Medical Image Decomposition for Depth Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging technologies face challenges in distinguishing between objects with similar intensity values and in visualizing regions of interest obscured by high-intensity structures, such as bones, which can occlude tissue voxels of greater interest for analysis.
Innovation Solution
The method involves decomposing digital medical images into component volumes based on anatomical structures, generating 2D projective views by moving a projection viewpoint around each structure, and displaying a cine loop to create a rotating effect, while also applying intensity thresholding to separate voxels above a certain threshold from the rest, allowing for clearer visualization of structures like tumors or organs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If maximum intensity projection (MIP) is used to visualize 3D medical data, then computational speed is improved, but depth perception and distinction between structures with similar intensity values deteriorates
Solution Approach 1:
The patent segments the 3D medical volume into multiple 2D slice sections at different depths. Each slice is rendered separately with depth information preserved, allowing the human visual system to perceive depth through the sequence of slices while maintaining computational efficiency through optimized rendering of individual 2D sections.
Solution Approach 2:
The patent transitions from traditional 2D MIP projection to a multi-slice 3D visualization approach where depth information is preserved by displaying multiple 2D slices stacked in the depth dimension. This dimensional approach allows simultaneous preservation of computational speed (through 2D rendering) and depth perception (through multi-slice arrangement).
2Loss of information
If bones and high intensity structures are included in MIP images, then complete anatomical information is preserved, but occlusion of underlying soft tissue structures of interest occurs
Solution Approach 1:
The patent extracts and removes high-intensity structures such as bones from the visualization by applying intensity thresholding. Voxels above a specified intensity threshold are excluded from the MIP calculation, allowing underlying soft tissue structures to be visualized without occlusion while preserving complete anatomical information in the processed output.
Solution Approach 2:
The patent applies different processing qualities to different regions of the image based on intensity characteristics. High-intensity regions (bones) are processed differently through thresholding and removal, while mid-range intensity regions (soft tissues) are preserved with full detail, optimizing visualization quality for each anatomical type.
3Productivity
If traditional 2D MIP rendering is used, then computational efficiency is maintained, but the ability to distinguish left from right and clockwise from counter-clockwise rotation is lost
Solution Approach 1:
The patent segments the volume into multiple 2D slice sections that are rendered in sequence. By preserving the depth ordering and spatial relationships between slices, the visualization maintains chirality information (left/right distinction and rotation direction) while using computationally efficient 2D rendering techniques for each slice.
Solution Approach 2:
The patent employs dynamic rendering where the viewpoint can be moved through the 3D volume by adjusting slice positioning and ordering. This dynamic repositioning of slices allows the viewer to perceive rotational direction and chirality without requiring complex 3D rendering, maintaining computational efficiency while recovering directional information.
Data Source
AI summary
A method for decomposing digital medical images includes providing a digital medical image, segmenting the image into one or more biological structures, extracting one or more segmented biological structures from the image by extracting all voxels within a spatial extent of each of the biological structures to construct one or more new component volumes of the biological structures. For each of the one or more new component volumes, generate a sequence of 2-dimensional projective views by moving a projection viewpoint around each the biological structure in the one or more new component images, and generate a 2-dimensional projective view from each viewpoint, and display a cine loop of the sequence of projective views where the biological structures appear to be rotating in the display.