Displacement Calculation via Image Subset Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radiation therapy accuracy is compromised due to respiratory motion of the patient body, leading to challenges in tracking and measuring the displacement of objects of interest within the body, resulting in inaccuracy and high computation complexity in current tracking methods.
Innovation Solution
A method that calculates a displacement model from pre-acquired images, identifies subsets of images with similar similarity levels, and selects reference images to determine displacement between them, reducing computation complexity and improving accuracy in tracking the object's position during image acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If markers are put on the skin and tracked using ultrasound imaging, then object tracking is enabled, but computation complexity becomes high
Solution Approach 1:
The patent segments the large set of pre-acquired images into multiple subsets based on temporal intervals. Instead of comparing each newly-acquired image with all pre-acquired images, the method divides the search space into manageable segments, significantly reducing the computational burden while maintaining tracking accuracy.
Solution Approach 2:
The patent performs preliminary organization of pre-acquired images into subsets before actual tracking occurs. By pre-segmenting the image set and establishing temporal relationships in advance, the system avoids computationally expensive real-time processing during tracking operations.
2Measurement precision
If all pre-acquired images are processed for each newly-acquired image, then accurate displacement calculation is achieved, but processing time increases
Solution Approach 1:
The patent divides the complete set of pre-acquired images into multiple subsets based on temporal intervals. Each subset contains a limited number of images that are temporally close to each other. This segmentation allows the system to process only relevant subsets for each newly-acquired image rather than processing the entire image set, thereby reducing processing time while maintaining accuracy.
Solution Approach 2:
The patent applies partial action by processing only the necessary subsets of pre-acquired images that are temporally relevant to each newly-acquired image, rather than processing all pre-acquired images. This selective processing approach maintains sufficient accuracy for displacement calculation while significantly reducing the computational workload and processing time.
3Productivity
If image registration is performed for real-time tracking, then tracking speed is improved, but computation complexity increases
Solution Approach 1:
The patent segments the image registration process by dividing pre-acquired images into temporal subsets. This segmentation enables real-time tracking performance by limiting the registration computation to only the relevant subset of images rather than the entire image set, thus improving tracking speed while controlling computation complexity.
Solution Approach 2:
The patent performs preliminary segmentation and organization of images into temporal subsets before real-time tracking begins. This pre-processing step establishes the structure needed for efficient real-time registration, allowing the system to achieve real-time tracking speed without the full computational burden of processing all images during tracking operations.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The invention relates to a method of calculating a displacement of an object of interest comprising a step of calculating (101) a displacement model of said object of interest from adjacent images of a set of pre-acquired images of said object of interest, said displacement model reflects the position of said object of interest along the time. The method is characterized in that the method further comprises the following. A step of determining (102) a first sub-set of images (S1) from said set of pre-acquired images within one periodical time cycle of said set of pre-acquired images on the basis of the displacement model. A first step of identifying (103) a second sub-set of images (S2) from newly-acquired images, wherein images in said second sub-set of images (S2) are consecutive and have the same most similar image in said first sub-set of images (S1), wherein a first set of similarity levels is determined by comparing a given image in said newly acquired images with each image of said first sub-set of images (S1), and wherein said most similar image has the largest similarity level in said first set of similarity levels. A first step of selecting (104) a given image in said second sub-set of images (S2) as a first reference image (I1). A second step of identifying (105) a third sub-set of images (S3) from said newly-acquired images, wherein images in said third sub-set of images (S3) are consecutive and have the same most similar image in said first sub-set of images (S1), wherein a set of similarity levels is determined by comparing a given image in said newly acquired images with each image of said first sub-set of images (S1), and wherein said most similar image has the largest similarity level in said set of similarity levels. A second step of selecting (106) a given image in said third sub-set of images (S3) as a second reference image (I2). A step of calculating (107) the displacement between said second reference image (I2) and said first reference image (I1). The invention also relates to a corresponding system of displacement calculation.