Dynamic X-ray Collimation Adaptation via Optical Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for collimation in X-ray imaging, particularly in fluoroscopy, face challenges in adapting to dynamic processes, leading to inadequate exposure of necessary body parts and excessive radiation due to incorrect collimator settings, often resulting from human error and a lack of techniques for sequential X-ray images.
Innovation Solution
A method for automatically adapting collimation in dynamic X-ray imaging using AI-based models, which receives criteria data from a library, acquires optical image data, and adjusts the collimation based on both optical and X-ray frame data to ensure optimal exposure and minimize radiation, incorporating explicit and implicit checks to verify the accuracy of the collimation settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the collimator area is made wider to ensure all necessary body parts are visible, then the completeness of anatomical coverage is improved, but unnecessary radiation exposure increases violating the ALARA principle
Solution Approach 1:
The patent implements dynamic collimation adjustment that continuously adapts the collimator settings based on real-time optical image analysis and sequence position. The collimation boundaries are not fixed but dynamically modified frame-by-frame to tightly enclose the region of interest, ensuring complete anatomical coverage while minimizing the irradiated area to comply with ALARA principles
Solution Approach 2:
The system employs feedback mechanisms where optical images are analyzed to detect anatomical landmarks and adjust collimation boundaries accordingly. The AI model continuously monitors the sequence and provides feedback for optimizing collimator settings, ensuring that the collimation area is precisely adapted to the visible anatomy while preventing unnecessary radiation exposure
2Object-affected harmful factors
If the collimator area is reduced to minimize radiation exposure according to ALARA, then radiation dose is decreased, but necessary anatomical structures may become invisible
Solution Approach 1:
The system dynamically adjusts collimation boundaries based on real-time optical image analysis, ensuring that the collimator area is precisely sized to include all necessary anatomical structures. The AI model continuously evaluates the region of interest and adapts the collimation area frame-by-frame, preventing both over-collimation (excessive radiation) and under-collimation (incomplete coverage)
Solution Approach 2:
The patent performs preliminary detection of anatomical landmarks using optical images and AI models before finalizing collimation settings. By pre-identifying the region of interest and necessary anatomical structures, the system can confidently set the collimator area to the minimum required size, ensuring complete coverage while minimizing radiation exposure
3Measurement precision
If manual collimation adjustment is used to ensure accurate positioning, then collimation precision can be improved, but human error and subjectivity increase leading to inconsistent settings
Solution Approach 1:
The system implements automated collimation adjustment using AI models that independently analyze optical images, detect anatomical landmarks, and determine optimal collimation boundaries without human intervention. This self-service approach eliminates human error and subjectivity, providing consistent and reliable collimation settings across all sequences while maintaining high positioning accuracy
Solution Approach 2:
The patent replaces manual mechanical collimation adjustment with an automated digital system based on AI image analysis. The mechanical process of manual boundary setting is substituted with algorithmic detection of anatomical landmarks and automatic calculation of collimation parameters, eliminating human error while maintaining or improving positioning accuracy and consistency
4Reliability
If automated collimation adjustment is implemented to reduce human error, then reliability of collimation settings is improved, but the complexity of the system increases
Solution Approach 1:
The patent employs a multi-functional AI system that performs multiple tasks: optical image analysis, anatomical landmark detection, collimation boundary determination, and real-time sequence monitoring. This universal system handles all collimation-related functions through a single integrated platform, improving reliability while managing complexity through functional consolidation rather than proliferation of separate components
Data Source
AI summary
A method for automatically adapting a collimation for a dynamic X-ray imaging, comprises: receiving criteria data for adapting the collimation, wherein the criteria data include first criteria and second criteria; acquiring optical image data from an examination object; generating an adapted collimation based on the acquired optical image data and the first criteria; acquiring an X-ray frame from the examination object with the adapted collimation; performing an automatic check of the adapted collimation of the acquired X-ray frame by checking whether the adapted collimation meets the second criteria based on the acquired X-ray frame; and generating a re-adapted collimation such that the second criteria are more likely fulfilled by the next acquired X-ray frame in response to the automatic check not being passed, or maintaining the adapted collimation in response to the automatic check being passed.