Combined Rib And Spine Image Processing For Trauma Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging techniques for visualizing ribs and spine, such as the 'filet view' and 'visceral cavity view', suffer from discontinuities, incorrect rib lengths, and distortions, making them inefficient for time-critical trauma assessments.
Innovation Solution
A method involving deep learning for segmentation and centerline detection, followed by a deformed two-dimensional manifold plane alignment and interpolation, generates a continuous and straightened visualization of the rib cage and spine, combining the advantages of both views while avoiding their disadvantages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the filet view method is used to segment and visualize each rib independently, then the rib centerlines can be accurately inspected in a normalized view, but discontinuities between ribs occur leading to imaging artifacts
Solution Approach 1:
The patent combines multiple rib segments into a unified manifold structure that preserves continuity between adjacent ribs. Instead of processing each rib independently as in the filet view, the invention creates a continuous surface that spans across multiple ribs, eliminating the discontinuities and artifacts while maintaining accurate rib centerline representation through the integrated manifold geometry.
2Ease of operation
If the visceral cavity view is used to segment the inside of the rib cage as a continuous manifold, then the rib cage can be inspected as a whole, but the relative rib lengths are not maintained and unrealistic distortions occur
Solution Approach 1:
The patent applies local quality by allowing different regions of the manifold to have different properties: the overall manifold provides continuous coverage for easy inspection, while local constraints preserve the accurate relative lengths of individual ribs. The manifold is constructed to maintain rib-specific geometric fidelity in critical areas while providing continuous visualization across the entire rib cage structure.
3Area of stationary object
If the MIP visualization is used in the visceral cavity view, then the rib cage surface can be projected, but subtle fractures become invisible and undetectable
Solution Approach 1:
The patent creates a manifold copy or representation of the rib cage surface that preserves fine geometric details. Instead of using MIP projection which averages and obscures subtle features, the manifold approach generates a surface model that maintains the precision of the underlying CT data, allowing subtle fractures to remain visible and detectable while still providing comprehensive surface coverage.
4Loss of information
If whole-body CT scans are used to capture complete anatomy, then comprehensive diagnostic information is obtained, but large amounts of imaging data must be thoroughly inspected which is time-consuming
Solution Approach 1:
The patent extracts and isolates the rib cage and spine structures from the whole-body CT scan data, creating a focused manifold visualization that contains only the critical anatomical regions for trauma assessment. This extraction approach maintains comprehensive coverage of the relevant anatomy while removing unnecessary data from other body regions, significantly reducing the inspection time required for trauma evaluation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A visualization scheme is generated based on automated segmentation to display anatomies of a patient. Data representative of a three-dimensional diagnostic image is received. The ribs and spine are segmented, rib centerlines and vertebra center landmarks are detected and labeled. Three-dimensional position coordinates corresponding to the received data are mapped to a defined two- dimensional manifold plane followed by an interpolation technique to complete the remaining three- dimensional position coordinates and deform the two-dimensional manifold plane. Image intensities are sampled from three-dimensional diagnostic image space, at each coordinate of the deformed two-dimensional manifold plane. A reformatted image is generated as a manifold slice from the sampled image intensities displaying a continuous and straightened visualization of a rib cage. A three-dimensional visualization of the rib cage is generated from a plurality of manifold slices.