WO2019192970A1

WO2019192970A1 - Ultrasonic shear wave imaging with improved accuracy and reliability

Info

Publication number: WO2019192970A1
Application number: PCT/EP2019/058211
Authority: WO
Inventors: James Robertson JAGO
Original assignee: Koninklijke Philips N.V.
Priority date: 2018-04-02
Filing date: 2019-04-01
Publication date: 2019-10-10

Abstract

An ultrasonic imaging system for analyzing shear wave characteristics comprises a confidence factor processor producing confidence factors corresponding to points of a stiffness or velocity display map. A stiffness/velocity weighting processor generates a weighted average stiffness or velocity value based on confidence factor values for stiffness or velocity measurements in a region of interest.

Description

ULTRASONIC SHEAR WAVE IMAGING

WITH IMPROVED ACCURACY AND RELIABILITY

This application claims the benefit of and priority to U.S. Provisional No. 62/651285, filed April 2, 2018, which is incorporated by reference in its entirety.

This invention relates to medical ultrasound imaging systems and, in particular, to ultrasound systems which perform measurements of tissue

stiffness or elasticity using shear waves.

An advantage of ultrasound imaging and other imaging modalities is that, in addition to depicting the structure of tissue and pathology in the body, it is also possible to anatomically visualize

characteristics and functionality of the tissue or pathology being imaged. This is done by acquiring two images of the anatomy, one structural and another which is parametric. The two images are then

overlaid for display in anatomical registration. A basic parametric image in ultrasound is a colorflow image, whereby a B mode image of tissue structure is overlaid with a color image representing the

direction and velocity of blood flow in vessels and other structure of the tissue. The structure of blood vessel walls frames the blood flow information, showing the clinician parameters of the blood flow at the locations where it is occurring. The clinician can diagnose blood flow functionality at specific locations in the body by observing parameters of the flow such as its velocity and direction at anatomical locations defined by the surrounding tissue. Other parametric imaging procedures are also well known in ultrasound, such as tissue motion imaging, contrast imaging of tissue perfusion, and strain imaging of tissue elasticity.

Another parametric imaging procedure which has evolved more recently is shear wave imaging. Like strain imaging, shear wave imaging is an

elastographic technique which attempts to provide indications of tissue stiffness. For example,

stiffer tissue regions of the breast or liver might be malignant or scarred, whereas softer and more compliant areas are more likely to be benign and healthy. Since the stiffness of a region is known to correlate with malignancy or benignity, and scarred or healthy cells, elastography provides the clinician with another piece of evidence to aid in diagnosis and determination of a treatment regimen.

In order to form a shear wave image, shear wave measurements are made throughout a region of interest. The physiological phenomena behind an ultrasonic shear wave measurement are as follows. When a point on the body is compressed, then released, the

underlying tissue undergoes local axial displacement in the direction of the compression vector, then rebounds back when the compressive force is released. But since the tissue under the compressive force is continuously joined to surrounding tissue, the

uncompressed tissue lateral of the force vector will respond to the up-and-down movement of the local axial displacement. A rippling effect in this

lateral direction, referred to as a shear wave, is the response in the surrounding tissue to the

downward compressive force. Furthermore, it has been determined that the force needed to push the tissue downward can be produced by radiation pressure from an ultrasound pulse, and ultrasound reception can be used to sense and measure the tissue motion induced by the shear waves. Shear wave velocity is determined by local tissue mechanical properties.

The shear wave will travel at one velocity through soft tissue, and at another, higher velocity through stiffer tissue. By measuring the velocity of the shear wave at a point in the body, information is obtained as to characteristics of the tissue

stiffness at that point, such as its shear elasticity modulus and Young's modulus. The laterally

propagating shear wave travels slowly, usually a few meters per second or less, making the shear wave susceptible to detection, although it attenuates rapidly over a few centimeters or less. See, for example, US Pat. 5,606,971 (Sarvazyan) and US Pat.

5,810,731 (Sarvazyan et al . ) The shear wave velocity is virtually independent of the amplitude of tissue displacement, and tissue density normally has little variance, which make the technique suitable for objective quantification of tissue characteristics with ultrasound.

