Breast Calcification Shape Determination via Tomosynthesis Variance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing techniques for diagnosing breast calcification in tomographic and synthesized two-dimensional images struggle to accurately represent the shape of calcification images due to noise and overlap issues, making it difficult to determine malignancy or benignancy.
Innovation Solution
An image processing apparatus and method that detects calcification images through tomosynthesis imaging, generates region-of-interest image groups, calculates variance values of feature amounts, and determines the shape type of calcification images based on these values, enabling accurate shape type determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If calcification images are displayed in tomographic images or synthesized two-dimensional images, then diagnostic information is provided, but the calcification image shape becomes blurred due to noise and visibility is lowered
Solution Approach 1:
The patent segments the analysis by extracting region-of-interest images containing calcification images from projection images at multiple irradiation angles, then calculates variance values for each segmented region across different angles. This segmentation approach isolates the calcification images from surrounding noise and tissue, enabling accurate shape type determination without the blurring effects present in traditional tomographic or synthesized two-dimensional images.
2Shape
If synthesized two-dimensional images are generated from multiple tomographic images, then overlap of mammary glands is reduced, but the calcification image shape is not accurately represented
Solution Approach 1:
Instead of synthesizing two-dimensional images from multiple tomographic images (which loses shape accuracy), the patent transitions to analyzing the third dimension by calculating variance values of calcification image features across multiple irradiation angles. This dimensional approach to analysis preserves the accurate representation of calcification shapes while maintaining the advantage of reduced mammary gland overlap through multi-angle imaging.
3Measurement precision
If variance values of feature amounts are calculated from region-of-interest images at multiple irradiation angles, then calcification image shape type can be accurately determined, but processing complexity increases
Solution Approach 1:
The patent first segments and extracts only the region-of-interest images containing calcification images from the full projection images at multiple irradiation angles. This segmentation reduces the amount of data requiring variance calculation, thereby managing processing complexity while maintaining the accuracy benefits of multi-angle variance analysis for calcification shape type determination.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for precise determination of calcification image shapes, improving diagnostic accuracy by reducing noise effects and enhancing visibility of calcification image shapes in both tomographic and synthesized images.
Implementation Method 1
a series of a plurality of projection images obtained by tomosynthesis imaging of a breast
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There are provided an image processing apparatus, an image processing method, and a program capable of accurately determining a type of a shape of a calcification image. An image processing apparatus includes at least one processor (16). The processor is configured to execute calcification image detection processing of detecting a calcification image based on a plurality of tomographic images obtained from a series of a plurality of projection images obtained by tomosynthesis imaging of a breast, region-of-interest image group generation processing of generating a region-of-interest image group by cutting out, as a region-of-interest image, a region including the calcification image detected by the calcification image detection processing from each of the plurality of projection images, variance value calculation processing of calculating a variance value of feature amounts of each of the region-of-interest images included in the region-of-interest image group, and shape type determination processing of determining a type of a shape of the calcification image based on the variance value calculated by the variance value calculation processing.