3D Fluoroscopic Tool-to-Lesion Visualization for Precise Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluoroscopic imaging struggles to resolve small soft-tissue objects like lesions due to their low density, and the 2D projections hinder accurate navigation of medical devices within the body.
Innovation Solution
A method and system for generating a 3D rendering of a medical tool relative to a lesion using fluoroscopic images, involving pose estimation, 3D reconstruction, and user interface marking to accurately visualize the tool's position and alignment with the lesion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard fluoroscopic imaging is used, then highly dense objects such as metal tools and bones are visualized well, but small soft-tissue objects such as lesions cannot be resolved
Solution Approach 1:
The patent transitions from 2D fluoroscopic projection to 3D volumetric reconstruction, enabling differentiation of structures at various depths. This dimensional change allows simultaneous visualization of both highly dense objects (metal tools, bones) and small soft-tissue objects (lesions) by providing depth information that was lost in the 2D projection.
2Measurement precision
If 2D fluoroscopic projection is used, then imaging is simple and fast, but accurate navigation of medical devices within the body is hindered
Solution Approach 1:
The system performs 3D volumetric reconstruction from multiple 2D fluoroscopic projections, adding the depth dimension to enable accurate spatial navigation. The reconstruction algorithm processes the projection data to generate a 3D representation that preserves navigation accuracy while maintaining the simplicity and speed of fluoroscopic imaging.
Solution Approach 2:
The patent introduces a 3D reconstruction algorithm as an intermediary processing step between the 2D fluoroscopic projections and the final navigation display. This intermediary transforms the limited 2D projection data into a comprehensive 3D volumetric representation, enabling accurate device navigation without requiring direct complex 3D imaging hardware.
3Measurement precision
If 3D reconstruction is performed from fluoroscopic images, then navigation accuracy is improved, but processing time and computational complexity increase
Solution Approach 1:
The system performs 3D reconstruction from a limited set of pre-acquired fluoroscopic projections rather than continuously processing real-time image streams. By preparing the 3D volumetric data structure in advance from a finite dataset, the system achieves high spatial localization accuracy while minimizing processing time during the actual navigation procedure.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods of visualizing a current view of a tool relative to a lesion by processing current fluoroscopic images from a current fluoroscopic sweep occurring after an initial fluoroscopic sweep. The processing includes determining the locations and/or orientations of a tool and a lesion in a current 3D reconstruction of the current fluoroscopic images or in a subset of the current fluoroscopic images, generating a 3D rendering based on the locations and/or orientations of the tool and the lesion, and displaying the 3D rendering. The locations and/or orientations of the tool and the lesion may be obtained from a user interface enabling a user to mark the current locations and/or orientations in the current 3D reconstruction or in a subset of the current fluoroscopic images, or by segmenting the current 3D reconstruction or a subset of the current fluoroscopic images.