Vascular Contour Overlay in Fluoroscopy via Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical imaging methods face challenges in clearly representing vascular structures in X-ray-based fluoroscopic images without repeated administration of contrast agents, as contrast agents dilute rapidly and are difficult to maintain visibility of both vascular structures and surrounding anatomy simultaneously.
Innovation Solution
A method involving capturing a mask image without contrast, injecting contrast to record filling images at different times, subtracting these to generate a vessel image, segmenting vessels using a threshold value based on structural features, and superimposing the segmented vessels onto a live X-ray image to maintain visibility without repeated contrast administration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If contrast agent is repeatedly injected to maintain visibility of vascular structures, then vascular structures remain visible in fluoroscopic images, but image noise increases and dynamic range is reduced
Solution Approach 1:
The patent segments the vascular tree into meaningful sections based on structural features (e.g., vessel diameter, branching patterns) and selectively processes only those sections that meet predetermined criteria. This segmentation allows the system to maintain vascular visibility without processing the entire image, thereby reducing image noise and preserving dynamic range while still providing reliable vascular structure visualization.
Solution Approach 2:
The patent applies different processing qualities to different regions of the image. Specifically, it enhances vascular structures in regions that meet structural feature thresholds while leaving other regions with lower enhancement. This local quality approach ensures that vascular visibility is maintained where needed without uniformly increasing image noise across the entire fluoroscopic image.
2Reliability
If contrast agent is repeatedly injected to maintain visibility of vascular structures, then vascular structures remain visible in fluoroscopic images, but the complexity of the imaging system increases
Solution Approach 1:
The patent implements automated image processing that independently identifies vascular structures, segments them based on structural features, and enhances them without requiring repeated manual contrast agent injection. The system performs self-service by automatically adapting the enhancement process to the specific anatomical structures present, reducing the need for repeated contrast administration and simplifying the overall imaging workflow.
Solution Approach 2:
The patent performs preliminary segmentation and identification of vascular structures before the actual enhancement process. By pre-identifying which vascular sections meet the structural feature criteria, the system prepares the image data in advance, allowing for more efficient processing and reducing the need for repeated contrast agent injection during the procedure, thereby reducing system complexity.
3Loss of information
If the entire vascular tree is displayed in fluoroscopic images, then complete vascular information is provided, but the ability to navigate to specific regions of interest is reduced
Solution Approach 1:
The patent segments the vascular tree into distinct sections based on structural features and spatial relationships. This segmentation allows the system to selectively display and enhance specific vascular sections that are relevant to the region of interest while maintaining the option to access other segments. The segmented approach preserves complete vascular information in the underlying data structure while presenting only the relevant portions during navigation, thereby improving ease of operation.
Solution Approach 2:
The patent applies enhanced visualization quality locally to specific vascular sections that are identified as regions of interest, while using standard visualization for other areas. This local quality enhancement allows operators to easily navigate to and focus on specific vascular regions without being overwhelmed by the entire vascular tree, improving operational ease while preserving access to complete vascular information when needed.
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
This approach allows for persistent representation of vascular trees without repeated contrast agent injection, reducing image noise and maintaining dynamic range, while improving navigation by clearly delineating vessel contours and limiting the display to regions of interest.
Implementation Method 1
X-ray imaging methods can be used to locate the object
Implementation Method 2
a contrast agent, for example iodine dioxide or carbon dioxide, is administered to the patient—often repeatedly—which allows the vascular system to be made visible in the X-ray images
Data Source
AI summary
Medical imaging systems and methods for determining contours of anatomical structures of vessels and vessel sections for representation in X-ray-based fluoroscopic images. Methods can include capturing a mask image, recording at least one filling image including a vascular tree filled with contrast agent, subtracting the mask image from the at least one filling image to produce at least one vessel image, generating a segmented vessel image by segmenting the vascular tree filled with contrast agent in the at least one vessel image and removing from the at least one segmented vessel image such vessels and vessel sections that do not meet a threshold value, wherein the threshold value represents a function of at least one predetermined structural feature, and superimposing the at least one segmented vessel image with a live X-ray image and reproduction on a display device.