An ultrasonic imaging system which measures shear wave velocity with focused received lines called tracking lines is described in US pat. pub. no. 2013/0131511 (Peterson et al . ) One or more push pulses are transmitted into tissue with an ultrasound probe to ultrasonically compress the tissue in the vectorial direction of the push pulses. Immediately thereafter, focused tracking pulses are transmitted and received by the probe in the vicinity of the push pulse vector which generates the shear wave. Each tracking pulse vector is repetitively sampled in a time-interleaved manner so that tissue motion

produced by a shear wave can be detected when it occurs at each tracking pulse vector location,

preferably by correlating the echo data from

successive interrogations of the vector. As the shear wave moves laterally away from the push pulse vector, the positioning of the tracking pulses can also be moved laterally to follow the propagation of the shear wave. The data from the repetitively sampled tracking pulse vectors is processed to find the times at which the shear wave causes a peak displacement at each point of the tracking pulse vector, preferably by cross-correlation, curve fitting or interpolating successive displacement measurements. Analysis of the times at which points on adjacent sampling vectors experience peak

displacement produces a measurement corresponding to the velocity of the shear wave at particular vector locations, with velocity variations indicating tissues of different stiffness or elasticity. Since the shear waves attenuate rapidly, it is generally not possible to acquire shear wave data from an entire image field with a single push pulse vector. Thus, the process is repeated at other locations in the tissue to acquire shear wave velocity

measurements in other regions of the tissue. The process is repeated until shear wave data has been acquired over the desired image field. The velocity information is then presented as a two- or three- dimensional overlay of the structural B mode image of the tissue, color-coded by the shear wave velocity data or corresponding stiffness values at points in the image. In addition to reviewing the shear wave velocity or stiffness values as an image, the

clinician may also choose to define one or more regions of interest (ROI) within the image which are used to calculate average values of shear wave speed or stiffness within each ROI or across multiple ROIs. The benefit of calculating average values within or across ROIs is that the shear wave speed or stiffness of the patient's tissues may now be represented by a number, instead of an image, which can make it easier to classify severity of disease and also track changes in severity over time.

The tissue motion caused by shear wave travel is very subtle, however. Peak shear wave tissue

displacements are at best about 10 ym, and under more common, less favorable circumstances are closer to 1 ym. The precision of displacement estimates for accurate shear wave measurements should be at least on the order of 100 nm. Thus, the tissue

displacement resulting from shear wave travel can easily be overwhelmed by motional effects due to patient heartbeat and hand-held transducer movement. Efforts to enhance these measurements include the taking of repetitive measurements and the application of sophisticated filtering algorithms. Recently, Philips Healthcare of Andover, MA has provided clinicians with a confidence map for shear wave measurement images. The confidence map extends over the same region of the body as the shear wave image and depicts confidence factors over a range of percentages of reliability for areas where shear wave measurements were made. The clinician can study the map to see what the confidence factors are in regions of interest in the ultrasound image and is then able to select a confidence factor below which

measurements will be excluded from the ultrasound image and shear wave computations, including for example the calculation of average shear wave speed or stiffness values within or across ROIs. For example, the clinician may decide that only shear wave measurements with a confidence factor above 50% should be used in the image and for the calculation of average values. The 50% percentage is selected and measurements with confidence factors below 50% are excluded from calculations of stiffness in the ultrasound image. The clinician thus has a greater confidence in the accuracy and reliability of the stiffness measurements depicted in his diagnostic shear wave image and related computations such as average values.

A problem with this approach, however, is

inconsistent application. One clinician can use a 50% confidence factor cutoff, while another clinician favors a 70% cutoff. The different selections can yield different stiffness measurements for the same patient, which consequently will result in different values for average stiffness for that patient. A clinician can select different confidence factors for different patients or even for the same patient on different days, depending for instance upon the technical difficulty of the exam that day. Shear wave measurements and images with these differing results may thus not be suitable for serial studies or clinical decision-making. Accordingly, it is desirable to make shear wave measurements and form shear wave images which are consistent in terms of accuracy and reliability and that are less operator dependent .

The present disclosure generally relates to an ultrasonic imaging system for shear wave analysis.

According to certain aspects, the system includes an ultrasonic array configured to transmit a push pulse along a predetermined vector to generate a shear wave and to receive echo signals indicative of shear wave tissue displacement in a region of interest. The ultrasound array is typically included in a probe, such as a handle-held probe. One or more processors are in communication with the array and at least memory comprising instructions that when read by the processor cause the processor to execute one or more steps. In certain aspects, the processor is

configured to receive the echo signals, determine tissue stiffness or shearwave velocity values for points of the region of interest, produce confidence or reliability values for the points of the region of interest, and produce a weighted average tissue stiffness or weighted average velocity value for the region of interest based at least in part on the confidence or reliability values.

It is understood that the at least one processor configured to execute one or more steps may be one processor or a combination of processors. For example, processors may be particularized to carry out one or more actions. For example, the present disclosure describes several processors, including a shear wave processor, confidence factor processor, weighting processor, etc., and these processors may be the same or different.

In some aspects, the present disclosure includes ultrasonic imaging systems for shear wave analysis.

An ultrasound system of the present disclosure can include an ultrasonic array probe configured to transmit a push pulse along a predetermined vector to generate a shear wave, and to receive echo signals indicative of shear wave tissue displacement in a region of interest, a shear wave processor that is responsive to the received echo signals and is configured to determine tissue stiffness or shearwave velocity values for points of the region of interest, a confidence factor processor adapted to produce confidence or reliability values for the points of the region of interest, and a stiffness/velocity weighting processor that is adapted to produce a weighted average tissue stiffness or weighted average velocity value for the region of interest based at least in part on the confidence or reliability values.

