IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 32, NO. 8 ......comb-push beams and track shear wave motions with a linear array transducer L7-4 (Philips Healthcare, Andover, MA, USA)

IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 32, NO. 8, AUGUST 2013 1435

Comb-Push Ultrasound Shear Elastography (CUSE)With Various Ultrasound Push Beams

Pengfei Song, Student Member, IEEE, Matthew W. Urban, Member, IEEE, Armando Manduca, Member, IEEE,Heng Zhao, Member, IEEE, James F. Greenleaf, Life Fellow, IEEE, and Shigao Chen*, Member, IEEE

Abstract—Comb-push ultrasound shear elastography (CUSE)has recently been shown to be a fast and accurate 2-D elasticityimaging technique that can provide a full field-of-view (FOV)shear wave speed map with only one rapid data acquisition.The initial version of CUSE was termed U-CUSE because un-focused ultrasound push beams were used. In this paper, wepresent two new versions of CUSE—focused CUSE (F-CUSE)and marching CUSE (M-CUSE), which use focused ultrasoundpush beams to improve acoustic radiation force penetrationand produce stronger shear waves in deep tissues (e.g., kidneyand liver). F-CUSE divides transducer elements into severalsubgroups which transmit multiple focused ultrasound beamssimultaneously. M-CUSE uses more elements for each focusedpush beam and laterally marches the push beams. Both F-CUSEand M-CUSE can generate comb-shaped shear wave fields thathave shear wave motion at each imaging pixel location so that afull FOV 2-D shear wave speed map can be reconstructed withonly one data acquisition. Homogeneous phantom experimentsshowed that U-CUSE, F-CUSE, and M-CUSE can all producesmooth shear wave speed maps with accurate shear wave speed es-timates. An inclusion phantom experiment showed that all CUSEmethods could provide good contrast between the inclusion andbackground with sharp boundaries while F-CUSE and M-CUSErequire shorter push durations to achieve shear wave speed mapswith comparable SNR to U-CUSE. A more challenging inclusionphantom experiment with a very stiff and deep inclusion showsthat better shear wave penetration could be gained by usingF-CUSE and M-CUSE. Finally, a shallow inclusion experimentshowed that good preservations of inclusion shapes could beachieved by both U-CUSE and F-CUSE in the near field. Safetymeasurements showed that all safety parameters are below FDAregulatory limits for all CUSE methods. These promising resultssuggest that, using various push beams, CUSE is capable ofreconstructing a 2-D full FOV shear elasticity map using only onepush-detection data acquisition in a wide range of depths for softtissue elasticity imaging.

Manuscript received March 01, 2013; revised April 04, 2013; accepted April05, 2013. Date of publication April 12, 2013; date of current version July 27,2013. This work was supported by the National Institute of Biomedical Imagingand Bioengineering under Grant EB002167 and Grant DK082408. This workwas supported by the National Institutes of Health (NIH) under Grant EB002167and Grant DK082408. The content is solely the responsibility of the authorsand does not necessarily represent the official views of NIH. Mayo and someof the authors have financial interest in the technology described here. Asteriskindicates corresponding author.P. Song is with Mayo Graduate School and the Department of Physiology

and Biomedical Engineering, Mayo Clinic College of Medicine, Rochester, MN55905 USA (e-mail: [email protected]).M. W. Urban, A. Manduca, H. Zhao, and J. F. Greenleaf are with the De-

partment of Physiology and Biomedical Engineering, Mayo Clinic College ofMedicine, Rochester, MN 55905 USA (e-mail: [email protected]).*S. Chen is with the Department of Physiology and Biomedical Engineering,

Mayo Clinic College of Medicine, Rochester, MN 55905 USA (e-mail: [email protected]).Color versions of one or more of the figures in this paper are available online

at http://ieeexplore.ieee.org.Digital Object Identifier 10.1109/TMI.2013.2257831

Index Terms—Acoustic radiation force, comb-push, comb-pushultrasound shear elastography (CUSE), focused ultrasound beam,shear wave, ultrasound elastography, unfocused ultrasound beam.

I. INTRODUCTION

F AST and accurate tissue elasticity imaging is an essen-tial task in ultrasound shear wave elastography. By in-

ducing shear waves into the tissue and ultrasonically trackingshear wave motion, ultrasound shear elastography is capable ofsolving for the shear modulus, , of soft tissue (assuming in-compressibility, isotropy, linearity, and pure elasticity) by [1]

(1)

where is shear wave propagation speed, is density andcan be assumed to be 1000 for all soft tissues [2].Shear waves can be produced by pushing the soft tissue withacoustic radiation force, as proposed by Sarvazyan et al. [1] inshear wave elasticity imaging (SWEI). Nightingale et al. devel-oped acoustic radiation force impulse (ARFI) imaging whichwas used to remotely palpate soft tissues [3] or generate shearwaves in the tissue [4]. Bercoff et al. invented supersonic shearimaging (SSI), in which a conical-shaped shear wave was gen-erated frommultiple ultrasound push beams focused at differentdepths [5]–[7].McAleavey et al. developed spatially-modulatedimpulse acoustic radiation force (SMURF) to generate shearwaves with known spatial frequency so that shear wave speedcan be calculated from motion measurements at a single spatiallocation [8]. Chen et al. proposed shear wave dispersion ultra-sound vibrometry (SDUV) which generates shear waves at mul-tiple frequencies with acoustic radiation force to characterizetissue elasticity and viscosity with shear wave dispersion anal-ysis [9], [10]. Hazard et al. implemented acoustic radiation forceto produce a synthetic crawling wave and solve for shear wavespeed from the interfering crawling wave patterns [11], [12].Zhao et al. recently used unfocused acoustic radiation force toproduce shear waves and achieved robust shear wave speed es-timates throughout a long axial extent in both phantoms and bi-ceps muscles [13].For elasticity imaging methods that use acoustic radiation

