Radiotherapy CT Image Boundary Detection via Top-Down Pixel Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for processing radiotherapy positioning CT images struggle to accurately identify human body data amidst complex environments, such as those with multiple accessories and varied shapes, leading to incomplete and inaccurate image reconstruction, which affects RT planning.
Innovation Solution
A method that defines pixels with a CT value greater than or equal to a threshold as human body pixels, counts these pixels from top to bottom to determine boundaries, and uses binarization and contour identification to separate human body image data from non-human body data, ensuring accurate image reconstruction regardless of the supporting environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the prior art method of removing images from bottom to top with preset heights is used, then the processing is simple, but the human body identification accuracy deteriorates in complex environments with multiple accessories
Solution Approach 1:
The patent inverts the traditional processing approach by changing the scanning direction from bottom-to-top to top-to-bottom, and by selecting pixels based on CT value thresholds rather than positional heights. This inversion allows the method to effectively distinguish human body pixels from accessories regardless of their vertical positions, resolving the contradiction between processing simplicity and identification accuracy
Solution Approach 2:
The patent changes the selection parameter from positional height (y-coordinate) to CT value threshold. By selecting pixels whose CT values fall within a specific range (e.g., -200 to 200 HU for human body tissues), the method can accurately identify human body regions even when accessories are present at various heights, thereby improving identification accuracy while maintaining algorithmic simplicity
2Adaptability or versatility
If more accessories are added to support patient postures, then the patient positioning capability is improved, but the difficulty of identifying human body data increases
Solution Approach 1:
The patent applies local quality by assigning different CT value ranges to different tissue types. Human body soft tissues have characteristic CT values (e.g., -200 to 200 HU) that differ from accessories (which may have higher or lower values depending on material). By selecting pixels within the human body CT value range, the method can locally identify human body regions even when surrounded by various accessories, thus maintaining identification capability despite increased positioning versatility
Solution Approach 2:
The patent changes the detection parameter from spatial position to material property (CT value). Since different materials have distinct CT values, this parameter change enables the system to distinguish human body tissues from accessories based on their inherent material properties rather than their positions, thereby maintaining identification accuracy even as the number and types of accessories increase
3Adaptability or versatility
If the scanning field is enlarged to accommodate patient postures, then the positioning flexibility is improved, but the image completeness deteriorates due to cut-off portions
Solution Approach 1:
The patent applies preliminary action by first identifying and selecting pixels that belong to the human body based on CT value thresholds before performing any image reconstruction or processing. This preliminary selection ensures that only relevant human body data is processed, and subsequent reconstruction can focus on completing the human body image without being interfered by accessories or cut-off regions, thereby maintaining image completeness while preserving positioning flexibility
Data Source
AI summary
Embodiments of the present invention provide a method for processing a radiotherapy CT positioning image, comprising: defining a pixel having a CT value greater than or equal to a first threshold in an original CT image as a human body pixel; counting a number of human body pixels of each pixel row in the original CT image in an order from top to bottom; and determining a boundary of the human body according to the counting result.