Breast Tomosynthesis Calcification Detection With Noise Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to accurately distinguish between calcification images and noise in tomographic images, leading to potential misdiagnosis in breast tissue analysis.
Innovation Solution
An image processing apparatus and method that includes calcification candidate image detection, group generation, and determination processing using machine-learned models to differentiate between calcification images and noise, employing calcification candidate image detection from projection and tomographic images, and generating weighted tomographic and synthesized two-dimensional images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a synthesized two-dimensional image is generated by synthesizing multiple tomographic images, then the overlap of mammary glands is reduced and image quality is improved, but noise is drawn on the synthesized image as pseudo calcification images leading to misdiagnosis
Solution Approach 1:
The patent extracts and separates calcification candidate images from noise by detecting calcification candidates in individual tomographic images and generating a synthesized two-dimensional image only from confirmed calcification candidates, thereby removing noise that would otherwise be drawn as pseudo calcification images
Solution Approach 2:
The patent performs preliminary detection of calcification candidates in each tomographic image before synthesis, and uses machine learning models to pre-judge whether detected candidates are true calcifications or noise, preventing noise from being included in the final synthesized image
2Measurement precision
If the weight of lesions is increased to emphasize calcification in tomographic images, then calcification detection sensitivity is improved, but noise cannot be distinguished from small calcification images
Solution Approach 1:
The patent segments the detection process into multiple stages: first detecting calcification candidates in individual tomographic images, then using machine learning models to classify each candidate as either true calcification or noise, and finally synthesizing only from confirmed calcifications. This segmentation allows sensitive detection while maintaining noise separation capability
Solution Approach 2:
The patent introduces machine learning models as an intermediary between the detection stage and synthesis stage, where the models act as a mediator that filters and classifies detected candidates before they are included in the final synthesized image, enabling distinction between calcification and noise
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
There is provided an image processing apparatus including: at least one processor, in which the processor is configured to execute: calcification candidate image detection processing of detecting a calcification candidate image estimated to be a calcification image from a series of a plurality of projection images obtained by tomosynthesis imaging of a breast or a plurality of tomographic images obtained from the plurality of projection images; calcification candidate image group generation processing of generating a calcification candidate image group by cutting out a region including the calcification candidate image detected by the calcification candidate image detection processing, from each of the plurality of projection images; and calcification determination processing of determining whether or not the calcification candidate image is a calcification image based on the calcification candidate image group generated by the calcification candidate image group generation processing.