Ultrasound Filter Coefficient Acquisition for Clutter Noise Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In ultrasound diagnosis, the pulse wave Doppler method struggles to effectively separate blood flow signals from clutter noise, especially when there are discontinuities in the input data or when the Doppler frequency of clutter overlaps with that of the blood flow, leading to inaccurate FFT displays.
Innovation Solution
The implementation of a filter coefficient acquiring circuitry that uses principal component analysis to derive a filter coefficient for suppressing clutter components, combined with MTI filter processing to extract a second data string from target data within a region of interest, and subsequent frequency analysis to generate a Doppler waveform.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high pass filter (HPF) of the infinite impulse response (IIR) type is applied to time-series data before FFT, then clutter noise is suppressed and blood flow signals are enhanced, but discontinuities in the input data cause artifacts and inaccurate FFT displays
Solution Approach 1:
The patent applies preliminary filtering actions before FFT processing. Specifically, it performs mean subtraction and variance normalization on the time-series data before applying the HPF and FFT. This preliminary preparation of the data removes discontinuities and stabilizes the signal, allowing the HPF to suppress clutter noise without creating artifacts that would corrupt the FFT display.
2Productivity
If fixed-point computation is used for FFT, then computational efficiency is improved, but the dynamic range is insufficient when signals near the direct-current component with high signal intensity are included
Solution Approach 1:
The patent changes the parameters of the time-series data by performing mean subtraction and variance normalization before FFT processing. This parameter transformation centers the data around zero mean and scales it to a standardized variance, which expands the effective dynamic range for fixed-point computation. This allows fixed-point FFT to accurately represent both high-intensity direct-current components and low-intensity blood flow signals without losing precision.
3Reliability
If the Doppler frequency of clutter overlaps with that of blood flow, then frequency-based separation becomes difficult, but principal component analysis enables effective clutter suppression
Solution Approach 1:
The patent introduces principal component analysis (PCA) as an intermediary processing step between data acquisition and FFT. PCA transforms the original time-series data into a new coordinate system where the first principal component captures the dominant clutter signal and subsequent components capture blood flow signals. By separating the data into these orthogonal components, the system can selectively process and suppress clutter while preserving blood flow information, even when their Doppler frequencies overlap.
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 reduces noise and allows for the clear display of blood flow signals in FFT displays, even in cases of discontinuities or overlapping frequencies, enhancing the accuracy of Doppler waveform representation.
Implementation Method 1
an ultrasound probe 1, a monitor 2, an input device 3, and an apparatus main body 10
Implementation Method 2
The pulse wave Doppler (hereinafter, PWD) method is one method for displaying signals from a blood flow. In the PWD method, a Doppler waveform is generated by performing fast Fourier transform (FFT) on time-series data of signals received from the same location
Data Source
AI summary
An ultrasound diagnosis apparatus includes filter coefficient acquiring circuitry configured to, based on a result of a principal component analysis using first data strings that are sets of data generated based on echo signals caused by transmission of ultrasound waves on the same scan line, obtain a filter coefficient that suppresses clutter components; deriving circuitry configured to use the filter coefficient to obtain, from target data strings contained in a region of interest among the first data strings, a second data string that is a set of data derived from echo signals based on a moving body present in the region of interest, and derive waveform information indicating temporal changes of the moving body by performing a frequency analysis on the second data string; and control circuitry configured to generate a waveform information image based on the wave information and cause a monitor to display the waveform information image.