In certain aspects, the present disclosure includes methods for shear wave analysis. A method of the present disclosure can include generating a shear wave, receiving echo signals indicative of shear wave tissue displacement in a region of

interest, determining tissue stiffness or shearwave velocity values for points of the region of interest, producing confidence or reliability values for the points of the region of interest, producing a

weighted average tissue stiffness or weighted average velocity value for the region of interest based at least in part on the confidence or reliability values.

In some aspects, the stiffness/velocity

weighting processor is further adapted to produce a plurality of weighted average tissue stiffness or weighted average velocity values for a plurality of regions of interest. The system can include an image processor that is adapted to produce a shear wave stiffness or velocity image. The image processor can be further adapted to overlay the shear wave

stiffness or velocity image onto an ultrasound image. The confidence factor processor can be further

adapted to produce confidence or reliability values based on a strength of detected shear waves. The confidence factor processor can be further adapted to produce confidence or reliability values based on at least one of tissue motion and blood flow location in the region of interest. The stiffness/velocity weighting processor can be further adapted to produce confidence or reliability values with a weight of zero for blood flow locations in the region of

interest. The system can include a confidence map memory adapted to store a map of confidence or reliability values. An image processor can be coupled to the confidence map memory and adapted to produce a confidence map image for display. The image processor can be further adapted to display both the confidence map image and a shear wave image. The system can further include a multiline receive beamformer. The shear wave processor can include a shear wavefront peak detector, an A-line r.f. cross correlator, and/or an A-line memory. The

stiffness/velocity weighting processor can be

configured to calculate the average tissue stiffness or weighted average velocity value using the

following equation:

Saverage = ( Si*Ci+S2*C2+...+S_n*C_n) / (Ci+C2+...C_n) , where S_aVerage is the weighted average tissue stiffness or the weighted average shear wave velocity value, S_n is the measured tissue stiffness or shear wave velocity value for a point n in a region of interest, and C_n is the confidence value for the point n .

In the drawings:

FIGURE 1 illustrates in block diagram form an ultrasonic imaging system in accordance with the principles of the present invention.

FIGURE 2 spatially illustrates a sequence of pulses along a push pulse vector, the resultant shear wavefront, and a series of tracking pulse vectors.

FIGURE 3 illustrates four laterally adjacent groups of 4x multiline tracking pulse vectors.

FIGURE 4 illustrates the transmission and reception of a 4x multiline pulse for the production of four adjacent multiline tracking pulse vectors in a region of interest.

FIGURE 5 illustrates shear wave displacement curves at two locations as it progresses through tissue .

FIGURE 6 illustrates an example embodiment for calculating a weighted average velocity or stiffness value according to the present invention.

In accordance with the principles of the present invention, an ultrasonic shear wave imaging system is described which improves the accuracy and reliability of shear wave measurements. In some embodiments, a confidence map is formed representing confidence factors for shear wave velocity or stiffness

measurements made at points in a shear wave image region. The confidence factors are used in

calculating a weighted average value for a shear wave velocity or stiffness measurement for the shear wave image region. The velocity or stiffness values are thereby compensated for variability in measurement confidence without the inconsistency arising from differences in user input. This is particularly significant for single stiffness or velocity

measurements calculated for regions of interest in an image .

Referring first to FIGURE 1, an ultrasound system constructed for the measurement of shear waves is shown in block diagram form. An ultrasound probe 10 has an array 12 of transducer elements for

transmitting and receiving ultrasound signals. The array can be a one dimensional or a two-dimensional array of transducer elements. Either type of array can scan a 2D plane and a two-dimensional array can be used to scan a volumetric region in front of the array. The array elements are coupled to a transmit beamformer 18 and a multiline receive beamformer 20 by a transmit/receive (T/R) switch 14. Coordination of transmission and reception by the beamformers is controlled by a beamformer controller 16. The

multiline receive beamformer produces multiple, spatially distinct receive lines (A-lines) of echo signals during a single transmit-receive interval.

The echo signals are processed by filtering, noise reduction, and the like by a signal processor 22, then processed by a shear wave processor comprising the following components. The received A-lines are stored in an A-line memory 24. Temporally distinct A-line samples relating to the same spatial vector location are associated with each other in an

ensemble of echoes relating to a common point in the image field. In a typical shear wave measurement system, the r.f. echo signals of successive A-line sampling of the same spatial vector are cross- correlated by an A-line r.f. cross-correlator 26 to produce a sequence of samples of tissue displacement for each sampling point on the vector. Alternatively, the A-lines of a spatial vector can be Doppler

processed to detect shear wave motion along the vector, or other phase-sensitive techniques can be employed. A wavefront peak detector 28 is responsive to detection of the shear wave displacement along the A-line vector to detect the peak of the shear wave displacement at each sampling point on the A-line.

