Vascular Image Classification Without a 3D-DSA Mask Run
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing X-ray imaging methods, such as 3D-DSA, are suboptimal due to motion artifacts, require double the recording time, and involve additional X-ray dose, making it inefficient for capturing contrast-enhanced vascular sections.
Innovation Solution
A method for providing a classified data set by identifying image points with predefined variability in time-intensity curves, allowing efficient classification of vascular sections without a mask run, using medical imaging devices like MRI, CT, or X-ray devices, and employing machine learning for improved identification and graphical representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a mask run is performed before contrast-enhanced imaging, then bone structures do not interfere with visualization, but recording time doubles and additional X-ray dose is applied
Solution Approach 1:
The patent extracts and removes the mask run phase from the conventional 3D-DSA protocol. By using machine learning algorithms to automatically identify and classify vascular sections directly from contrast-enhanced images, the system eliminates the need for separate mask image acquisition, thereby reducing recording time while maintaining visualization quality through intelligent image processing rather than sequential imaging steps
2Reliability
If a mask run is performed before contrast-enhanced imaging, then bone structures do not interfere with visualization, but additional X-ray dose is applied
Solution Approach 1:
The patent removes the unnecessary mask run step that consumes additional X-ray dose. By directly processing contrast-enhanced images through machine learning classification, the system achieves bone structure visualization and vascular section identification without the extra radiation exposure required for separate mask imaging, thereby reducing total X-ray dose while maintaining diagnostic reliability
3Reliability
If conventional 3D-DSA is used, then contrast-enhanced vascular sections can be captured, but motion artifacts occur and recording time is doubled
Solution Approach 1:
The patent implements feedback mechanisms through machine learning algorithms that continuously analyze time-intensity curves of image points to identify and classify vascular sections. This intelligent feedback system automatically adjusts the analysis to distinguish true vascular contrast enhancement from motion artifacts, enabling reliable vascular section capture while minimizing the impact of motion during the single-pass contrast-enhanced imaging
Data Source
AI summary
A method for providing a classified data set includes capturing an image data set of an examination object by a medical imaging device. The image data set has a plurality of image points in each case with a time-intensity curve. The image points map an examination area of the examination object with at least one contrast-enhanced vascular section. The method further includes identifying first image points in the image data set whose time-intensity curves have a predefined variability as image points that map the at least one contrast-enhanced vascular section, and providing the classified data set based on the image data set and the first image points, wherein the classified data set has a classification between the first image points and further image points of the image data set.