Clutter suppression in ultrasonic imaging systems

• US 9,451,932 B2
• Filed: 02/29/2012
• Issued: 09/27/2016
• Est. Priority Date: 12/30/2004
• Status: Active Grant

• Current Assignee: Crystalview Medical Imaging Limited
• Original Assignee: Crystalview Medical Imaging Limited
• Inventors: Zwirn, Gil
• Primary Examiner(s): Park, Patricia
• Application Number:
  US13/408,136
• Publication Number:
  US 20120157851A1
• Time in Patent Office: 1,672 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: A61B 8/0883 for diagnosis of the heart
• CPC Class Codes: A61B 8/4488 the transducer being a phased array
• CPC Class Codes: A61B 8/483 involving the acquisition of a 3D volume of data
• CPC Class Codes: A61B 8/488 involving Doppler signals
• CPC Class Codes: A61B 8/5207 involving processing of raw data to produce diagnostic data, e.g. for generating an image reconstruction G06T11/00
• CPC Class Codes: G01S 15/8993 Three dimensional imaging systems
• CPC Class Codes: G01S 7/52026 Extracting wanted echo signals Doppler systems G01S15/50; Doppler short range imaging systems G01S15/8979
• CPC Class Codes: G01S 7/52047 for elimination of side lobes or of grating lobes; for increasing resolving power
• CPC Class Codes: G01S 7/52077 with means for elimination of unwanted signals, e.g. noise or interference

Abstract

A method of ultrasound imaging is provided herein. The method includes the following stages: transmitting ultrasound radiation towards a target and receiving reflections of the ultrasound radiation from a region of the target in a main reflected signal and one or more auxiliary reflected signals, wherein each one of the reflected signals comprises an input dataset and is associated with a beam with a different and distinct beam pattern; compounding the input datasets from the main reflected signal and one or more auxiliary reflected signals, by the use of a compounding function, said compounding function using parameters derived from spatial analysis of the input datasets.

Claims

1. A method of ultrasound imaging, said method comprising:

transmitting ultrasound radiation towards a target and receiving reflections of the ultrasound radiation from a region of the target in a main reflected signal and one or more auxiliary reflected signals, wherein each one of the reflected signals comprises an input dataset and is associated with a beam with a different and distinct beam pattern;

compounding the input datasets from the main reflected signal and one or more auxiliary reflected signals, by the use of a compounding function, said compounding function using parameters derived from spatial analysis of the input datasets, said compounding function parameters being derived from at least one of;

(a) A local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ratio between difference receive beams, and/or functions of one or more of the aforementioned parameters;

(b) Spatial functions of the local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ration between difference beams;

(c) A local difference and/or ratio between spatial functions of the local magnitude and/or phrase and/or complex signal in the different receive beams;

(d) Applying multiple temporal band-pass filters to the local magnitude and/or phase and/or complex signal in the different receive beams, applying spatial analysis to the outputs of multiple temporal band-pass filters and compounding the results;

(e) Applying multiple temporal band-pass filters to the local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ratio between different receive beams, applying spatial analysis to the outputs of multiple temporal band-pass filters and compounding the results;

or

(f) Spatial-temporal functions of the local phase difference and/or magnitude difference and/or magnitude ratio between different beams, wherein spatial derivatives of the local phase difference and/or magnitude difference and/or magnitude ratio between different beams, are applied after local temporal filtering;

or wherein the local difference and/or ratio between spatial functions of the magnitude and/or phase and/or complex signal in the different beams are applied after local temporal filtering,

wherein the compounding function parameter is a generalized metric and is indicative of the probability for the corresponding voxel to be substantially affected by clutter effects only, as derived using certain physical and/or physiologic assumptions,

wherein the generalized metric is a normalized generalized metric and is defined as real numbers whose values range from 0.0 to 1.0, wherein the value 0.0 is assigned tovoxels that are substantially unaffected by clutter effects, and the value 1.0 is assigned to voxels in which substantially all of the measured signal emanates from clutter effects, and wherein the normalized generalized metric is the estimated probability for the corresponding voxel to be substantially affected by clutter effects only, and

wherein the compounding function is defined as a linear function of the local information for a main reflected beam and one or more associated auxiliary beams and wherein the compounding function is defined by the following equation;

S_out=[1−

m]S_main

wherein m is the normalized generalized metric;

S_main is the local measured signal for the main reflected beam; and

S_out is the clutter suppressed local signal.

