Adaptive Volume Ray Casting for Ultrasound Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional volume ray casting in ultrasound diagnostic systems samples data at a constant interval, leading to decreased rendering speed and exclusion of smaller objects, as it includes unnecessary spaces and fails to accurately sample minute objects.
Innovation Solution
The method adjusts the sampling interval by casting virtual rays and checking voxel opacity, sampling in a shorter interval when encountering object space and a longer interval in empty space, with accumulated opacity determining the completion of sampling processes to enhance rendering speed and include minute objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If volume data are sampled at a constant sampling interval in conventional volume ray casting, then the sampling process is simple to implement, but the rendering speed decreases and minute objects smaller than the sampling interval are excluded
Solution Approach 1:
The patent applies dynamic sampling by adjusting the sampling interval based on the detected object boundaries. The sampling interval is shortened when objects are detected and lengthened in empty spaces, making the sampling process adaptive rather than static. This resolves the contradiction by allowing the system to maintain high rendering speed in empty regions while ensuring precise sampling of minute objects when encountered.
Solution Approach 2:
The patent implements local quality enhancement by applying different sampling strategies to different regions of the volume data. In regions identified as empty space, a longer sampling interval is used to maintain speed, while in regions containing objects, a shorter interval is applied to ensure accurate capture of minute details. This localized adaptation resolves the contradiction between speed and precision.
2Productivity
If volume data are uniformly sampled in a ray propagation direction, then desirable image quality is obtained, but unnecessary spaces are also sampled leading to decreased rendering speed
Solution Approach 1:
The patent extracts and identifies empty space regions from the volume data using opacity-based detection. Once empty spaces are identified, the sampling process skips or reduces sampling in these regions, effectively taking out the unnecessary sampling operations. This resolves the contradiction by maintaining image quality in object regions while eliminating redundant sampling in empty spaces to improve rendering speed.
Solution Approach 2:
The patent applies partial sampling action by selectively sampling only the portions of the volume data that contain objects, rather than uniformly sampling the entire volume. The sampling is intensified (excessive action) in object regions to ensure quality, while being reduced or skipped in empty regions to improve speed, thus resolving the contradiction through selective application of sampling density.
3Measurement precision
If a shorter sampling interval is used to capture minute objects, then sampling precision is improved, but the rendering speed decreases due to increased number of samples
Solution Approach 1:
The patent dynamically adjusts the sampling interval based on the local content of the volume data. When minute objects are detected through opacity checking, the sampling interval is shortened to ensure precise capture. When empty spaces are detected, the sampling interval is lengthened to maintain rendering speed. This dynamic adaptation resolves the contradiction by applying short intervals only where necessary for precision while maintaining speed in other regions.
Solution Approach 2:
The patent enhances local quality by applying high-density sampling (short interval) only in specific local regions where objects are present, rather than uniformly across the entire volume. In regions identified as empty space, lower-density sampling (long interval) is used. This localized quality enhancement resolves the contradiction by concentrating computational resources where precision is needed while maintaining speed elsewhere.
Data Source
AI summary
The present invention relates to an apparatus and a method for rendering volume data in an ultrasound diagnostic system. A method for rendering volume data including an object space and an empty space acquired from ultrasound data, comprises the following steps: a) casting a virtual ray from at least one pixel comprising a viewing plane into the volume data; b) sampling voxels of the volume data along the virtual ray in a first sampling interval; c) checking whether a currently sampled voxel corresponds to the object space or the empty space by using opacity based on a voxel value of a sampled voxel; d) at step c), if it is determined that the currently sampled voxel corresponds to the object space, then returning to the previously sampled voxel and sampling voxels in a second sampling interval shorter than the first sampling interval; e) checking whether accumulated opacity calculated based on the sampled voxels along the virtual ray is greater than a critical value; f) at step e), if it is determined that the accumulated opacity is greater than the critical value, then completing the sampling process for the current virtual ray and calculating a rendering value by using the voxel values and the opacity of the sampled voxels; and g) repeating the steps a) to f) until the sampling process for the pixels comprising the viewing plane is completed.