Mitotic Figure Detection in Digital Pathology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital pathology systems for detecting and counting mitotic figures in tissue samples are not reliable enough for automatic diagnostic use, leading to inconsistent and costly manual counting by pathologists.
Innovation Solution
A system and method that involves color filtering, shape filtering, and machine learning techniques, including convolutional neural networks, to accurately identify and count mitotic figures in biopsy samples, using feature extraction and classification to distinguish between mitotic and non-mitotic figures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual counting by pathologists is used, then diagnostic accuracy can be maintained, but time consumption and cost increase significantly
Solution Approach 1:
The patent replaces the mechanical manual counting process with an automated computer-based system that uses image processing and machine learning algorithms to detect and count mitotic figures, thereby reducing time consumption while maintaining diagnostic accuracy
Solution Approach 2:
The system creates a digital copy of the tissue sample image and performs automated analysis on the copy, allowing pathologists to review only the detected mitotic figures rather than manually examining the entire image, thus significantly reducing time consumption
2Productivity
If simple feature extraction with thresholding is used, then processing speed increases, but classification reliability deteriorates
Solution Approach 1:
The system performs preliminary processing steps including color filtering to identify candidate mitotic figures and shape filtering to refine candidate selection before applying the final classification algorithm, which improves both processing efficiency and classification reliability by reducing the number of figures requiring detailed analysis
Solution Approach 2:
The patent transforms the classification problem from using simple intensity thresholds to using learned parameters from training data, where the system learns optimal feature combinations and decision boundaries that improve classification reliability while maintaining reasonable processing speed
3Productivity
If automated detection systems are implemented, then productivity increases, but measurement precision and reliability are insufficient
Solution Approach 1:
The system dynamically adjusts its detection strategy by first identifying candidate mitotic figures using color and shape filters, then applying more computationally intensive classification algorithms only to these candidates, which maintains high detection accuracy while improving processing speed
Solution Approach 2:
The system uses training data to learn from examples of mitotic and non-mitotic figures, incorporating feedback from labeled data to continuously improve its classification accuracy and detection precision
Data Source
AI summary
A method and system for detecting and counting mitotic figures in an image of a biopsy sample stained with at least one dye, includes color filtering the image in a computer process to identify pixels in the image that have a color which is indicative a mitotic figure; extracting the mitotic pixels in the image that are connected to one another in a computer process, thereby producing blobs of mitotic pixels; shape-filtering and clustering the blobs of mitotic pixels in a computer process to produce mitotic figure candidates; extracting sub-images of mitotic figures by cropping the biopsy sample image at the location of the blobs; extracting two sets of features from the mitotic figure candidates in two separate computer processes; determining which of the mitotic figure candidates are mitotic figures in a computer classification process based on the extracted sets of features; and counting the number of mitotic figures per square unit of biopsy sample tissue.