Mammography Schema Image Selection Using Breast Shape Feature Amounts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The variability in breast shape between individuals and the compression effect during mammography imaging result in significant differences between mammography images and standard schema images, making it time-consuming and difficult to match lesion positions accurately.
Innovation Solution
An image processing device calculates feature amounts relevant to the breast shape from mammography images and selects a corresponding schema image from a set of predetermined types, allowing for accurate matching without the need for a different schema image for each imaging session.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a standard schema image is used for all mammography images, then the schema image database remains simple and manageable, but the breast shape in the mammography image and the schema image may be significantly different, making it time-consuming to match lesion positions
Solution Approach 1:
The patent segments the single standard schema image into multiple schema images with different breast shapes. Each schema image is associated with specific shape feature amount ranges, allowing the system to divide the matching task into more precise segments that correspond to actual breast shape variations, thereby reducing the time to match lesion positions while maintaining manageable database complexity.
Solution Approach 2:
The patent introduces shape feature amounts as parameters to differentiate between various breast shapes. By calculating shape feature amounts from mammography images and selecting schema images based on matching these parameters, the system achieves accurate shape correspondence without requiring an excessively complex database structure.
2Measurement precision
If a schema image showing the breast shape of each mammography image is created, then the breast shape correspondence is accurate, but the schema image becomes different for each imaging session even for the same subject, making it difficult to match lesion positions from past examinations
Solution Approach 1:
The patent performs preliminary classification of breast shapes by calculating shape feature amounts and selecting appropriate schema images from predetermined types before lesion position matching. This preliminary action ensures that the same breast shape type always corresponds to the same schema image type, providing consistency for longitudinal comparison while maintaining accurate shape correspondence.
Solution Approach 2:
The patent uses shape feature amounts as stable parameters to classify and select schema images. By basing schema image selection on these quantitative parameters rather than direct image copying, the system achieves both accurate shape correspondence and consistent schema images for the same breast shape type across different imaging sessions.
3Measurement precision
If multiple types of schema images are prepared for different breast shapes, then the matching accuracy between mammography images and schema images improves, but the complexity of selecting the appropriate schema image increases
Solution Approach 1:
The patent implements an automated system that calculates shape feature amounts from mammography images and automatically selects the appropriate schema image type based on these calculations. This self-service approach eliminates the need for manual schema image selection, reducing the perceived complexity for users while maintaining high matching accuracy through systematic parameter-based selection.
Data Source
AI summary
Provided is an image processing device that includes a hardware processor. The hardware processor calculates a feature amount relevant to a breast shape from a mammography image. The hardware processor selects a schema image corresponding to the breast shape of the mammography image from a plurality of types of predetermined schema images based on the feature amount relevant to the breast shape calculated by the hardware processor.