Moving Object Tracking Apparatus for Radiotherapy Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radiotherapy systems face challenges in accurately detecting targets under severe fluoroscopic radiation conditions with excessive noise, leading to increased operation loads and misrecognition issues, as existing methods require high radiation intensity to prevent misrecognition and distortion.
Innovation Solution
A moving object tracking apparatus that applies a target emphasizing process to fluoroscopic images using filters to enhance specific frequency components and image regions, reducing the intensity of fluoroscopic radiation while maintaining accurate target detection by calculating candidate positions based on certainty degrees and common perpendicular lines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the intensity of fluoroscopic radiation is increased to prevent misrecognition and distortion of target images, then measurement precision is improved, but use of energy by stationary object increases and operation load increases
Solution Approach 1:
The system performs preliminary actions by acquiring a reference image before the fluoroscopic image, extracting feature amounts from both images, and calculating correlation values in advance. This preprocessing enables accurate target detection even with low-intensity fluoroscopic radiation, resolving the contradiction between measurement precision and energy consumption.
Solution Approach 2:
The patent introduces an intermediary mechanism using reference images and correlation value calculations as mediators between the fluoroscopic radiation detector and target detection. By comparing feature amounts from the reference image with those from the fluoroscopic image through correlation analysis, the system achieves accurate detection without requiring high radiation intensity, thus reducing detector load while maintaining precision.
2Manufacturing precision
If the intensity of fluoroscopic radiation is increased to avoid distortion of projected target image, then manufacturing precision is improved, but loss of energy increases
Solution Approach 1:
The system acquires a reference image beforehand and extracts feature amounts from it. By having this preliminary reference data, the system can accurately detect targets in subsequent fluoroscopic images even when those images have low intensity and potential distortion, thereby maintaining manufacturing precision while reducing radiation energy loss.
Solution Approach 2:
The patent changes the parameter of image analysis by using correlation value calculations between reference images and fluoroscopic images. This parameter change allows the system to achieve accurate target detection without relying on high image intensity, thus maintaining manufacturing precision while reducing the radiation energy required.
3Measurement precision
If template matching is used to detect target positions, then measurement precision is improved, but reliability deteriorates under excessive noise conditions
Solution Approach 1:
The patent introduces reference images as intermediaries in the detection process. By extracting feature amounts from reference images and calculating correlation values with fluoroscopic images, the system creates a more reliable detection mechanism that is less susceptible to noise compared to direct template matching, thereby improving reliability under noisy conditions while maintaining measurement precision.
Solution Approach 2:
The system implements a feedback mechanism by using detected target positions from the correlation analysis to guide subsequent detections. This feedback approach allows the system to adapt to noise conditions and maintain reliable detection, resolving the contradiction between measurement precision and reliability under excessive noise.
Data Source
Figure 1
Figure 2~3
Figure 4~5B
AI summary
The moving object tracking apparatus 104 emphasizes (S902A, S902B) an image with specific size in each of fluoroscopic images A, B derived from two or more paired fluoroscopic radiographic devices 4A, 4B, obtains (S903A, S903B) a value indicating certainty degree of detecting a candidate position of the object 3 to be tracked on the image subjected to the emphasizing process, extracts (S904A, S904B) the candidate position based on the value indicating the certainty degree of detection, calculates (S905; S906) a value indicating a correlation between the candidate position extracted from images picked up from two or more directions, and a position of the fluoroscopic radiation generator, detects (S907; S908) the position of the object to be tracked based on the value indicating the certainty degree of detection, and the value indicating the correlation, and controls irradiation of radiation to an irradiation target based on the detected position of the object to be tracked.