measurement process more manageable and accurate, rather than directly counting all nuclei as a single group
Data Source
AI summary
Aspects of the disclosure are directed to the field of artificial intelligence technologies and provides a method and an apparatus for obtaining a feature of duct tissue based on computer vision, an intelligent microscope, a storage medium, and a computer device. The method can include the steps of obtaining an image including duct tissue, determining, in an image region corresponding to the duct tissue in the image, at least two feature obtaining regions adapted to duct morphology of the duct tissue, obtaining cell features of cells of the duct tissue in the feature obtaining regions respectively, and obtaining a feature of the duct tissue based on the cell features of the cells of the duct tissue in the feature obtaining regions respectively.