Real-Time Co-Registration of Intravascular and Extravascular Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current co-registration techniques for angiography and intravascular ultrasound (IVUS) images are typically performed offline, limiting real-time insights during medical procedures and hindering dynamic decision-making.
Innovation Solution
A system configured to co-register angiography and IVUS images in real-time, allowing for immediate feedback and enabling clinicians to observe dynamic changes in vessel anatomy and plaque characteristics during procedures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If co-registration is performed offline after imaging procedures, then processing accuracy can be improved through detailed analysis, but real-time feedback and dynamic decision-making are hindered
Solution Approach 1:
The system performs preliminary actions by pre-processing and establishing coordinate transformations between CTA and IVUS image spaces before the actual co-registration is needed. This allows the system to rapidly execute co-registration when real-time feedback is required, resolving the contradiction between processing accuracy and real-time performance
Solution Approach 2:
The system implements dynamic co-registration that can adapt between offline detailed processing and online rapid processing modes. The coordinate transformation framework allows flexible adjustment of processing depth and speed based on whether real-time feedback or maximum accuracy is the priority
2Area of stationary object
If multiple image frames are co-registered manually, then comprehensive vessel anatomy coverage is achieved, but procedural complexity and time consumption increase
Solution Approach 1:
The system segments the co-registration task by identifying and matching key anatomical landmarks (such as side branches) across multiple image frames. This breaks down the complex full-frame co-registration into manageable landmark-matching sub-tasks, reducing overall complexity while maintaining comprehensive vessel coverage
Solution Approach 2:
The system creates simplified representations (copies) of vessel anatomy through centerline extraction and landmark identification from multiple frames. These copied features are then used for efficient co-registration without requiring processing of complete high-resolution image data, reducing computational complexity
3Productivity
If real-time co-registration is implemented, then immediate feedback and dynamic decision-making are enabled, but processing speed requirements increase system complexity
Solution Approach 1:
The system extracts only the essential features needed for co-registration (landmarks, centerlines, key vessel segments) from complete image frames. This extraction approach enables real-time processing by reducing data volume while maintaining the accuracy needed for clinical decision-making
Solution Approach 2:
The system replaces traditional mechanical/manual co-registration methods with automated computational algorithms. This substitution enables real-time processing speeds unattainable through manual operations, though it introduces computational complexity that is managed through efficient algorithm design
Data Source
AI summary
The present disclosure provides to co-register intravascular images of a vessel with one or more extravascular images of the vessel in real-time, such as, during acquisition of the extravascular images. Key points in a first frame of the extravascular images can be identified and then tracked across other frames of the extravascular images. The intravascular images can be co-registered to a frame (or frames) of the extravascular images based on the locations of the key points in the frame.