Brick-Based Fusion Renderer for Multi-Volume Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fusion renderers require all volumes to be aligned and resampled to the same resolution, leading to increased processing time and memory usage, especially when volumes have different orientations and coordinate systems, and existing bricking techniques are inefficient due to graphics hardware limitations.
Innovation Solution
Divide each 3D image dataset into bricks along three orthogonal axes, building hierarchical structures with expanded nodes to cover the entire 3D space, and render each brick in order of visibility, using cut-planes to ensure all visible portions are rendered exactly once, while sharing graphics memory to optimize performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all volumes are aligned and resampled to the same resolution using existing fusion renderers, then fusion rendering can be performed, but processing time and memory usage increase significantly
Solution Approach 1:
The patent divides the 3D volumetric dataset into multiple smaller sub-volumes called bricks along three orthogonal axes. Each brick is an independent unit that can be processed separately. This segmentation allows the rendering system to process only visible bricks rather than the entire volume, significantly reducing processing time while maintaining fusion rendering capability across multiple datasets with different resolutions and orientations
Solution Approach 2:
The patent implements dynamic brick visibility determination based on the current view frustum and camera position. The system dynamically identifies which bricks are visible and processes only those, rather than statically processing all bricks. This dynamic approach adapts to different viewing angles and positions, optimizing processing time for each rendering operation while preserving the ability to render fused images from multiple datasets
2Reliability
If all volumes are aligned and resampled to the same resolution using existing fusion renderers, then fusion rendering can be performed, but memory consumption increases
Solution Approach 1:
By segmenting the volume into bricks, the system loads only the necessary bricks into memory during rendering rather than the entire volume. This reduces memory consumption from storing all voxels of multiple high-resolution volumes to storing only the visible subset, enabling fusion rendering of multiple datasets without excessive memory requirements
Solution Approach 2:
The patent introduces a hierarchical spatial indexing structure that organizes bricks in a tree-like hierarchy. This allows efficient memory management by loading only the relevant portion of the brick hierarchy into memory based on visibility, rather than loading entire volumes. The hierarchical structure enables progressive loading and memory release, optimizing memory usage for fusion rendering
3Manufacturing precision
If volumes are registered with user interaction and visual feedback, then accurate alignment is achieved, but rendering speed decreases
Solution Approach 1:
The patent performs preliminary registration of volume coordinate systems and orientations before the rendering process. The registration parameters are determined in advance and stored with the volume data. During rendering, these pre-computed parameters are used directly without requiring real-time interactive adjustment, achieving both accurate alignment and fast rendering speed
Solution Approach 2:
The system dynamically adjusts the rendering process based on pre-computed registration parameters. The brick visibility determination and processing order are dynamically optimized based on the registered coordinate systems, allowing fast rendering while maintaining accurate spatial relationships between multiple datasets
4Reliability
If re-sampling is performed to align volumes to reference volume grid, then fusion rendering is enabled, but processing time increases for initial and each rendering
Solution Approach 1:
Instead of resampling all volumes to match a reference volume grid, the patent inverts the approach by keeping each volume at its native resolution and orientation, and transforming the rendering process to accommodate different grids. The brick-based rendering system samples from each volume's native grid according to its registration parameters, eliminating the need for time-consuming resampling operations while maintaining accurate volume alignment in the fused output
Data Source
AI summary
A method for combining at image data sets to generate a composite image. The method divides each of the data sets into a plurality of bricks along three mutually orthogonal axes. The method includes (a) building a hierarchical structure for each one of the at least two data sets, each structure comprising higher level blocks of voxels and lower level blocks of voxels, the higher level blocks having a larger number of voxels than the lower level blocks; (b) expanding boundaries of each one of the structures into corresponding expanded hierarchical structures, such boundary expanding comprising adding additional virtual processing bricks to the initial processing bricks, such virtual processing bricks comprising semi-unbounded blocks to provide the expanded boundaries of the expanded hierarchical structures; and (c) rendering each one of the bricks in each one of the expanded hierarchical structures into a two dimension image.