In one embodiment this is done by curve-fitting, although cross-correlation and other interpolative techniques can also be employed if desired. The time at which the peak of the shear wave displacement occurs on an A-line is noted in relation to the times of the same event at other A-line locations, all to a common time reference, and this information is

coupled to a wavefront velocity detector 30 which differentially calculates the shear wave velocity from the peak displacement times on adjacent A-lines. This velocity information is coupled to a stiffness or velocity display map storage unit 32. A completed map indicates the tissue stiffness or velocity of the shear wave at spatially different points in a 2D or 3D image field scanned by push pulse transmission and shearwave acquisition by the probe 10. The stiffness or velocity display map is coupled to an image

processor 34 which processes the map for display, typically overlaying the velocity map on an

anatomical (B mode) ultrasound image of the tissue for display on an image display 36. The image

processor can also compute aggregate shear wave measurement values for display, such as the average stiffness of a region of interest of an image, using the stiffness values of a region of interest, e.g., pixel values .

FIGURE 2 is an illustration of the use of four push pulses to create a composite shear wavefront.

The four push pulses are transmitted along vectors 44, 54, 64 and 74 which are seen to be aligned along a single vectorial direction in FIGURE 2. When the shallowest push pulse of vector 44 is transmitted first, followed by successively deeper push pulses 54, 64, and 74, the shear wavefronts of the respective push pulses will have propagated as indicated by waves 46, 56, 66, and 76 by a time shortly after the last push pulse (vector 74) has been transmitted. As the shear waves 46, 56, 66, and 76 travel outward from the push pulse vector, they are interrogated by tracking pulses 80 shown in spatial sequence along the top of the drawing. Tracking pulses can occur between as well as after push pulses.

The velocity of the laterally traveling shear wave is detected by sensing the tissue displacement caused by the shear wave as it proceeds through the tissue. This is done with time-interleaved sampling pulses transmitted adjacent to the push pulse vector as shown in FIGURE 3. In this example the push pulse (s) 40 is transmitted along push pulse vector 44 to cause a laterally traveling shear wave. A-line vectors adjacent to the push pulse vector 40 are sampled by sampling pulses Tl, T2, T3, T4, and T5 transmitted along each vector in a time-interleaved sequence. For example, the first vector location A1 is sampled by a first pulse Tl, then the second vector location A2 by the next pulse T2, then A3, A4, and A5. Then vector location A1 is sampled again, and the sequence repeats at a pulse repetition frequency (PRF) . The interval between pulse

transmissions is referred to as the pulse repetition interval (PRI) . Since the sampling is time- interleaved, each of the five vector locations is sampled once in every five sampling pulses in this example. In this example every vector location is pulsed fifty-five times for a total tracking time of 27.5 msec. Each pulse results in echoes returning from along the vector which are sampled by a high speed A/D converter. Thus, for every sampled point along each vector there is an ensemble of 55 samples, with each sample taken at one-fifth the pulse rate of the T1-T5 sampling pulse sequence. The typical ensemble length at each echo location on a sampling vector is 40-100 samples. The sampling rate will be chosen in consideration of the frequency content of the shear wave displacement being detected so as to satisfy the Nyquist criterion for sampling. Since the purpose of the sampling is to sense and track the displacement effect of the shear wave as it

progresses through the tissue, the vector locations may be located closer together for slowly moving shear waves and further apart for more rapidly moving shear waves. Other sequences of time-interleaving the vector sampling may also be employed. For

example, odd-numbered vectors could be sampled in sequence, followed by sampling of the even-numbered vectors. As another example, vector locations A1-A3 could be sampled in a time-interleaved manner, then vector locations A2-A4, then vector locations A3-A5 to track the shear wave displacement as it progresses. Other sequences may also be employed based upon the exigencies of the situation. The ensembles of time- interleaved samples at each point along each sampling vector are then processed to find the time of peak tissue displacement at each point of each vector.

In accordance with one implementation, multiline transmission and reception is employed so that a single tracking pulse can simultaneously sample a plurality of adjacent, tightly spaced, A-line

locations. Referring to FIGURE 4, one technique for multiline transmission and reception is shown. In FIGURE 4 a single A-line tracking pulse with a beam profile 82a, 82b is transmitted as indicated by the wide arrow A#. The broad beam profile insonifies multiple receive line locations as shown in the drawing. Preferably the tracking pulse is a so- called "fat pulse" as described in US Pat. 4,644,795 (Augustine), for example. In this example four receive line locations Al-1, Al-2, Al-3 and Al-4 are insonified. Echoes from the four receive lines (4x multiline) are received in response to the single transmit pulse and are appropriately delayed and summed to produce coherent echo signals along each of the receive line locations. Beamformers capable of producing such simultaneous multilines are described, for instance, in US Pats. 5,318,033 (Savord) , 5,345,426 (Lipschutz) , 5,469,851 (Lipschutz) , and 6,695,783 (Henderson et al . ) These multiline

