Adaptive X-Ray Collimation With Machine Learning for Interventional Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current X-ray imaging systems require cumbersome and time-consuming manual adjustments of the collimator, which are often not adaptable to changing imaging geometries, leading to increased radiation exposure and inefficient use of the field of view.
Innovation Solution
A system utilizing a machine learning model to estimate and adjust collimator settings based on user input and acquired images, allowing for rapid and precise collimator adjustments, including a collimator setting estimator and a segmentor to process feature maps for robust collimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual collimator adjustment is used, then the user can control the field of view, but the adjustment process is cumbersome and time-consuming
Solution Approach 1:
The system enables automatic collimator adjustment by using the imager to capture images and the processor to analyze them, automatically determining optimal collimator settings without requiring manual user input for each adjustment. The system serves itself by using its own imaging capabilities to guide the collimation process.
Solution Approach 2:
The patent replaces manual mechanical adjustment of collimator components with an automated computational system. The processor analyzes images and automatically controls the collimator settings, substituting the mechanical manual adjustment process with an automated image-processing-based system.
2Object-affected harmful factors
If tight collimation is applied to focus on the current ROI, then radiation exposure is reduced, but the collimator settings become unusable when imaging geometry changes
Solution Approach 1:
The system dynamically adjusts collimator settings based on changing imaging geometries. Instead of using fixed tight collimation, the system continuously monitors imaging geometry changes and automatically recalibrates collimator settings to maintain optimal FOV while adapting to new geometric configurations.
Solution Approach 2:
The system uses feedback from image analysis to continuously optimize collimator settings. The processor analyzes captured images to determine whether the current collimation is appropriate, and automatically adjusts settings based on this feedback, creating a closed-loop system that adapts to changing conditions.
3Measurement precision
If multiple collimator components are adjusted manually, then precise field of view control is achieved, but the complexity of operation increases
Solution Approach 1:
The system merges the control of multiple collimator components into a single automated process. Instead of requiring separate manual adjustment of multiple shutters and wedges, the processor analyzes the image and automatically coordinates all collimator components together as an integrated system.
Solution Approach 2:
The imager serves multiple functions: it captures images for diagnostic purposes and simultaneously provides the data needed for automatic collimator adjustment. This multi-functionality eliminates the need for separate control mechanisms, as the imaging system itself guides the collimation process.
4Adaptability or versatility
If imaging geometry is changed to view different ROIs, then comprehensive imaging is achieved, but procedure duration increases
Solution Approach 1:
The system performs preliminary analysis of captured images to pre-determine optimal collimator settings before actual imaging geometry changes are needed. By analyzing images in advance and preparing collimation parameters beforehand, the system reduces the time required when geometry changes are actually executed.
Data Source
AI summary
A system (SYS) and related method for facilitating collimator adjustment in X-ray imaging or a radiation therapy delivery. The system comprises an input interface (IN) for receiving input data including i) an input image and/or ii) user input data including a partial collimator setting for a collimator (COL) of an X-ray imaging apparatus (IA). A collimator setting estimator (CSE) of the system computes a complemented collimator setting for the collimator based the input data. Preferably, the system uses machine learning.