Automatic Intravascular Pullback Alignment Using Junction-Point Offsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual alignment of intravascular data representations from pre- and post-intervention pullbacks can lead to errors due to misidentification of corresponding points, which is a challenge in aligning intravascular imaging data.
Innovation Solution
An automated method and device using processors to align intravascular data representations by determining a junction point, calculating distance offsets, and horizontally offsetting end points to vertically align representations, utilizing extraluminal images and AI masks for precise alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual alignment of intravascular data representations is performed, then user control and flexibility are maintained, but alignment accuracy deteriorates due to misidentification of corresponding points
Solution Approach 1:
The patent replaces the manual mechanical alignment process with an automated image processing and pattern recognition system. The system uses extraluminal images to automatically identify corresponding anatomical landmarks and compute alignment transformations, eliminating the need for manual point identification and significantly improving alignment accuracy while reducing operational complexity
Solution Approach 2:
The patent introduces extraluminal images as an intermediary element that bridges the two intravascular data sets. These extraluminal images serve as a common reference frame that contains identifiable anatomical landmarks, allowing the system to establish correspondences between pre- and post-intervention pullbacks without requiring direct manual comparison of the intravascular data themselves
2Productivity
If automated alignment using extraluminal images and AI masks is implemented, then alignment accuracy and productivity are improved, but device complexity increases
Solution Approach 1:
The system performs preliminary processing of extraluminal images to generate AI masks that pre-identify anatomical landmarks and structures before the actual alignment operation. This preliminary action creates ready-to-use reference data that accelerates the alignment process and reduces the computational complexity during the actual alignment execution
Solution Approach 2:
The patent segments the alignment task into distinct modular components: extraluminal image processing, AI mask generation, landmark detection, transformation computation, and final alignment application. This segmentation allows each component to be independently optimized and processed, improving overall productivity while managing system complexity through modular architecture
Data Source
Figure 1
Figure 1B
Figure 2A~2B
AI summary
The present disclosure provides systems and methods for automatically aligning intravascular data taken during a plurality of pullbacks in a target blood vessel. Extraluminal images may be taken to determine the location of the guide catheter in the target vessel. Two or more pullbacks may then be performed in the target vessel. The start and end point of each pullback may be determined. A distance between the end point of each pullback and the proximal tip, or junction point, of the guide catheter may be determined. A difference between the end point of each pullback and the junction point may be determined. The difference between the distances from the end of the pullback and the junction point may correspond to the distance to offset one of the representations of the pullbacks in order to automatically align the representation of the pullbacks.