Intervention Planning Visualization Using Signed Distance Fields
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical intervention planning techniques lack accuracy in visualizing complex tumor surfaces and multiple objects, particularly in procedures like tumor ablation and resection, due to the use of simple geometric shapes and low-order surfaces.
Innovation Solution
The use of a signed distance field (SDF) to determine the surface of target objects, enabling accurate visualization of safety margins and resection surfaces during medical intervention planning, based on orthogonal distances from the target object surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple geometric shapes (sphere proxies) are used for tumor visualization, then the device complexity is reduced, but the manufacturing precision and measurement precision deteriorate
Solution Approach 1:
The patent creates a digital copy of the complex tumor surface by generating a signed distance field (SDF) that mathematically represents the exact geometry. This digital copy allows accurate representation of arbitrary tumor shapes without physical models, resolving the contradiction between simplicity and precision.
Solution Approach 2:
The patent transforms the representation parameters from simple geometric primitives to continuous distance field values. By using SDF where each voxel stores the signed distance to the tumor surface, the system achieves high precision while maintaining computational efficiency through parameter optimization.
2Ease of operation
If low-order Bezier surfaces are used for resection planning, then the ease of operation is improved, but the manufacturing precision deteriorates
Solution Approach 1:
The patent replaces the mechanical/mathematical system of Bezier surface fitting with a distance field-based system. The SDF naturally handles complex geometries through straightforward distance calculations, eliminating the need for complex surface fitting operations while maintaining ease of use.
3Measurement precision
If accurate complex object representation is implemented, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent moves from representing surfaces as 2D manifolds (requiring complex mesh structures) to representing them as 3D distance fields. This dimensional transformation simplifies the computational model while enabling arbitrary surface complexity, as the SDF naturally encodes geometric information in three-dimensional space.
4Productivity
If real-time visualization of safety margins is provided, then the productivity is improved, but the use of energy increases
Solution Approach 1:
The patent performs preliminary computation by pre-calculating the signed distance field from the tumor surface. This pre-computed SDF can then be used for rapid safety margin visualization at any distance threshold without repeated complex calculations, achieving real-time performance while controlling energy consumption.
Data Source
Figure 1
Figure 2~3
AI summary
Medical imaging data (18) comprising a representation of an ROI including a segmented representation of a target object (8) is received. A surface of the target object (8) is determined and an SDF (19) comprising an orthogonal distance of the respective voxel from the surface of the target object (8), is generated. A rendered output representation (7) comprises: i. a visualization (12) of a safety margin (9) surrounding the target object (8) generated based on the SDF (19); and/or ii. a visualization (13) the target object (8) depending on distances of a resection surface from the surface of the target object (8), which are determined based on the SDF (19) map; and/or iii. a visualization (14) of the resection surface, which depends on the distances of the resection surface from the surface of the at least one target object (8), which are determined based on the SDF (19) map.