beamformers are typically used to decrease the acquisition time and thereby increase the frame rate of live ultrasound images, which is particularly useful when imaging the beating heart and blood flow in real time echocardiography. They are also useful in 3D ultrasound imaging so that real time frame rates of display can be attained. See, in this regard, US Pat. 6,494,838 (Cooley et al . ) In an implementation of the present invention, the benefit of multiline acquisition is two-fold: it enables a closely-spaced sampling line density and rapid acquisition of a short duration shear wave which only travels a short distance through tissue before being dissipated by attenuation. While higher order multiline may be employed which acquires samples along a greater number of A-lines at the same time and thus a higher sampling rate, this will require a broader transmit beam (A#) to simultaneously insonify the greater number of receive lines. The broader transmit beam will consequently diminish the signal- to-noise performance of the higher order

implementation .

FIGURE 3 illustrates the use of 4x multiline reception for transmission and reception along each sampling vector A1-A5. A first tracking pulse Ti is transmitted close to the push pulse vector 44, insonifying four receive line locations Al-1 to Al-4 and four multiline A-lines are received in response from the lateral region Al . When the four multilines are centered with respect to the transmitted tracking pulse, echoes from two A-lines are received on each side of the center of the tracking pulse beam center, shown by Al-1 and Al-2 to the left of center and Al-3 and Al-4 to the right of center. In a preferred embodiment the A-lines are spaced 0.5mm apart from each other. Shear waves generally move at a speed of 1-10 meters per second, and consequently tracking pulses are repetitively transmitted down regions Al- A5 in a time-interleaved manner and A-line samples received from the A-line locations during the time intervals between push pulses (when there are such intervals), and for 20 msec after the last push pulse, after which the shear wave has propagated out of the one centimeter A1-A5 sampling window. Since shear waves can have frequency components in the range of about 100Hz to about 1000Hz, sampling theory dictates that each A-line should have a sampling rate of 2kHz. This results in a set (ensemble) of fifty-five A-line samplings of each sampling point on each multiline A- line .

In the example of FIGURE 3, five tracking pulses, Ti-T₅, are transmitted over successive sampling

windows A1-A5 adjacent to the push pulse vector 44 to sample the shear wave displacement effect as the wave propagates. A typical sampling pulse is a short pulse, usually only one or two cycles, at a frequency suitable for penetrating the depth being studied, such as 7-8 MHz. Each tracking pulse is offset by 2mm from its adjacent neighbors, resulting in twenty A-lines spaced 0.5mm apart with 4x multiline over a total distance of one centimeter. There are various ways the interrogate the sampling windows. One is to just sample region A1 until the shear wave is

detected, then to begin sampling in region A2, then A3, and so on. Another is to time interleave the sampling in the regions as described above, sampling with tracking pulses T₁-T₅ in succession, then

repeating the sequence. With the latter approach five sampling windows with twenty tracking A-line positions can track the shear wave effect

simultaneously. After the shear wave has passed through the closest A1 sampling window and into the adjacent windows, sampling of the near window can be terminated and the sampling time can be devoted to the remaining sampling windows through which the shear wave is still propagating. Sampling continues until the shear wave has propagated out of the one cm. sampling region, by which time the shear wave has usually attenuated below a detectable level. Shear waves on average have a relaxation time of 10 msec.

It is necessary that the sampling times of the tracking A-line positions be related to a common time base when the tracking pulses are time-interleaved so that the results can be used to make a continuous measurement of time, and hence velocity, across the one cm. sampling region. For example, since the sampling pulses for sampling window A2 do not occur until 50 microseconds following the corresponding sampling pulses for window Al, a 50 microsecond time offset exists between the sampling times of the two adjacent windows. This time difference must be taken into account when comparing the peak times of

displacement in the respective windows, and must be accounted for in an accumulated manner across the full one centimeter sampling window. Referencing the sampling times of each sampling vector to a common time reference can resolve the problem of the offset sampling times.

Since a diagnostic region-of-interest (ROI) is generally greater than one centimeter in width, the procedure of FIGURES 2-5 is repeated with push pulses transmitted at different lateral locations across the image field. An image field is thereby interrogated in one-centimeter wide regions, and the results of the interrogations are displayed adjacent to each other in anatomical relationship to present an image of the full ROI . A four centimeter wide image field can be interrogated in four adjacent or overlapping one cm. regions, which are then displayed side-by- side or wholly or partially overlaid on the display.