2. The method of claim 1, wherein the output of the compounding function for any region of the target in any time frame may be used instead of the measured main reflected signal in any further processing performed by ultrasound scanners, said further processing being at least one of:

(a) Log-compression;

(b) Time-gain control;

(c) Global gain control;

or

(d) Doppler processing,

wherein said Doppler processing is one of;

color-Doppler flow imaging;

tissue-Doppler imaging;

pulse-Doppler studies; and

/or continuous-wave Doppler studies.

3. The method of claim 1, wherein the main reflected signal and the one or more auxiliary reflected signals result from a single transmitted pulse or from two or more transmitted pulses.

4. The method of claim 1, wherein the main reflected signal and the one or more auxiliary reflected signals are obtained in one of the following ways:

(a) All obtained concurrently, wherein the main reflected signal and the one or more auxiliary reflected signals are all associated with the same transmit pulse;

(b) All obtained at different times, wherein each of the main reflected signal and the one or more auxiliary reflected signals is associated with a different transmit pulse;

or

(c) Divided into groups of reflected signals, wherein each group of reflected signal may comprise a main reflected signal and/or one or more auxiliary reflected signals, wherein each group of reflected signals is obtained concurrently but different groups are obtained at different times, and wherein each group of reflected signals is associated with a different transmit pulse.

5. The method of claim 1, wherein the local phase difference and/or magnitude difference and/or magnitude ratio are indicative of the angular direction of the signal source.

6. The method of claim 1, wherein for non-clutter signals which originate from angular directions approximately corresponding to the center of the mainlobe of the beam associated with the main reflected signal, the signal level and/or the magnitude of the signal level change as a function of spatial location approximately the same way in the main reflected signal and in the one or more auxiliary reflected signals, wherein the change as a function of spatial location is defined by at least one of:

the local signal level;

the local spatial derivatives; and

the local normalized spatial derivatives, said local normalized spatial derivative being defined by the local spatial derivative divided by the local signal.

7. The method of claim 1, wherein temporal filters are employed to estimate the signal levels for target elements with predefined dynamic properties.

8. The method of claim 1, wherein the dataset collected by one or more reflected signals, in each time-swath or frame, is organized in a 1D, 2D or 3D scanned data array and the compounding function parameter is computed for a predefined subset of the frames, and in each frame, for each entry into the scanned data array or group of adjacent entries, and applied to all frames, wherein the parameters applied in a specific frame are the ones computed for the closest frame index, or alternatively, the closest frame index which is equal to or lower than the current frame index.

9. The method of claim 8, wherein spatial interpolation and/or interpolation in time is used in order to derive the compounding function parameters applied in a given frame.

10. The method of claim 1, wherein the generalized metric for a set of voxels in a set of frames is derived using the following four-step process:

(a) Select one or more compounding function parameters to be used (compounding function parameter set) from the following parameters;

(i) parameter #1;

the local magnitude ratio between two different reflected beams (a main and an auxiliary beam);

 

parameter #2;

the local phase difference between two different reflected beams;

(ii) parameter #1;

the local ratio of spatial magnitude derivatives along a first spatial axis between two different reflected beams;

 

parameter #2;

the local ratio of spatial magnitude derivatives along a second spatial axis between two different reflected beams;

(iii) parameter #1;

the local ratio of normalized spatial magnitude derivatives along a first spatial axis between two different reflected beams;

 

parameter #2;

the local ratio of normalized spatial magnitude derivatives along a second spatial axis between two different reflected beams;

 

wherein a normalized local signal derivative is defined as the local signal derivative divided by the local signal;

(iv) parameter #1;

the ratio of local magnitude spatial derivatives between two different receive beams, wherein the spatial derivative is computed using a “

Laplacian filter”

;

 

parameter #2 ;

the local magnitude ratio between two different beams;

(v) parameter #1;

the ratio of local normalized magnitude spatial derivatives between two different reflected beams, wherein the normalized spatial derivative is computed by dividing the local output of a “

Laplacian filter”

(applied to the local signal magnitude) by the local signal magnitude;

 

parameter #2;

the local magnitude ratio between two different reflected beams; and

(vi) parameter #1;

the ratio of the normalized magnitude spans between two different reflected beams, wherein a normalized magnitude span is defined as the difference between the maximal and minimal magnitudes (with or without outlier rejection) within a local region, divided by the mean magnitude in said region, wherein the local region is defined so that its width is greater than its height;

 

parameter #2;

the ratio of the normalized magnitude spans between two different beams, wherein the local region is defined so that its height is greater than its width;

(b) Define a model for at least one of;

(i) the probability (or a simple function thereof, such as the probability density function) for a voxel to be substantially affected by clutter effects only;

