Ultrasound Displacement Map via Dynamic Programming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ultrasound elasticity imaging techniques suffer from artifacts and noise due to speckle decorrelation, leading to inaccurate tumor imaging during thermal ablation procedures, and are computationally expensive, causing significant lag in image display.
Innovation Solution
A method using dynamic programming to process ultrasound data by generating a displacement map that identifies corresponding pixels between two images, refining the map to obtain intermediate displacement values, and calculating physical properties of the region of interest, while incorporating Kalman filtering to reduce noise and improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ultrasound elasticity imaging techniques are used to track tissue deformation, then tumor contour can be extracted from surrounding tissue, but artifacts and noise are introduced due to speckle decorrelation
Solution Approach 1:
The patent introduces an intermediary statistical model that represents the probability distribution of speckle patterns. Instead of directly comparing speckle patterns which suffer from decorrelation, the model serves as a mediator that accounts for the statistical properties of speckle, thereby reducing artifacts and noise while maintaining measurement precision in tumor imaging.
2Measurement precision
If related art image processing techniques are used to track speckle motion, then tissue deformation can be measured, but computational cost increases significantly causing lag in image display
Solution Approach 1:
The patent extracts only the essential statistical features of speckle patterns rather than processing the complete speckle data. By taking out and processing only the critical statistical moments needed for deformation measurement, the computational burden is reduced while maintaining measurement precision, thereby improving image display speed.
3Measurement precision
If ultrasound probe pressure is applied to surrounding tissue to image tumor deformation, then tumor contour becomes visible, but out-of-plane motion increases causing speckle decorrelation
Solution Approach 1:
The patent applies a dynamic statistical model that adapts to changes in speckle patterns caused by probe pressure. The model dynamically adjusts to accommodate out-of-plane motion and speckle decorrelation, maintaining tumor contour visibility while accounting for the instability introduced by applied pressure through real-time statistical parameter updates.
Data Source
AI summary
A method of processing ultrasound data includes receiving ultrasound data for a first ultrasound image, the first ultrasound image being represented as a first set of discrete pixels corresponding to positions of a region of interest; receiving ultrasound data for a second ultrasound image, the second ultrasound image being represented as a second set of discrete pixels corresponding to positions of the region of interest; generating a displacement map by minimizing a cost function using a dynamic programming procedure that identifies each of the first set of discrete pixels with a corresponding one of the second set of discrete pixels; refining the displacement map to obtain intermediate displacement values corresponding to positions between the discrete pixels based on minimizing a local approximation to the cost function; and calculating a physical property of the region of interest based on the displacement map.