FIGURE 5 illustrates a sequence of displacement values for two laterally adjacent points of tissue on two adjacent A-lines such as Al-3 and Al-4 in FIGURE 3. Curve 100 represents the displacement over time caused by passage of a shear wave through a point on A-line Al-3, and curve 120 the displacement at an adjacent point of A-line Al-4. Points 102-118 of tissue displacement values are calculated from local cross correlations of r.f. data (e.g., 10-30 r.f. samples in depth) acquired around a sampling point depth on Al-3 over time to yield the local

displacement values over time at the depth point.

The points 102-118 of displacement values detected at successive times (y-axis) , when plotted as a function of time, are joined to form the first displacement curve 100. At a point on second A-line A-l-4 spaced to the right of the point on the first A-line, the succession 122-136 of displacement values produced by local cross correlation can be joined to form a second displacement curve 120. Since the shear wave is traveling from left to right in this example, the second curve 120 for the right-most A-line is shifted to the right (in time) of the first displacement curve 100. A precise time reference of the passage of the wavefront from one point to the next is measured by the detected peak or inflection point of each displacement curve, indicated at 200 and 220 in this example. Various techniques can be used to find the curve peak. In one implementation the

displacement values of each curve are processed by fitting curves to the values to form complete

displacement curves 100, 120 and the curve peaks. Another technique is to interpolate additional points between the detected points to find the peak. Yet another technique is to determine the slopes of the curve on either side of the peak and determine the peak from the intersection of slope lines. Still another approach is cross-correlation of the curve data. When the peaks of the shear wave displacement at successive A-line positions are found by the waveform peak detector 28, their times of occurrence in relation to the detection of the points on the curves are noted. The difference of these times, At, taking into consideration sampling time offsets, and the spacing between the A-lines (e.g., 0.5mm) can then be used by the wavefront velocity detector 30 to determine the velocity of the shear wave as it traveled between the two A-line locations. After the entire ROI has been interrogated in this manner and displacement curves and times of peak occurrence determined for each sample point on each A-line vector, the velocity of shear wave travel can be calculated from one image point to another across the entire region of interest. This two- or three- dimensional matrix of velocity values is color-coded or otherwise coded with corresponding stiffness estimates to form a velocity or stiffness display map which is overlaid and in spatial alignment with a B mode image of the region of interest for display on image display 36.

It is seen that the ultrasound system of FIGURE 1 has a B mode image processor 48. The B mode processor receives echo signals acquired in response to image pulse transmission and performs amplitude detection of the echoes for production of B mode images of tissue. Generally, B mode processing also includes logarithmic compression of the amplitude values to produce a more diagnostic range of

grayscale values for the B mode image. It is with such a B mode image that the parametric map of stiffness or velocity values can be overlaid in anatomical registration for display on image display 36.

The ultrasound system of FIGURE 1 further comprises a confidence factor processor 40 which produces confidence or reliability values. The confidence or reliability values can be stored in a confidence map memory 42 for a confidence value map of the anatomy spatially associated with a stiffness or shear wave velocity map. The confidence factor processor 40 operates by combining several types of information which bear upon the reliability of the shear wave measurements. For example, the confidence factor processor receives A-line values from the A- line memory 24, which indicate the amplitudes of the tissue displacements detected by the tracking pulses. These values indicate the strength of the shear waves detected at points in the measurement field. Strong shear wave signals are generally more reliable than weaker shear wave signals. The confidence factor processor also receives B mode images which provide two other types of information. Successive B mode images are compared or correlated to determine whether local or global tissue motion has occurred during shear wave measurement, either due to motion of the tissue or of the probe. This motion can be determined by correlation of pixel values at the same location in successive images, or by block matching of pixel areas as is known in the art. The presence of tissue motion indicates that shear wave tissue displacement estimates are contaminated by motion from other sources. The B mode images can also reveal the locations of blood vessels in the image field, which appear as dark areas in B mode images. Since the shear wave measurements interrogate tissue, signals from vessel lumens should be omitted from shear wave measurement computation. Alternatively, color Doppler, power Doppler, or contrast detection can be used to identify regions of blood flow. These sources of information are combined for points in an image, e.g., in each pixel location, in the stiffness or velocity map to determine a confidence or

reliability factor for each point in the map. The types of information can be combined in various ways. For example, the confidence factor processor can calculate a confidence factor using the expression

CF = 1.0 * Ampl_sw + 0.5 * TM + 1.0 * I_BM where the confidence factor (CF) for an image point is the sum of the shear wave amplitude (Ampl_s„) plus one-half of the tissue motion value (TM) plus the B mode image pixel value, where vessel lumens have a value of zero and tissue pixels a value of one. The raw values of these three variables may be used, or they may be normalized if desired. For instance, if the shear wave amplitude varies between zero and eight pm, this range can be converted to a range of zero to one or 0% to 100%. The converted range can be linear, or nonlinear to emphasize (or de-emphasize) one or both range extremes. The confidence factor processor produces a confidence factor value for each location in the stiffness or velocity display map.

