3D Tissue Mesh Visualization with 2D Image Feature Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-assisted medical imaging technologies struggle to accurately and comprehensively visualize surrounding tissue features, particularly those located outside or inside the surface of a tissue, due to limitations in integrating detailed 2D image data with 3D models.
Innovation Solution
A computer-implemented method that receives a 3D model of a tissue and integrates detailed 2D image data by segmenting regions of interest, extracting local characteristics, and associating these characteristics with mesh elements in the 3D model, creating a sub-mesh labeled with these characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional imaging methodologies (MRI, CT scans, ultrasound) are used to visualize internal structures, then the capability to perceive internal bodily structures is broadened, but the discernment of adjacent or surrounding tissue features outside or inside a given tissue surface remains suboptimal
Solution Approach 1:
The patent segments the 2D image data into multiple regions of interest (ROIs) corresponding to different surrounding tissue features. Each ROI is processed independently to extract local characteristics, then associated with corresponding mesh elements in the 3D model. This segmentation enables precise visualization of specific surrounding features without being overwhelmed by the complexity of the entire anatomical structure.
Solution Approach 2:
The patent transforms 2D image data into 3D mesh elements by associating extracted local characteristics with spatial coordinates on the 3D model surface. This dimensional transformation allows surrounding tissue features visualized in 2D images to be accurately represented and positioned on the 3D model, enhancing spatial understanding and visualization accuracy.
2Measurement precision
If 3D models of organs and tissues are adopted to facilitate immersive understanding of anatomical structures, then the understanding of anatomical structures is enhanced, but the representation of surrounding or adjacent tissue features is inadequately represented
Solution Approach 1:
The patent extracts local characteristics from 2D image data corresponding to surrounding tissue features. These extracted characteristics (such as intensity, texture, or other image properties) are then mapped to the 3D model's mesh elements, ensuring that surrounding tissue information is not lost but rather integrated into the 3D representation for comprehensive anatomical understanding.
Solution Approach 2:
The patent uses mesh elements as intermediaries to bridge the gap between 2D image data and 3D model representation. The mesh elements serve as carriers that transmit the extracted local characteristics from the 2D images to the 3D model, enabling accurate representation of surrounding tissue features while maintaining the immersive 3D visualization capability.
3Measurement precision
If static high-fidelity representations of internal structures are provided, then the fidelity of internal structure representation is improved, but the seamless integration of real-time or static imaging data with pre-existing 3D models is insufficient
Solution Approach 1:
The patent performs preliminary processing of 2D image data by segmenting it into regions of interest and extracting local characteristics before integrating with the 3D model. This preliminary action prepares the imaging data in advance, making the subsequent integration with the 3D model faster and more automated, thus improving productivity while maintaining high fidelity representation.
Solution Approach 2:
The patent replaces manual intervention and complex computational methodologies with automated image processing techniques. By using algorithmic segmentation and characteristic extraction, the system automatically integrates imaging data with the 3D model, significantly improving integration speed and productivity while preserving the high-fidelity representation of internal structures.
4Ease of operation
If rudimentary thresholding techniques and simple region-of-interest extraction methodologies are employed, then the ease of processing is improved, but the precision of identifying and segmenting relevant data is insufficient
Solution Approach 1:
The patent extracts local characteristics from 2D image data at specific regions of interest corresponding to surrounding tissue features. Instead of using uniform thresholding across the entire image, the method focuses on local image properties (such as local intensity, texture, or gradient characteristics) within each ROI, enabling precise identification of surrounding tissue features while maintaining ease of processing through automated extraction.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention concerns a computer-implemented method for the visualization of surrounding tissue features located out of a surface of a tissue of a patient's organ, the method comprising the following steps : receiving a 3D model comprising a mesh representing a surface of said tissue; receiving at least one image representing said tissue and at least one surrounding tissue feature located out of said surface; segmenting, in said image, at least one region of interest corresponding to said surrounding tissue feature and being formed by a plurality of first sub-parts; extracting, for each first sub-part, at least one value of a predetermined local characteristic; associating each sub-part to at least one mesh element of said mesh, based on said first sub-part and mesh element's positions; wherein said associated mesh elements form together a sub-mesh; each mesh element being labelled with the value of the predetermined local characteristic of its associated first sub-part.