Bow-tie Filter Removal in Radiation Therapy Bed Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In radiation therapy, accurately removing the bow-tie filter from X-ray images for high-precision 3D/2D registration is challenging due to its capture in various X-ray image conditions, leading to errors and prolonged treatment throughput when manually detached.
Innovation Solution
A bed positioning system that captures X-ray transparent and CT images using a rotating X-ray tube and detector, generates subtraction images to correct for the bow-tie filter, and performs 3D/2D registration without pre-prepared correction data, enabling accurate positioning by averaging images in the CT rotation axis direction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the bow-tie filter is detached from the image capturing system to accurately perform 3D/2D registration, then positioning accuracy is improved, but treatment throughput deteriorates due to the time-consuming detachment process
Solution Approach 1:
The invention extracts and removes the bow-tie filter structure from the captured X-ray transparent image through image processing techniques. By generating a subtraction image that isolates and eliminates the bow-tie filter components, the system achieves accurate 3D/2D registration without physically detaching the filter, thus maintaining both positioning accuracy and treatment throughput
Solution Approach 2:
The invention creates a digitally reconstructed radiograph (DRR) that replicates the expected X-ray image without the bow-tie filter interference. By comparing the actual captured image with this digital copy, the system can accurately register positions without the physical presence or removal of the filter
2Productivity
If the bow-tie filter is left attached in the image capturing system, then treatment throughput is maintained, but positioning accuracy deteriorates due to errors in 3D/2D registration
Solution Approach 1:
The invention converts the harmful effect of the bow-tie filter structure appearing in captured images into a benefit by using image subtraction techniques. The system generates a correction image that specifically targets and removes the filter's contribution, transforming what was previously a source of registration error into an opportunity for enhanced positioning accuracy while maintaining continuous filter attachment
3Measurement precision
If manual detachment of the bow-tie filter is performed, then positioning accuracy is improved, but treatment time increases
Solution Approach 1:
The invention replaces the mechanical process of physically detaching the bow-tie filter with an automated image processing system. By using computational methods to generate subtraction images and remove filter artifacts digitally, the system achieves the same positioning accuracy as manual detachment would provide, but without the time-consuming physical intervention
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 method allows for high-accuracy bed positioning by removing the bow-tie filter from images used in 3D/2D registration without pre-prepared correction data, enhancing treatment efficiency and accuracy.
Implementation Method 1
capable of capturing X-ray transparent images and CT images by rotating an X-ray tube and an X-ray detector around a subject on a bed
Implementation Method 2
generates a subtraction image from a first X-ray transparent image captured by the X-ray tube and the X-ray detector and a digitally reconstructed radiograph generated from a treatment plan CT image
Data Source
AI summary
Provided is a bed positioning system 100 for a radiation therapy system 3 capable of capturing X-ray transparent images 200 and CT images by rotating an X-ray tube 4 and an X-ray detector 5 around a subject 1 on a bed 2, comprising: a transparent image registration system 12 which generates a subtraction image 220 from a first X-ray transparent image 200 captured by the X-ray tube 4 and the X-ray detector 5 and a digitally reconstructed radiograph 210 generated from a treatment plan CT image, corrects the first X-ray transparent image 200 by use of a correction image 230 generated by processing the subtraction image 220 in a previously specified direction, and compares the first X-ray transparent image 200 after the correction and the digitally reconstructed radiograph 210 to determine a movement amount of the bed 2.