Vessel Segmentation via Intensity Value Substitution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current segmentation methods for anatomical structures in radiological scans, such as blood vessels, face challenges in accurately delineating the boundaries of thrombus and wall due to low contrast with surrounding tissues, leading to incomplete or inaccurate segmentations, especially in cases like abdominal aortic aneurysms.
Innovation Solution
A method that temporarily replaces intensity values of disturbing anatomical structures within a specified volume with substitute values representative of healthy tissue, allowing for accurate segmentation of the outer boundary, followed by restoration of original values for precise identification of lesions and other components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional segmentation approaches are used, then the segmentation process is simple, but the segmentation accuracy deteriorates due to low contrast between thrombus, wall and surrounding tissue
Solution Approach 1:
The patent applies preliminary action by performing intensity value replacement before the segmentation process. Disturbing anatomical structures are identified and their intensity values are replaced with substitute values that match the surrounding healthy tissue, thereby eliminating contrast interference before segmentation algorithms are applied. This preliminary modification of the image data enables conventional segmentation methods to achieve accurate results without requiring complex specialized algorithms.
2Measurement precision
If intensity values of disturbing structures are not modified, then the original image information is preserved, but the segmentation of the outer boundary becomes inaccurate due to strong responses from neighboring objects
Solution Approach 1:
The patent uses an intermediary approach by introducing substitute intensity values as a temporary mediator. The original intensity values of disturbing structures are replaced with substitute values that match the surrounding healthy tissue, allowing the segmentation algorithm to correctly identify the outer boundary. After segmentation is complete, the original intensity values are restored, ensuring that no permanent information is lost while achieving accurate segmentation during the intermediate processing stage.
3Productivity
If threshold-based approaches are used, then the processing is computationally efficient, but the segmentation fails because the same intensity value is found inside the thrombus and in the neighboring structures
Solution Approach 1:
The patent applies local quality by modifying intensity values specifically in regions containing disturbing anatomical structures, rather than applying uniform processing to the entire image. The substitution of intensity values is performed locally only where needed to eliminate confusion between thrombus and surrounding tissues, allowing threshold-based segmentation to work correctly in affected regions while maintaining processing efficiency.
Data Source
AI summary
A method and Computer Program Product for segmenting a three-dimensional digital representation of an anatomical structure, in which a temporary replacement of intensity values is performed in a specified volume with a substitute intensity value corresponding to healthy anatomical structure. The segmentation of an outer boundary of the blood vessel is performed, before the original intensity values are restored. After restoration of the original intensity values, a thorough segmentation of the specified volume and any lesions contained therein is performed.