Kidney Segmentation from MR Images Using Spatial Priors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for segmenting kidneys from magnetic resonance images in patients with ADPKD are resource-intensive, time-consuming, and prone to analyst bias and error, lacking a fully automated system for accurate volumetric measurement.
Innovation Solution
A fully automated system and method using a spatial prior probability map for regional mapping and boundary refinement with a level set framework, employing propagated shape constraints to generate segmented MR organ volumes, enabling accurate kidney segmentation and volumetric measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If manual delineation and adaptive thresholding methods are used for kidney segmentation, then the segmentation can be completed with straightforward procedures, but the process becomes resource intensive, time consuming, and subject to analyst bias and error
Solution Approach 1:
The patent applies preliminary action by performing preliminary segmentation to generate an initial kidney segmentation result before the refinement step. This preliminary segmentation creates a starting point that guides subsequent boundary refinement, eliminating the need for complete manual delineation while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical manual delineation process with an automated image processing system using level set methods and boundary refinement algorithms. This substitution eliminates human analysts from the segmentation process, removing subjectivity and bias while dramatically increasing processing speed.
2Ease of operation
If manual delineation methods are used for kidney segmentation, then the procedure remains straightforward, but the method requires heavy observer input and is prone to analyst bias and error
Solution Approach 1:
The patent applies self-service by designing a system where the segmentation algorithm automatically refines its own boundaries through level set evolution and boundary refinement mechanisms. The system uses image data and mathematical constraints to self-correct and optimize segmentation results without human intervention, ensuring consistency and eliminating analyst bias.
Solution Approach 2:
The patent implements feedback through iterative boundary refinement where the level set method continuously evaluates and adjusts segmentation boundaries based on image gradients and shape constraints. This feedback loop ensures that the final segmentation accurately reflects the true kidney boundaries while maintaining operational simplicity.
3Extent of automation
If semi-automated methods are used for kidney segmentation, then some automation is achieved, but observer perceptual guidance and input are still required
Solution Approach 1:
The patent applies segmentation by dividing the kidney segmentation process into distinct phases: preliminary segmentation to establish initial boundaries, followed by boundary refinement to optimize accuracy. This phased approach allows each stage to be fully automated while maintaining operational simplicity, as the system transitions automatically between phases without requiring observer intervention.
Data Source
AI summary
A method of segmenting an MR organ volume includes performing regional mapping on the MR organ volume using a spatial prior probability map of a location of the organ to create a regionally mapped MR organ volume, and performing boundary refinement on the regionally mapped MR organ volume using a level set framework that employs the spatial prior probability map and a propagated shape constraint to generate a segmented MR organ volume.