Tissue Boundary Determination Using Dual-Energy CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In medical imaging, particularly in radiation therapy, accurately determining tumor boundaries without contrast agents is challenging due to unclear and variable tumor boundaries, which affects treatment planning and evaluation.
Innovation Solution
A tissue boundary determination apparatus and method that uses CT spectral imaging to acquire multiple slice images, performs feature extraction on each pixel, initializes a seed point, and conducts region expansion based on pre-configured criteria to accurately determine tumor boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If contrast agents are used to highlight tumor tissue, then tumor boundary visibility is improved, but treatment evaluation accuracy deteriorates due to uncontrollable contrast agent accumulation
Solution Approach 1:
The patent extracts and removes the contrast agent dependency from the tumor boundary detection process. By using dual-energy CT imaging technology, the system directly captures tumor boundary information from the patient's body without requiring external contrast agents, thereby eliminating the problem of uncontrollable contrast agent accumulation while maintaining clear boundary visibility
Solution Approach 2:
The patent introduces dual-energy CT imaging technology as an intermediary between the X-ray source and the tumor tissue. This intermediary technology enables direct visualization of tumor boundaries by exploiting the differential attenuation properties of tissues at different energy levels, replacing the need for contrast agents as mediators
2Difficulty of detecting and measuring
If spectral imaging is used to enable tissue examination, then tumor detection capability is improved, but processing load increases and operation efficiency deteriorates
Solution Approach 1:
The patent extracts only the essential tumor boundary information from the dual-energy CT data through intelligent algorithms, rather than processing all spectral imaging data. This selective extraction approach maintains high tumor detection capability while significantly reducing processing load and improving operation efficiency
Solution Approach 2:
The patent segments the complex spectral imaging processing into distinct modules: data acquisition, preliminary processing, tumor boundary detection, and result generation. This segmentation allows for optimized processing at each stage and enables parallel computation, thereby reducing overall processing time and improving efficiency
3Extent of automation
If conventional AI technologies are used to process tumor boundaries, then automation is improved, but boundary determination accuracy deteriorates due to unclear and variable tumor characteristics
Solution Approach 1:
The patent changes the parameter space for tumor boundary detection by using dual-energy CT values and their ratios as input features, rather than conventional single-energy CT values. This parameter transformation enhances the contrast between tumor and normal tissue, providing more discriminative features for AI algorithms and improving boundary determination accuracy while maintaining automation
Solution Approach 2:
The patent combines multiple imaging parameters and AI algorithms into a composite detection system. By integrating dual-energy CT data, region expansion algorithms, and intelligent boundary detection methods, the system achieves superior boundary determination accuracy that overcomes the limitations of individual conventional AI approaches
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 enables efficient and accurate acquisition of tumor boundaries, providing a reliable reference for analyzing internal and external tumor tissues, thus improving treatment planning and evaluation.
Implementation Method 1
multiple slice images are spectral images acquired by performing CT scanning on the multiple positions of the tissue to be examined
Implementation Method 2
CT spectral imaging may allow the doctor to gain tissue examination capabilities
Data Source
AI summary
Embodiments of the present application provide a tissue boundary determination apparatus and method. The tissue boundary determination method includes acquiring multiple slice images of multiple positions of a tissue to be examined, determining, with regard to at least one first selected slice image of the multiple slice images, a tissue boundary on the first selected slice image, and determining, according to the tissue boundary on the first selected slice image, tissue boundaries on a predetermined number of other slice images adjacent to the first selected slice image. Therefore, an accurate tissue (such as malignant tumor) boundary may be acquired efficiently so as to provide a reference basis for analysis on internal and external tissue of a tumor.