Cardiac Chamber Skeletonization from Incomplete Noisy Voxels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing skeletonization algorithms for electroanatomical maps from cardiac chamber data are prone to errors due to incomplete and noisy volume data, leading to inaccurate and disconnected skeletons.
Innovation Solution
A multi-step algorithm that includes filtering out outlier candidate locations, spatially compressing and optimizing candidate points using geometrical and statistical models, and applying Principal Component Analysis (PCA) to connect these points into a well-defined 1D curve.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing skeletonization algorithms are applied to incomplete and noisy voxel data, then processing speed is maintained, but skeleton accuracy and completeness deteriorate
Solution Approach 1:
The patent applies preliminary actions by first filling missing voxels using morphological operations and distance transform before skeletonization. This preprocessing step ensures that incomplete data is reconstructed, providing a more complete input for subsequent skeleton extraction and improving both accuracy and reliability of the final skeleton
Solution Approach 2:
The patent segments the skeletonization process into multiple independent steps: voxel filling, distance transform, skeleton extraction, and graph construction. This segmentation allows each step to be optimized independently and enables error correction at intermediate stages, improving overall skeleton quality from noisy input data
2Measurement precision
If noise filtering is applied to voxel data, then skeleton accuracy improves, but processing time increases
Solution Approach 1:
The patent extracts and removes outlier voxels from the dataset before skeletonization using statistical methods. By taking out only the problematic noisy voxels rather than processing the entire dataset with heavy filtering, the method achieves noise reduction with minimal impact on processing time
Solution Approach 2:
The patent changes parameters by applying adaptive thresholding based on local voxel density and distance metrics. This dynamic parameter adjustment allows efficient noise filtering that adapts to different regions of the data, maintaining high accuracy while optimizing processing speed through localized rather than global processing
3Measurement precision
If complex preprocessing is applied to noisy data, then skeleton quality improves, but algorithm complexity increases
Solution Approach 1:
The patent applies self-service by using the data's own statistical properties (mean, standard deviation, distance metrics) to automatically determine filtering and reconstruction parameters. This eliminates the need for manual parameter tuning and complex adaptive algorithms, achieving high skeleton quality through simple, data-driven operations
Solution Approach 2:
The patent creates a corrected copy of the voxel data through morphological operations and distance transform, preserving the original noisy data while generating a cleaned version for skeletonization. This copying approach isolates the complexity to a single preprocessing step without requiring complex iterative modifications to the core skeletonization algorithm
Data Source
AI summary
A method for generating a skeleton in an image of a cavity of an organ of a body includes receiving a map of the cavity, the map including surface voxels and interior voxels. A subset of the interior voxels is generated, of candidate locations to be on the skeleton. The subset is pruned by removing outlier candidate locations. Using a geometrical model including a statistical model, the candidate locations remaining in the pruned subset are spatially compressed. The compressed candidate locations are connected to produce one or more centerlines of the skeleton. At least the skeleton is displayed to user.