(ii) the probability (or a simple function thereof, such as the probability density function) for a voxel to be substantially unaffected by clutter effects; and

(iii) the percentage of the energy within the corresponding voxel that originates from clutter effects (or a simple function thereof);

as a function of the compounding function parameter set values or the values for a subset of the compounding function parameter set (compounding model);

(c) Compute the compounding function parameter set for each relevant voxel and/or its immediate vicinity, said immediate vicinity being defined in space and/or in time;

(d) For each relevant voxel, use the one or more compounding models to estimate at least one of the following (or a simple function thereof);

(i) the probability for its being substantially affected by clutter effects only;

(ii) the probability for its being substantially unaffected by clutter effects; and

(iii) the percentage of the energy within the corresponding voxel that originates from clutter effects.

11. The method of claim 1, wherein the compounding function parameter is a generalized metric, wherein a compounding function parameter set is a set of one or more compounding function parameters to be used, wherein a compounding function parameter subset is either a subset or a full set of a compounding function parameter set, wherein a compounding model is a model of at least one of:

(i) the probability (or a simple function thereof, such as the probability density function) for a voxel to be substantially affected by clutter effects only;

(ii) the probability (or a simple function thereof, such as the probability density function) for a voxel to be substantially unaffected by clutter effects; and

(iii) the percentage of the energy within the corresponding voxel that originates from clutter effects (or a simple function thereof);

as a function of the compounding function parameter set values or the values for a subset of the compounding function parameter set, and

wherein the compounding model may be defined in one of more of the following ways;

(a) A predefined model, which may be theoretically determined and/or experimentally derived;

(b) A simplified adaptive model, based on an assumption regarding the shape of the probability density function for a voxel to be substantially affected (or substantially unaffected) by clutter effects as a function of the applicable compounding function parameter subset (compounding model probability density function);

(c) An adaptive model, wherein the compounding model probability density function is computed directly, without any prior assumptions regarding the shape of the probability density function.

12. The method of claim 11, wherein voxels that are substantially unaffected by clutter are assumed to be more ubiquitous than voxels that are strongly affected by clutter.

13. The method of claim 11, wherein the compounding function parameter subset provides information regarding the local clutter levels and/or the angular direction of the primary source of the measured signal energy.

14. The method of claim 11, wherein the compounding model probability density function is computed using all relevant voxels in all frames at once.

15. The method of claim 11, wherein the dataset collected by one or more reflected signals, in each time-swath or frame, is organized in a 1D, 2D or 3D scanned data array, and wherein the compounding model probability density function is computed separately for each frame and/or each group of voxels, wherein the groups of voxels are separated in one or more scanned data array axes.

16. The method of claim 11, wherein the compounding model probability density function is utilized for estimating the local beam pattern within a given target region, and/or features of that beam pattern.

20. The method of claim 1, wherein the dataset collected by one or more reflected signals, in each time-swath or frame, is organized in a 1D, 2D or 3D scanned data array, wherein at least one of the following post-processing steps is applied to the output of the compounding function:

(a) For each applicable entry into the scanned data array and/or each applicable frame, replace the output of the compounding function with the corresponding voxel of the main receive beam if a certain criterion is met;

(b) For each applicable entry into the scanned data array and/or each applicable frame, replace the local output of the compounding function with the mean signal, weighted mean signal or median signal within a small local region of the compounding function output, defined in space and/or in time; and

(c) For each applicable entry into the scanned data array and/or each applicable frame (current voxel), apply the following process;

(i) Define a block within the scanned data array, approximately centered at the current voxel (data block), wherein said block may be one-dimensional, two-dimensional, three dimensional or four-dimensional;

(ii) For each of the main receive beam and the output of the compounding function, calculate the mean magnitude or the weighted mean magnitude within the data block, including or excluding the current voxel, and divide the result by the magnitude of the current voxel or by the mean magnitude in a smaller block within the scanned data array; and

(iii) Replace the value of the current voxel within the compounding function output by one of;

the value of the corresponding voxel in the main receive beam, the mean signal, weighted mean signal or median signal within the data block of the compounding function output;

if the ratio between the output of step (ii) for the main receive beam and for the output of the compounding function is higher-than or lower-than than a predefined constant.

21. The method of claim 1, wherein a plurality of main receive beams are used concurrently, and wherein at least one of the following:

(a) A compounding model probability density function;

(b) A generalized metric and/or a normalized generalized metric;

(c) A clutter suppression weight; and

(d) The compounding function;

is computed in at least one of the following ways;

(i) Employing each pair of main receive beam and associated auxiliary receive beam used (main-auxiliary beam pair) separately; and

(ii) Once for at least two main-auxiliary beam pairs, wherein transformations are applied to data relating to each main-auxiliary beam pair.

