Ultrasound Strain Imaging Dropout Correction via Displacement Error Reset
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ultrasound systems face challenges in strain imaging due to decorrelation errors, which result in dropouts or horizontal lines in strain images, especially when the initial displacement error occurs, making it difficult to accurately visualize tumors or cancers with similar tissue reflectivity.
Innovation Solution
The ultrasound system acquires both pre-compression and post-compression frame data, sets windows to compute displacements, and uses a second window to check for displacement computation errors, resetting pixel values when errors occur, thereby correcting dropouts and enhancing image accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the window is moved by initial displacement to compute displacement, then decorrelation error is reduced, but error propagation in axial direction occurs causing dropout
Solution Approach 1:
The patent implements feedback by checking the computed displacement values and detecting outliers that indicate dropout errors. When a dropout is detected in the current frame, the system uses displacement information from the previous frame as feedback to correct the erroneous value, thereby preventing error propagation and maintaining computational accuracy.
Solution Approach 2:
The patent performs preliminary action by pre-computing displacements in the axial direction before lateral direction processing. This preliminary computation establishes a baseline that can be used to detect and correct errors, as the system compares axial displacements with lateral displacements to identify inconsistencies indicating dropout errors.
2Measurement precision
If correlation operation is performed to obtain displacement, then tissue deformation is measured, but decorrelation error occurs when position gap exceeds phase range
Solution Approach 1:
The patent applies segmentation by dividing the displacement computation into two separate stages: axial direction displacement computation followed by lateral direction displacement computation. This segmentation allows each stage to operate within optimal correlation ranges, preventing decorrelation errors that would occur if a single large position gap exceeded the phase range.
Solution Approach 2:
The patent transitions from one-dimensional axial displacement measurement to two-dimensional displacement measurement by adding lateral direction computation. This dimensional expansion enables more accurate tissue deformation measurement while maintaining reliability through the use of reference frames and iterative correlation operations.
3Loss of information
If strain imaging is used to visualize tissue mechanical characteristics, then tumor detection is improved, but dropout errors reduce image quality
Solution Approach 1:
The patent converts the harmful effect of dropout errors into a benefit by using the detected dropout locations as indicators for where correction is needed. The system leverages the presence of errors to trigger corrective actions, such as using previous frame data or alternative computation paths, thereby transforming image degradation into an opportunity for error correction and improvement.
4Measurement precision
If window correlation is used to compute displacement, then axial and lateral displacements are obtained, but error accumulation occurs in axial direction
Solution Approach 1:
The system uses feedback mechanisms to detect and correct error accumulation. By comparing computed displacements with expected values and detecting outliers, the system triggers corrective actions that prevent error propagation, maintaining measurement precision across multiple computation steps.
Solution Approach 2:
The patent implements dynamic error correction by adaptively adjusting computation strategies based on detected error conditions. When errors are detected in axial direction computations, the system dynamically switches to alternative methods or uses data from previous frames, making the computation process flexible and resistant to error accumulation.
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 approach effectively corrects dropout errors in strain imaging, improving the visualization of tissues by accurately computing displacements and reducing errors, leading to clearer differentiation between tumor and normal tissues.
Implementation Method 1
the ultrasound image is displayed in a Brightness-mode (B-mode) by using reflectivity caused by an acoustic impedance difference between the tissues of the target object
Implementation Method 2
move the first window in a predetermined direction and compute a correlation between the first ultrasound frame data and the second ultrasound frame data within the first window to obtain a displacement
Data Source
AI summary
Embodiments for correcting a dropout in a strain image in an ultrasound system are disclosed. In one embodiment, a processing unit sets a first window on each of pre-compression ultrasound frame data and post-compression second ultrasound frame data, move the first window in a predetermined direction and compute a correlation between the pre-compression ultrasound frame data and the post-compression ultrasound frame data within the first window to obtain a displacement corresponding to a value of each pixel of target ultrasound frame data. The processing unit sets one of pixels of the target ultrasound frame data as a reference pixel, sets a second window to encompass predetermined numbers of pixels positioned around the reference pixel, checks whether a displacement computation error corresponding to a dropout occur based on the pixel values within the second window and resets, when the dropout occurs, the value of the reference pixel.