Vascular Image Registration Using Intravascular Trajectory Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-invasive vascular imaging, such as CTA, suffers from high inter-user variability in manual delineation of lumen and plaques due to limited resolution and imaging artifacts, hindering effective training and validation of AI/ML models for medical imaging analysis.
Innovation Solution
The registration of intravascular images, such as IVUS and OCT, with non-invasive vascular images like CTA is achieved by determining a trajectory using a cost function that measures similarity and optimizing rotation angles, allowing for the construction of a 3D mesh and angular registration, thereby providing accurate ground truth for training AI/ML models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual delineation is used to annotate CTA images, then training data can be obtained, but high inter-user variability occurs due to limited resolution and imaging artifacts
Solution Approach 1:
The patent introduces intravascular images (IVUS/OCT) as an intermediary reference standard with high-resolution ground truth. These intravascular images serve as a mediator between the non-invasive CTA images and the desired accurate delineation, allowing automated algorithms to learn from the high-quality reference data and produce consistent, reliable annotations that reduce inter-user variability.
Solution Approach 2:
The patent replaces manual mechanical delineation (human annotators visually interpreting low-resolution CTA images) with an automated computational system. The system uses machine learning algorithms trained on intravascular reference images to automatically segment and annotate CTA images, eliminating human variability and improving both precision and reliability of measurements.
2Measurement precision
If intravascular images are registered with non-invasive vascular images, then accurate ground truth is obtained, but complex registration algorithms are required
Solution Approach 1:
The registration process is segmented into distinct functional components: (1) extraction of vessel centerlines and contours from both intravascular and non-invasive images, (2) matching of corresponding anatomical landmarks between the two image types, (3) computation of transformation parameters, and (4) application of registration to align the images. This segmentation makes the complex overall process more manageable and implementable.
Solution Approach 2:
The system performs preliminary processing of both image types before registration, including extraction of key geometric features (centerlines, contours, landmark points) and preparation of matching data. This preliminary action simplifies the subsequent registration step by reducing the complexity of directly aligning the complete image datasets.
3Reliability
If automated algorithms are used for image analysis, then consistency is improved, but training data quality depends on accurate intravascular imaging
Solution Approach 1:
High-quality intravascular images (IVUS/OCT) are acquired as a preliminary step before any automated analysis of non-invasive CTA images. These intravascular images serve as pre-processed reference ground truth that encapsulates accurate vessel morphology and plaque characteristics. By obtaining this high-accuracy reference data beforehand, the system ensures that subsequent automated algorithms are trained on reliable data, maintaining both consistency and precision.
Data Source
AI summary
Systems and methods for determining a trajectory of one or more intravascular images in one or more non-invasive vascular images are provided. 1) one or more intravascular images of a vessel of a patient and 2) one or more non-invasive vascular images of the vessel of the patient are received. A trajectory of the one or more intravascular images in the one or more non-invasive vascular images is determined. The trajectory is output.