3D Ultrasound Image Segmentation Using Simulated Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ultrasound image segmentation methods fail to effectively segment 3D images due to shadow regions caused by energy absorption or reflection in body tissues, leading to incomplete image models.
Innovation Solution
A multi-target 3D ultrasound image segmentation method that uses simulated and measured data, involving presetting acoustic parameters, initial segmentation, probability variable representation, simulation operations, and iterative comparison to adjust probabilities and refine tissue classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional segmentation algorithms are used on raw 3D ultrasound data, then segmentation speed is maintained, but segmentation accuracy deteriorates due to shadow regions caused by energy absorption or reflection
Solution Approach 1:
The patent applies preliminary action by performing simulation operations before final segmentation. Synthetic ultrasound data is generated based on preliminary tissue parameter estimates, and this simulated data is used to guide the segmentation process before actual segmentation occurs, thereby improving accuracy in shadow regions without requiring complex iterative algorithms
Solution Approach 2:
The patent introduces simulated ultrasound data as an intermediary between the raw data and the segmentation algorithm. This simulated data acts as a mediator that fills in information for shadow regions, allowing conventional segmentation algorithms to work effectively without needing to be fundamentally complex
2Measurement precision
If iterative refinement with simulation operations is performed, then segmentation accuracy improves, but computational time increases
Solution Approach 1:
The patent applies partial action by performing simulation operations only on regions where they are most needed (shadow regions identified from initial segmentation), rather than uniformly processing the entire 3D volume. This selective approach improves accuracy in critical areas while minimizing overall computational time
Solution Approach 2:
The patent implements periodic action through iterative refinement where simulation operations are performed in cycles. Each iteration refines tissue parameters and regenerates simulated data, with the process repeating a limited number of times until convergence or maximum iterations are reached, balancing accuracy improvement with computational efficiency
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enables complete image modeling by iteratively refining tissue classification, overcoming the limitations of existing algorithms and achieving accurate segmentation even in shadowed regions.
Implementation Method 1
performing a simulation operation on the transitional image model by utilizing the principle of ultrasound imaging
Implementation Method 2
selecting the conventional acoustic parameters of a corresponding normal tissue or lesion tissue to substitute the actual acoustic parameters of the incomplete target tissue
Data Source
AI summary
The disclosure provides a multi-target 3D ultrasound image segmentation method based on simulated and measured data. The method includes: presetting conventional acoustic parameters; collecting raw 3D data; employing an initial segmentation algorithm to segment the raw 3D data; substituting with the conventional acoustic parameters according to probability in order to form a transitional image model; performing a simulation operation; performing transformation to obtain simulated data; performing a comparison operation; adjusting corresponding magnitude of the probability in each probability variable, and returning to the step of substituting with the conventional acoustic parameters. According to the probability, the conventional acoustic parameters are substituted into a model of an incomplete target tissue, the simulation operation is then performed, the probability is adjusted, and repeatedly, corrections are performed continuously in an iterative convergence manner till each incomplete target tissue is completely substituted by a certain normal tissue or lesion tissue.