The raw confidence factor values can be used, or the values can be converted to a range of 0% to 100% by use of a conversion look-up table. As before, the conversion can be either linear or nonlinear to emphasize or de-emphasize different parts of the range .

These confidence factor values for each location in the shear wave display map are stored as a map of confidence factors in confidence map memory 52. In one embodiment, the confidence map is continually updated for each newly acquired stiffness or shear wave velocity map. In accordance with the principles of the present invention, the confidence map values can be further processed by a stiffness/velocity weighting processor 50 where they are used to weight the spatially corresponding values of the stiffness or shear wave velocity display map when calculating a weighted average value of stiffness or shear wave velocity within an ROI or weighted average values of stiffness or shear wave velocity within multiple ROIs. The user can in certain embodiments engage this use of the confidence factors by actuating a control of the user interface, or the system can automatically engage the use of the confidence factors.

In an example embodiment, the stiffness/velocity weighting processor can produce one or more weighted average tissue stiffness or weighted average velocity values using the confidence or reliability values.

Determining the weighted average values can be

calculated using the following equation:

S_average = ( Si*Ci+S2*C2+...+S_n*C_n) / (Ci+C2+...C_n) , where S_aVerage is the weighted average tissue stiffness or the weighted average shear wave velocity value, S_n is the measured tissue stiffness or shear wave

velocity value for a point n in a region of interest (e.g., where such point can be a pixel in an image), and C_n is the confidence value for the point n . In this approach, the sum of the confidence values of all points in the region of interest being averaged are used in the denominator to reduce or remove bias in the calculated weighted average tissue stiffness or the weighted average shear wave velocity value. This weighting process can advantageously produce an image in which questionable stiffness and velocity values are only dimly perceived while more reliable values are brightly displayed, and questionable stiffness and velocity values play a lesser role in regional stiffness computations. The user can choose to display the weighted shear wave image alone, or together with a confidence map from the confidence map memory 52.

FIGURE 6 shows a visual example of an embodiment of the present disclosure. As shown, a display 90 shows an ultrasound image 92, which can include B- mode image information. A stiffness or shear wave velocity map bounded by a region of interest 94 can be overlaid on the ultrasound image 92. As described herein, confidence or reliability values can be calculated for points 96 in the map. For example, each gray block in FIGURE 6 represents a point for which a measured stiffness or shear wave velocity value S_n can be calculated. Similarly, a confidence or reliability value C_n can be calculated for each point. The sum of the confidence values of all points in the region of interest 94 being averaged are used in the denominator to reduce or remove bias in a calculated weighted average tissue stiffness or the weighted average shear wave velocity value for the region of interest 94 that includes points n. In some embodiments, a weighted average value of stiffness or shear wave velocity can be calculated for more than one region. For example, four different regions can be positioned or drawn in different areas of a tissue of interest, e.g., a liver. A weighted average a weighted average value of stiffness or shear wave velocity can be calculated for each, and the ultrasound system can further calculate additional information from the weighted average values of stiffness or shear wave velocity.

This use of the confidence factor values

provides a number of advantages. One advantage is that tissue stiffness and/or shear wave velocity measurements will be more consistent from one

clinician to another and from one exam to another, since the weighting process is automatic and involves no user-selected variable. This means that serial studies of a patient over a period of time will be more consistent and reliable. A second advantage is that weighted average stiffness or velocity

measurements of a region will be more consistent and reliable. This is due to the fact that the stiffness or velocity measurements throughout the region are less affected by a user-selected variable, and more reliable values weigh more strongly than less

reliable values in regional measurements and

computations. A third advantage is that there is no need to carefully trace around blood vessels in an image when drawing a region of interest. Since clinicians know that stiffness measurements should only come from tissue and not blood flow, past practice has been for a clinician to carefully outline a region of interest that includes only tissue and excludes blood vessel segments from the region. The use of the confidence factor computation described above, where regions of blood revealed in a B mode or motion image are given a weight of zero in the weighting process, automatically excludes these regions from stiffness and velocity computations through zero weighting, sparing the clinician from this tedious tracing task and its inherent user variability. The system of the present invention is well suited to clinicians such as hepatologists who generally prefer to spend less time adjusting

controls and studying images and are primarily concerned with the ease and reliability of obtained regional stiffness measurements.

The confidence factor weighting techniques of the present invention are applicable to ultrasound, and to other diagnostic modalities which measure tissue stiffness noninvasively . For instance, the techniques of the present invention are also

applicable to magnetic resonance elastography .