22. The method of claim 1, wherein the input datasets from the main reflected signal and the one or more auxiliary reflected signals are in one of the following forms:

(a) Analog signal;

or

(b) Digital signal.

23. The method of claim 1, wherein the input datasets from the main reflected signal and the one or more auxiliary reflected signals are in one of the following forms:

(a) Real representation;

or

(b) Complex representation.

24. The method of claim 1, wherein the input datasets from the main reflected signal and the one or more auxiliary reflected signals are taken from at least one of the following processing phases of ultrasound scanners:

(a) Before matched filtering and before down-conversion;

(b) Before matched filtering but after down-conversion;

(c) After matched filtering but before down-conversion;

(d) After matched filtering and after down-conversion.

17. A method of ultrasound imaging, said method comprising:

transmitting ultrasound radiation towards a target and receiving reflections of the ultrasound radiation from a region of the target in a main reflected signal and one or more auxiliary reflected signals, wherein each one of the reflected signals comprises an input dataset and is associated with a beam with a different and distinct beam pattern;

compounding the input datasets from the main reflected signal and one or more auxiliary reflected signals, by the use of a compounding function, said compounding function using parameters derived from spatial analysis of the input datasets, said compounding function parameters being derived from at least one of;

(a) A local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ratio between different receive beams, and/or functions of one or more of the aforementioned parameters;

(b) Spatial functions of the local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ratio between different beams;

(c) A local difference and/or ratio between spatial functions of the local magnitude and/or phase and/or complex signal in the different receive beams;

(d) Applying multiple temporal band-pass filters to the local magnitude and/or phase and/or complex signal in the different receive beams, applying spatial analysis to the outputs of multiple temporal band-pass filters and compounding the results;

(e) Applying multiple temporal band-pass filters to the local phase difference and/or magnitude difference and/or magnitude ratio and/or complex signal ratio between different receive beams, applying spatial analysis to the outputs of multiple temporal band-pass filters and compounding the results;

or

(f) Spatial-temporal functions of the local phase difference and/or magnitude difference and/or magnitude ratio between different beams, wherein spatial derivatives of the local phase difference and/or magnitude difference and/or magnitude ratio between different beams, are applied after local temporal filtering;

or wherein the local difference and/or ratio between spatial functions of the magnitude and/or phase and/or complex signal in the different beams are applied after local temporal filtering,

wherein the compounding function parameter is a generalized metric and is indicative of the probability for the corresponding voxel to be substantially affected by clutter effects only, as derived using certain physical and/or physiologic assumptions,

wherein the generalized metric is a normalized generalized metric and is defined as real numbers whose values range from 0.0 to 1.0, wherein the value 0.0 is assigned to voxels that are substantially unaffected by clutter effects, and the value 1.0 is assigned to voxels in which substantially all of the measured signal emanates from clutter effects, and wherein the normalized generalized metric is the estimated probability for the corresponding voxel to be substantially affected by clutter effects only, and

wherein the compounding function is computed by a linear combination of the outputs of another compounding function and the main reflected beam signal, wherein the weights are based on a normalized generalized metric, and wherein the compounding function is defined by the following equation;

18. The method of claim 17, wherein the clutter suppression weight for each applicable voxel {right arrow over (p)} and/or each applicable frame n is estimated by the following steps:

(a) Define a region surrounding ({right arrow over (p)}, u), wherein the region may be spatial and/or temporal (spatial-temporal region);

(b) Set the clutter suppression weight for({right arrow over (p)}, n) to be a function of at least one of;

(i) the values of one or more generalized metrics and/or normalized generalized metrics within the spatial-temporal region; and

(ii) the local complex ratio between the signal for the main receive beam, and the signal in one of the normalized auxiliary receive beams, within the spatial-temporal region (local complex main-auxiliary ratio).

19. The method of claim 18, wherein the clutter suppression weight for ({right arrow over (p)}, n) is set to the local complex main-auxiliary ratio for the voxel within the spatial-temporal region, whose generalized metric and/or normalized generalized metric values are most indicative of significant clutter effects;

wherein the computation of the clutter suppression weight is based on the following assumptions;

(a) The clutter signal contribution changes slowly enough in space and/or in time, such that it is approximately constant within small spatial-temporal regions; and

(b) Speckle effects in ultrasonic imaging cause local variations in the clutter free signal levels even for approximately homogeneous tissue regions, such that one may expect at least one of the voxels in the spatial-temporal region to include low free signal levels.