Real-time X-ray Collimator Adjustment via Landmark Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current X-ray imaging systems in interventional procedures face challenges in automatically and efficiently setting collimation and region-of-interest (ROI) filters to minimize radiation exposure to medical staff, as manual adjustments are labor-intensive and prone to errors, while existing solutions like eye-trackers are not mature enough for clinical use.
Innovation Solution
A non-transitory program storage device executes methods for real-time collimation and ROI-filter positioning by acquiring images, classifying procedures, detecting landmarks, loading pre-trained models, and recomputing collimator settings based on the presence and movement of anatomical structures and medical devices, using low-level features like edges and Haar-like features, and employing a 'tag-and-learn' method for initializing new landmarks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If manual adjustment of collimator and ROI-filter is used, then radiation exposure to medical staff is minimized, but labor intensity increases and procedural duration increases
Solution Approach 1:
The system automatically detects landmarks (anatomical structures and medical devices) in real-time X-ray images and autonomously adjusts collimator and ROI-filter settings without requiring manual intervention. The computer executes algorithms that identify relevant structures and compute optimal collimation parameters, enabling the system to serve itself rather than requiring continuous operator input.
Solution Approach 2:
The patent replaces manual mechanical adjustment of collimator and ROI-filter with an automated computational system. Instead of operators physically moving components based on visual assessment, a computer-based image processing system analyzes X-ray images, detects landmarks using algorithms, and automatically computes and applies optimal collimation settings, substituting mechanical manual operation with automated digital control.
2Object-affected harmful factors
If manual adjustment of collimator and ROI-filter is used, then radiation exposure to medical staff is minimized, but procedural duration increases
Solution Approach 1:
The system continuously monitors X-ray images throughout the procedure and maintains optimal collimator and ROI-filter settings in real-time. Rather than requiring periodic manual re-adjustment, the automated system continuously detects landmarks and updates collimation parameters as needed, ensuring continuous radiation protection without interrupting the procedural workflow.
Solution Approach 2:
The automated landmark detection and collimation adjustment system operates autonomously throughout the procedure, detecting anatomical structures and medical devices in real-time and automatically updating settings without requiring operator intervention. This self-service capability eliminates time loss associated with manual re-adjustment during procedural changes.
3Extent of automation
If automatic landmark detection is implemented, then real-time optimization of collimator settings is achieved, but device complexity increases
Solution Approach 1:
The patent introduces a computer-based image processing system as an intermediary between the X-ray imaging system and the collimator/ROI-filter control. This intermediary executes algorithms that detect landmarks in images and compute optimal collimation settings, serving as a intelligent mediator that translates visual information into automated control commands without requiring direct complex integration between imaging and collimation subsystems.
Solution Approach 2:
The system replaces complex manual mechanical adjustment mechanisms with a computational approach. Instead of designing mechanically complex automated adjustment devices, the patent uses software-based landmark detection and algorithmic computation of collimation parameters, substituting mechanical complexity with information processing that can be implemented through standard computer systems.
Data Source
Figure 1(a)~3
Figure 2
Figure 4(a)~5
AI summary
A method for real-time collimation and ROI-filter positioning in X-ray imaging in interventional procedures includes acquiring an image of a region-of-interest (ROI) at a beginning of a medical intervention procedure on a subject, classifying the image based on low-level features in the image to determine a type of procedure being performed, determining a list of landmarks in the image from the type of procedure being performed, and loading a pre-trained landmark model for each landmark in the list of landmarks, where landmarks include anatomical structures of the subject and medical devices being used in the medical intervention procedure, and computing collimator settings of an X-ray imaging device from ROI filter margins and bounding boxes of the landmarks calculated using the landmark models.