It should be noted that the ultrasound system of FIGURE 1 which measures shear wave speed and derived measurements of stiffness may be implemented in hardware, software or a combination thereof. The various embodiments and/or components of an

ultrasound system, for example, the modules, or components and controllers therein, also may be implemented as part of one or more computers or microprocessors. The computer or processor may include a computing device, an input device, a display unit and an interface, for example, for accessing the Internet. The computer or processor may include a microprocessor. The microprocessor may be connected to a communication bus, for example, to access a PACS system. The computer or processor may also include a memory. The memory devices such as the A-line memory 24 and the confidence map memory 52 may include Random Access Memory (RAM) and Read Only Memory (ROM) . The computer or processor further may include a storage device, which may be a hard disk drive or a removable storage drive such as a floppy disk drive, optical disk drive, solid-state thumb drive, and the like. The storage device may also be other similar means for loading computer programs or other instructions into the computer or processor.

As used herein, the term "computer" or "module" or "processor" as used in describing the B mode image processor 48, the confidence factor processor 40, and the stiffness/velocity weighting processor 50, may include any processor-based or microprocessor-based system including systems using microcontrollers, reduced instruction set computers (RISC) , ASICs, logic circuits, and any other circuit or processor capable of executing the functions described herein. The above examples are exemplary only, and are thus not intended to limit in any way the definition and/or meaning of these terms.

The computer or processor executes a set of instructions that are stored in one or more storage elements, in order to process input data. The

storage elements may also store data or other

information as desired or needed. The storage

element may be in the form of an information source or a physical memory element within a processing machine .

The set of instructions of an ultrasound system, including the shear wave measurement, confidence factor computing, and stiffness/velocity weighting computations described above, may include various commands that instruct a computer or processor as a processing machine to perform specific operations such as the methods, computations, and processes of the various embodiments of the invention. The set of instructions may be in the form of a software program. The software may be in various forms such as system software or application software and which may be embodied as a tangible and non-transitory computer readable medium. Further, the software may be in the form of a collection of separate programs or modules, a program module within a larger program or a portion of a program module. For example, the expressions calculated by the confidence factor and

stiffness/velocity weighting processors may be executed by software modules calculating the

equations. The software also may include modular programming in the form of object-oriented

programming. The processing of input data by the processing machine may be in response to operator commands, or in response to results of previous processing, or in response to a request made by another processing machine.

Furthermore, the limitations of the following claims are not written in means-plus-function format and are not intended to be interpreted based on 35 U.S.C. 112, sixth paragraph, unless and until such claim limitations expressly use the phrase "means for" followed by a statement of function devoid of further structure.

Claims

WHAT IS CLAIMED IS:

1. An ultrasonic imaging system for shear wave analysis comprising:

an ultrasonic array configured to transmit a push pulse along a predetermined vector to generate a shear wave, and to receive echo signals indicative of shear wave tissue displacement in a region of

interest; and

at least processor in communication with the array and at least memory comprising instructions that when read by the processor cause the processor to :

receive the echo signals;

determine tissue stiffness or shearwave velocity values for points of the region of interest;

produce confidence or reliability values for the points of the region of interest; and

produce a weighted average tissue stiffness or weighted average velocity value for the region of interest based at least in part on the confidence or reliability values.

2. The ultrasonic imaging system of Claim 1, wherein the processor is further configured to produce a plurality of weighted average tissue stiffness or weighted average velocity values for a plurality of regions of interest.

3. The ultrasonic imaging system of Claim 1, wherein the processor is further configured to produce a shear wave stiffness or velocity image.

4. The ultrasonic imaging system of Claim 3, wherein the processor is further configured to overlay the shear wave stiffness or velocity image onto an ultrasound image.

5. The ultrasonic imaging system of Claim 1, wherein the processor is further configured to produce confidence or reliability values based on a strength of detected shear waves.

6. The ultrasonic imaging system of Claim 1, wherein the processor is further configured to produce confidence or reliability values based on at least one of tissue motion and blood flow location in the region of interest.

7. The ultrasonic imaging system of Claim 6, wherein the processor is further configured to produce confidence or reliability values with a weight of zero for blood flow locations in the region of interest.

8. The ultrasonic imaging system of Claim 1, wherein the memory is further configured to store a map of confidence or reliability values.

9. The ultrasonic imaging system of Claim 8, wherein the processor is further configured to produce a confidence map image based on the

confidence or reliability values.

10. The ultrasonic imaging system of Claim 9, wherein the processor is further configured to display both the confidence map image and a shear wave image .

11. The ultrasonic imaging system of Claim 1, further comprising a multiline receive beamformer.

12. The ultrasonic imaging system of Claim 1, wherein the processor is further configured to detect a peak in a wavefront of the shear wave.

13. The ultrasonic imaging system of Claim 12, wherein the processor is further configured to determine a cross-correlation of r.f. data associated with the shear wave.

14. The ultrasonic imaging system of Claim 13, wherein the processor further comprises an A-line memory .

15. The ultrasonic imaging system of Claim 1, wherein the processor is configured to calculate the average tissue stiffness or weighted average velocity value using the following equation: