Radiographic Structure Recognition for Accurate Tissue Emphasis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiographic image processing technologies struggle to individually specify and emphasize structures within a subject, such as lesions or artifacts, due to small differences in attenuation coefficients and volumes among soft tissues, making accurate diagnosis difficult.
Innovation Solution
A radiographic image processing device and non-transitory computer readable medium that acquire multiple images with different energy levels, recognize structures using a structure recognition unit, calculate attenuation coefficients, and perform image processing to emphasize specific structures, using a learned model for structure identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subtraction processing is used to generate soft part images or bony part images, then images for diagnosis can be generated, but it is difficult to individually specify each soft part tissue due to small differences in attenuation coefficients and volumes
Solution Approach 1:
The patent segments the subject into multiple structures (e.g., lungs, heart, fat, muscle) and calculates attenuation coefficients for each structure individually. This segmentation allows the system to distinguish between different soft part tissues that have small differences in attenuation coefficients by treating each structure separately rather than as a collective soft part group.
Solution Approach 2:
The patent changes the parameter approach by calculating attenuation coefficients for each identified structure individually rather than using a single subtraction processing parameter for all soft parts. By varying the attenuation coefficient calculation for each structure based on its specific volume and composition, the system can individually specify structures that would otherwise be indistinguishable.
2Measurement precision
If multiple radiographic images with different energies are acquired, then attenuation coefficients can be calculated for individual structures, but the device complexity and processing time increase
Solution Approach 1:
The patent performs preliminary structure recognition and identification before calculating attenuation coefficients. By first identifying the structures in the subject and their respective volumes, the system prepares the necessary information in advance, which simplifies the subsequent attenuation coefficient calculation process and reduces overall processing complexity.
Solution Approach 2:
The patent introduces structure recognition and volume calculation as intermediary steps between acquiring multiple radiographic images and calculating attenuation coefficients. These intermediary processes organize and structure the data from multiple images, making the final attenuation coefficient calculation more manageable and less complex.
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
Enables accurate and efficient specification and emphasis of structures within radiographic images, reducing the impact of individual subject differences and improving diagnostic clarity.
Implementation Method 1
A radiographic image acquisition unit acquires a plurality of radiographic images of a specific subject using radiations having energies different from each other
Data Source
AI summary
A radiographic image processing device includes a radiographic image acquisition unit that acquires a plurality of radiographic images of a specific subject taken using radiations having energies different from each other, a structure recognition unit that recognizes structures, which is included in the subject, using the radiographic images, an attenuation coefficient calculation unit that calculates attenuation coefficients μ of the radiation for the structures, which are recognized by the structure recognition unit, using recognition results of the structure recognition unit and the plurality of radiographic images, and an image processing unit that performs image processing on the radiographic images using the attenuation coefficients.