Guidewire Shadow Detection in Intravascular Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intravascular imaging systems face challenges in accurately detecting guidewires within blood vessels due to shadows generated, which can obscure stent visualization and lead to diagnostic errors, particularly when multiple guidewires are present, complicating the identification of shadows, branches, and mergers in arterial images.
Innovation Solution
The method involves generating a carpet view of intravascular data, creating binary masks to detect guidewire shadows, and processing these masks to validate candidate guidewire segments, link them into contiguous representations, and remove spurious segments, ultimately displaying guidewire positions relative to the lumen in a user interface for improved diagnostic accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If guidewires are used to navigate the tortuous path through arteries, then the ability to reach locations of interest for data collection is improved, but guidewire shadows are generated that negatively affect the imaging of stents and other arterial elements
Solution Approach 1:
The patent extracts and separates guidewire shadow detection from the overall image processing pipeline. By independently identifying guidewire shadows through specialized algorithms (carpet view analysis, binary mask generation), the system removes these harmful artifacts from the diagnostic image data, allowing clean visualization of stents and arterial elements without guidewire interference
Solution Approach 2:
The patent converts the harmful guidewire shadows into useful diagnostic information. By detecting and analyzing the shadow patterns, the system not only removes artifacts but also provides enhanced guidewire location data that can be used for procedural guidance, transforming an obstacle into a beneficial feature for both artifact removal and improved guidewire visualization
2Adaptability or versatility
If multiple guidewires are present in the blood vessel, then the ability to perform complex procedures is improved, but the identification of shadows, branches, and mergers becomes more complex and error-prone
Solution Approach 1:
The patent segments the complex task of multi-guidewire analysis into distinct processing stages: carpet view generation, binary mask creation, shadow detection, and contour analysis. This segmented approach allows the system to handle multiple guidewires systematically by processing each shadow independently through the pipeline, reducing computational complexity and improving accuracy compared to attempting to analyze all guidewires simultaneously
Solution Approach 2:
The patent transforms the two-dimensional image data into a carpet view representation that unfolds the circular cross-sections along the longitudinal axis, creating a pseudo-3D view. This dimensional transformation allows multiple guidewire shadows that may overlap in cross-sectional views to be separated and traced independently along their longitudinal paths, making it easier to distinguish between multiple guidewires and their respective shadows
3Reliability
If guidewire shadows are not distinguished from other arterial features, then diagnostic accuracy is maintained, but stent deployment planning and stenosis evaluation are compromised
Solution Approach 1:
The patent introduces an intermediary processing layer between raw image acquisition and diagnostic analysis. The carpet view generation and binary mask creation act as intermediary steps that prepare the data by enhancing guidewire shadow characteristics while suppressing other features. This intermediary processing allows the system to selectively enhance guidewire detection without compromising the integrity of underlying arterial structures for diagnostic purposes
Solution Approach 2:
The patent applies different processing qualities to different regions of the image data. By generating binary masks that specifically highlight guidewire shadow regions while preserving the original image quality in non-shadow areas, the system enables localized analysis where guidewire detection is enhanced in shadow regions while maintaining diagnostic accuracy in arterial tissue regions, allowing simultaneous artifact removal and preserved diagnostic information
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In part, the disclosure relates to methods of guidewire detection in intravascular data sets such as scan lines, frames, images and combinations thereof. Methods of generating one or more indicia of a guidewire in a representation of blood vessel are also features of the disclosure. A carpet view is generated in one embodiment and regions of relatively higher contrast are detected as candidate guidewire regions. In one embodiment, the disclosure relates to selective removal of guidewire segments from a set of intravascular data and the display of a representation of a blood vessel via a user interface. Representations of a guidewire can be toggled on and off in one embodiment.