X-ray Imaging Geometry Correction via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile X-ray imaging systems face challenges in achieving correct imaging geometry due to the lack of rigid mechanical coupling between the X-ray source and detector, leading to high patient and personnel doses, as well as suboptimal image throughput.
Innovation Solution
A system utilizing a pre-trained or trainable machine learning module to compute personalized correction information for adjusting the imaging geometry of mobile X-ray imaging apparatuses, coupled with a modulator that provides user instructions based on this information to achieve a target imaging geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If mobile X-ray imaging apparatus is used to enable imaging under awkward conditions, then accessibility and flexibility are improved, but X-ray dosage to patients and personnel increases
Solution Approach 1:
The system uses feedback from scout images and collimator settings to automatically adjust imaging geometry parameters, enabling the mobile X-ray apparatus to achieve correct imaging geometry without requiring repeated high-dose exposures. The automatic adjustment mechanism learns from previous imaging attempts and refines geometry parameters to minimize retakes and reduce cumulative radiation dosage.
Solution Approach 2:
The system performs preliminary low-dose scout imaging and geometry calculation before the actual diagnostic exposure. By pre-determining the optimal imaging geometry using minimal radiation, the system avoids the need for multiple high-dose retakes, thereby reducing the overall X-ray dosage while maintaining imaging accessibility in mobile settings.
2Device complexity
If manual adjustment of imaging geometry is used in mobile X-ray apparatus, then device simplicity is maintained, but image throughput decreases due to repeated adjustments and retakes
Solution Approach 1:
The system performs self-adjustment of imaging geometry by automatically calculating correction parameters from scout images and executing the adjustments without requiring manual intervention. This self-service capability eliminates the time-consuming iterative process of manual adjustment and retakes, thereby significantly improving image throughput while maintaining the relative simplicity of the mobile apparatus.
Solution Approach 2:
The system replaces manual mechanical adjustment with automated computational geometry correction. By using image processing algorithms and automatic parameter calculation instead of manual mechanical positioning, the system reduces the time required for geometry adjustment and increases imaging productivity while keeping the physical device structure relatively simple.
3Measurement precision
If multiple adjustment iterations are performed to achieve correct imaging geometry, then imaging accuracy is improved, but time consumption increases
Solution Approach 1:
The system performs preliminary geometry calculation and correction parameter determination using low-dose scout images before the actual diagnostic imaging. By pre-establishing the correct imaging geometry through computational methods, the system achieves high imaging accuracy in a single setup rather than requiring multiple iterative adjustments, thereby reducing time consumption.
Solution Approach 2:
The system introduces an intermediary computational step that processes scout images and calculates the optimal imaging geometry parameters. This intermediary calculation layer acts as a mediator between the rough initial setup and the final diagnostic imaging, enabling accurate geometry establishment without requiring multiple time-consuming adjustment iterations.
Data Source
AI summary
System (SYS) for supporting X-ray imaging and related methods. The system (SYS) comprises a machine learning module (MLM), a logic (LG) configured to compute output correction information for adjusting an imaging geometry of an X-ray imaging apparatus to achieve a target imaging geometry. A modulator (MOD,L-MOD, H-MOD, S-MOD) is the system is configured to provide a user instruction for imaging geometry adjustment. The user instruction is modulated based on the output correction information. The machine learning module was previously trained on training data including a specific user's responses to previous instructions.