Membranous Septum Segmentation Using LVOT Wall Thickness Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The membranous septum is difficult to segment in 3D image data sets due to its small size and non-planar surface, making manual annotations tedious and costly, and existing methods lack efficient automated segmentation techniques.
Innovation Solution
A device and method that segments the left ventricular outflow tract (LVOT) to determine wall thickness information, maps this information onto the LVOT surface, and uses it to automate the segmentation of the membranous septum, employing techniques like thresholding, independent component analysis, and k-nearest neighbor clustering, with optional model-based segmentation and machine learning for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual annotation methods are used to segment the membranous septum, then segmentation accuracy can be achieved, but the process becomes tedious and costly
Solution Approach 1:
The system performs automated segmentation of the membranous septum by processing 3D image data independently without requiring manual annotation. The circuitry segments the LVOT, determines wall thickness information, and identifies the membranous septum automatically, making the system self-sufficient and eliminating the need for tedious manual work.
Solution Approach 2:
The system uses wall thickness information as a key parameter to identify and segment the membranous septum. By mapping wall thickness information onto the LVOT surface and using it as a segmentation criterion, the system transforms the segmentation problem into a parameter-based identification task that can be solved automatically.
2Ease of operation
If the membranous septum is segmented in standard planar slice views, then the segmentation process is simplified, but the small and unobtrusive structure becomes difficult to identify
Solution Approach 1:
The system transitions from standard planar slice views to a 3D surface representation of the LVOT. By mapping wall thickness information onto the 3D surface of the segmented LVOT, the system provides a more comprehensive view that makes the small and unobtrusive membranous septum structure visible and identifiable in its spatial context.
3Measurement precision
If extensive manual annotations are collected for neural network training, then segmentation accuracy improves, but the cost and complexity increase significantly
Solution Approach 1:
The system replaces the need for extensive manual annotations and neural network training with a deterministic algorithmic approach. Instead of using machine learning models that require large datasets, the circuitry uses geometric and anatomical relationships (wall thickness information) to automatically segment the membranous septum, substituting computational geometry for statistical learning.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The present invention relates to a device and method for segmenting the membranous septum. The method comprises segmenting the left ventricular outflow tract, LVOT, of the heart in a 3D image data set; determining wall thickness information indicative of the wall thickness of the septum at different locations of a part of the segmented LVOT that is oriented towards the right heart chambers and right atrium; mapping the determined wall thickness information onto the surface of said part of the segmented LVOT; and segmenting the membranous septum in the 3D image data set based on the mapped wall thickness information.