Volumetric Medical Image Rendering Voxel Visibility Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current volumetric medical imaging techniques face challenges in accurately determining voxel visibility, leading to artefacts and inefficient rendering due to the naive thresholding method, which fails to account for interpolation and results in missing critical voxels and blocky artefacts.
Innovation Solution
A method and apparatus that utilize a visibility thresholding process, where a region of interest is set around each voxel, determining maximum and minimum data values, and using a summed visibility table to designate voxels as visible or non-visible, considering interpolated data values and opacity thresholds, to improve voxel classification and reduce artefacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If naive thresholding method is used to determine voxel visibility, then rendering efficiency is improved by disregarding non-visible voxels, but voxel classification accuracy deteriorates leading to artefacts and missing critical voxels
Solution Approach 1:
The patent applies preliminary action by performing visibility classification on voxels before the actual rendering process. A classification stage is introduced that uses transfer function opacity values to pre-determine which voxels are visible and which are non-visible. This preliminary classification allows the rendering stage to efficiently skip non-visible voxels while ensuring that all visible voxels are correctly identified, thus resolving the contradiction between rendering efficiency and classification accuracy.
2Loss of energy
If naive thresholding disregards voxels with zero opacity, then rendering resources are reduced, but interpolation effects are lost causing blocky artefacts
Solution Approach 1:
The patent introduces an intermediary classification mechanism that acts between the raw voxel data and the rendering process. This classification stage evaluates opacity values from transfer functions and makes informed decisions about voxel visibility. By using this intermediary layer, the system can accurately identify voxels that contribute to interpolation effects even if their individual opacity is zero, while still disregarding truly non-visible voxels to save rendering resources, thus eliminating blocky artefacts without wasting computational resources.
3Measurement precision
If all voxels are processed in rendering to ensure accuracy, then voxel classification accuracy is maintained, but rendering time and computational resources increase significantly
Solution Approach 1:
The patent applies segmentation by dividing the rendering process into two distinct segments: a classification stage and a rendering stage. In the classification stage, voxels are evaluated and labeled as visible or non-visible based on transfer function opacity values. In the rendering stage, only voxels marked as visible are processed. This segmentation allows the system to maintain high classification accuracy through careful evaluation while significantly reducing rendering time by excluding non-visible voxels from the computationally intensive rendering process.
Data Source
AI summary
A medical processing apparatus comprises processing circuitry configured to: receive an image data set for rendering; for each of a plurality of a pixels or voxels in the image data set: set a region of interest around the pixel or voxel; determine a maximum data value and a minimum data value for pixels or voxels in the region of interest; and designate the pixel or voxel as visible or as non-visible based on the maximum data value and the minimum data value for the region of interest; and perform a rendering process using the pixels or voxels of the image data set that designated as visible.