X-ray Dose Adjustment via Region-Specific ROI Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current x-ray imaging technologies lack the ability to optimally adjust x-ray doses based on specific regions of interest within an image, often resulting in inadequate visibility of desired objects due to overall or weighted regional measurements that disregard the user's focus.
Innovation Solution
A method and system that utilize machine learning processes to identify and select regions of interest (ROIs) within an x-ray image, allowing users to adjust the x-ray dose based on specific measurement fields of pixels, using interfaces like mice, joysticks, or voice control, and indicating selected ROIs on the image for precise dose adjustment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If overall or weighted regional measurement is used to determine x-ray dose, then the dosing process is simplified, but the visibility of specific objects of interest is compromised
Solution Approach 1:
The patent divides the image into multiple regions of interest (ROIs) based on image content analysis, allowing separate dose optimization for each region. The system identifies different anatomical structures and objects in the image, creates distinct ROIs for each, and enables independent dose adjustment for each region, thereby resolving the contradiction between simplified dosing and precise object visibility.
Solution Approach 2:
The patent implements local quality by allowing different x-ray dose levels for different regions of interest within the same image. Instead of applying a uniform dose to the entire image, the system calculates and applies region-specific doses based on the unique characteristics and importance of each ROI, ensuring optimal visibility for each specific object while maintaining operational efficiency through automated region identification.
2Ease of manufacture
If fixed regions of interest are used for dose adjustment, then the dosing process is standardized, but the regions may not correspond to actual patient anatomy or objects of interest
Solution Approach 1:
The patent transforms fixed, static ROIs into dynamic, adaptive regions that automatically adjust based on the actual image content. The system uses image analysis algorithms to identify anatomical structures and objects of interest, then dynamically defines ROIs that conform to the actual patient anatomy and clinical context, eliminating the mismatch between fixed ROIs and real-world structures while maintaining standardized dosing protocols.
Solution Approach 2:
The system performs self-service by automatically analyzing the image content, identifying anatomical structures and objects of interest, and generating appropriate ROIs without requiring manual intervention. This automated ROI generation ensures that the regions correspond accurately to actual patient anatomy while maintaining the benefits of standardized dosing procedures, resolving the contradiction between standardization and anatomical accuracy.
3Productivity
If a single x-ray dose is applied to the entire image, then the imaging process is efficient, but the dose may be too low or higher than necessary for specific sections
Solution Approach 1:
The patent segments the image into multiple regions of interest, each with its own optimized dose level. By dividing the image into distinct ROIs based on anatomical structures and objects of interest, the system can apply different dose levels to different regions, ensuring that each section receives the precise dose needed while maintaining overall imaging efficiency through automated processing.
Solution Approach 2:
The patent implements parameter changes by adjusting the x-ray dose parameter locally for different regions of interest. Instead of using a single uniform dose for the entire image, the system varies the dose parameter across different ROIs based on their specific requirements, thereby achieving both imaging efficiency and dose precision for each specific section.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables optimal x-ray dose adjustment for improved visibility of selected objects, enhancing image quality by tailoring the dose to the specific region of interest, rather than relying on arbitrary or anatomically irrelevant fixed regions.
Implementation Method 1
an x-ray tube that irradiates a patient with x-rays and obtains an x-ray image of the patient
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A dose adjustment method according to an embodiment includes irradiating a patient (P) with x-rays and obtaining an image (300) of the patient (P), creating a list of regions of interest (ROIs) related to the image (300) based on the image (300), creating a list of measurement fields of pixels corresponding to the created list of ROIs, providing to a user an interface (600) for selecting a measurement field from the created list of measurement fields, determining an x-ray dose of x-rays to be irradiated the patient (P), based on the selected measurement field, and irradiating the patient (P) with the determined x-ray dose.