Weighted Active Shape Model for Intracochlear Anatomy Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for segmenting intracochlear anatomy in MR images fail to accurately separate the intracochlear anatomy from the entire labyrinth, which is necessary for conducting studies that relate cochlear signal to outcomes.
Innovation Solution
A weighted active shape model (wASM) is created from CT images and registered to MR images to automatically segment the intracochlear anatomy, using a combination of affine and nonrigid transformations to achieve accurate alignment and segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If level sets and statistical shape model are used to segment the labyrinth in MR images, then the segmentation of the entire labyrinth is achieved, but the intracochlear anatomy cannot be separated from the labyrinth
Solution Approach 1:
The patent applies segmentation by dividing the intracochlear anatomy into distinct separable structures (scala vestibuli, scala media, scala tympani) using a weighted active shape model that processes MR images to isolate individual anatomical components from the complete labyrinth, enabling precise measurement and analysis of each structure
Solution Approach 2:
The patent implements local quality by assigning different weights to different regions within the active shape model framework, allowing certain anatomical features to be emphasized or de-emphasized based on their importance for accurate segmentation of intracochlear structures while maintaining overall model coherence
2Productivity
If 3D U-Net-based method is used to segment the labyrinth, then automated segmentation is achieved, but the intracochlear anatomy remains unseparated
Solution Approach 1:
The patent achieves both automation and anatomical separation by using a weighted active shape model that automatically processes MR images to divide and identify individual intracochlear structures (scala vestibuli, scala media, scala tympani) as distinct segmented regions, providing both automated processing and precise anatomical differentiation
Solution Approach 2:
The patent applies parameter changes by modifying the energy function in the active shape model to include weighted terms that emphasize specific anatomical boundaries and features, allowing the automated segmentation process to differentiate between various intracochlear structures based on adjusted weighting parameters rather than uniform treatment
3Measurement precision
If CT images are used for segmentation, then intracochlear anatomy is visible, but MR images are needed for signal-outcome studies
Solution Approach 1:
The patent uses a weighted active shape model as an intermediary that was initially developed using CT image data but has been adapted to process and segment MR images, serving as a bridge that allows the model trained on CT anatomy to effectively segment MR images for signal-outcome research while maintaining anatomical accuracy
Data Source
AI summary
Methods and systems for automatic segmentation of structures of interest of an organ in an MR image. The method includes creating a weighted active shape model (wASM); registering model points of the structures in an MR atlas image to a target image that is the MR image to be segmented, as initial model points of the structures in the target image; and iteratively fitting the wASM to the target image, starting from the initial model points, until the shape converges, wherein the final shape is the segmentation of the structures of interest.