Registered Geometry Reconstruction for Complete Cardiac Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing catheter-based mapping methods for heart chambers may miss critical anatomical features like pulmonary veins, which are visible in fluoroscopy, leading to incomplete 3D anatomical maps.
Innovation Solution
A trained artificial neural network (ANN) generates a 3D anatomical map from two 2D fluoroscopic images, such as anterior-posterior and left anterior-oblique projections, and refines it using catheter-based mapping to ensure accuracy, leveraging magnetic and/or impedance-based location tracking for registration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If catheter-based mapping methods are used to map heart chambers, then electrical activity can be sensed and measured at multiple points, but critical anatomical features like pulmonary veins may be missed leading to incomplete 3D anatomical maps
Solution Approach 1:
The patent combines fluoroscopic imaging data with catheter-based mapping data to create a comprehensive 3D anatomical map. The fluoroscopy system captures anatomical structures including pulmonary veins, while the catheter provides electrical activity measurements. By merging these two data sources, the system achieves complete anatomical visualization without missing critical features.
Solution Approach 2:
The patent uses fluoroscopic images as an intermediary to register and guide catheter-based mapping. The fluoroscopy system provides a visual reference framework that helps locate and identify anatomical structures, which then guides the catheter navigation and electrical mapping process, ensuring no critical features are missed.
2Loss of information
If fluoroscopic images are integrated with catheter-based mapping data, then the 3D anatomical map completeness is improved, but the system complexity increases
Solution Approach 1:
The patent creates a unified mapping system that performs multiple functions: fluoroscopic imaging for anatomical visualization, catheter-based electrical mapping, and automated integration of both data streams. This multi-functional system eliminates the need for separate systems and reduces overall complexity by consolidating functions into a single integrated platform.
Solution Approach 2:
The system employs automated image processing and registration algorithms that self-align fluoroscopic images with catheter-based mapping data without requiring manual intervention. The automated feature matching and coordinate transformation processes perform the integration task autonomously, reducing the operational complexity for users.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In one embodiment, a method for generating a three-dimensional (3D) anatomical map, including applying a trained artificial neural network to (a) a set of two-dimensional (2D) fluoroscopic images of a body part of a living subject, and (b) respective first 3D coordinates of the set of 2D fluoroscopic images, yielding second 3D coordinates of the 3D anatomical map, and rendering to a display the 3D anatomical map responsively to the second 3D coordinates.