force to generate shear waves, shear waves propagate in oppo-site directions away from the push beam. Consequently, there isno propagating shear wave in the push beam region and shearwave speed cannot be calculated there. Meanwhile shear wavesmay be significantly attenuated in areas that are far away fromthe push beam region. Therefore, multiple data acquisitions with

0278-0062/$31.00 © 2013 IEEE

1436 IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 32, NO. 8, AUGUST 2013

Fig. 1. Schematic plots of different CUSE imaging sequences. (a) U-CUSE: transmit multiple unfocused ultrasound push beams simultaneously at time .(b) F-CUSE: transmit multiple focused ultrasound beams simultaneously at time . (c) M-CUSE: transmit single focused ultrasound beam using element sub-group 1 at time , then march laterally to push with subgroup 2 at time , subgroup 3 at time , and subgroup 4 at time . Time interval between the end andstart of consecutive push beams was 15 .

push beams transmitted at different locations are typically re-quired to reconstruct a full field-of-view (FOV) 2-D shear elas-ticity map [14]. Recently, Song et al. proposed the comb-pushultrasound shear elastography (CUSE) method that is capableof providing a full FOV 2-D shear wave speed map under theentire width of the transducer with only one rapid data acquisi-tion [15], [16]. CUSE uses a comb-push to generate multipleshear waves that can cover the entire FOV so that areas in-cluding the push beam regions can be reconstructed. Moreover,because each imaging pixel always has one or more push beamsnearby, higher SNR shear waves can be obtained in CUSE. Byuse of a directional filter, CUSE is capable of differentiatingthe left-to-right (LR) propagating and right-to-left (RL) propa-gating shear waves so that an accurate shear wave estimate canbe achieved at each pixel location using time-of-flight (TOF)calculations. This version of CUSE used unfocused ultrasoundpush beams to generate shear waves and thus is termed Unfo-cused CUSE (U-CUSE) here.To improve acoustic radiation force penetration and gen-

erate stronger shear waves in deeper tissue, in this paper, wepropose two new versions of CUSE that use focused ultra-sound push beams. The first version divides the transducerelements equally into subgroups, which transmit several fo-cused ultrasound beams simultaneously and is termed focusedCUSE (F-CUSE). The second version uses more transducerelements to transmit a focused ultrasound push beam with alower F-number and the push elements rapidly march alongthe lateral direction to push at different horizontal locations.This version of CUSE is termed marching CUSE (M-CUSE).Similar to U-CUSE, both F-CUSE and M-CUSE can generatecomb-patterned ultrasound push beams. As in U-CUSE, adirectional filter can be used for both F-CUSE and M-CUSE toremove the interferences and separate the LR and RL waves sothat robust shear wave speed estimates can be achieved at eachimaging pixel within the FOV.In this paper, we first introduce the principles of F-CUSE

and M-CUSE, including the push beam sequences, shear wavemotion detection, directional filtering, and the shear wave speed

map reconstruction. Then we describe phantom experimentsincluding homogeneous phantoms and inclusion phantoms toassess the relative performance of the three techniques in avariety of situations. We close the paper with discussion andconclusions.

II. MATERIALS AND METHODS

A. Principles of F-CUSE and M-CUSE

A Verasonics ultrasound system (Verasonics Inc., Red-mond, WA, USA) was used in this study to producecomb-push beams and track shear wave motions with alinear array transducer L7-4 (Philips Healthcare, Andover,MA, USA). Schematic plots of U-CUSE, F-CUSE, andM-CUSE are shown in Fig. 1. For all CUSE methods, thetransducer elements were divided into subgroups, i.e., sub-groups 1–9 for U-CUSE. In U-CUSE, subgroups 1, 3, 5,7, and 9 simultaneously transmit unfocused push beamscenter frequency MHz push durationwhile subgroups 2, 4, 6, and 8 are turned off. There are 12elements in each push beam and 17 elements in between pushbeams. For F-CUSE, the transducer elements were dividedinto four subgroups and each subgroup has 32 elements. Allsubgroups transmit focused ultrasound beams simultaneouslycenter frequency MHz push duration . ForM-CUSE, the transducer was divided into four subgroups aswell but with overlapping elements. Each subgroup has 64elements. Subgroup 1 transmits a single focused push beamcenter frequency MHz push durationat time , and then marches to subgroup 2 to transmit thesecond focused push beam at time . The marching continuesthrough subgroups 3 and 4 and terminates. The time intervalbetween the end and start of consecutive push beams was 15due to hardware limitations. Note that each focused beam forM-CUSE has 200 push duration, which is one third of bothU-CUSE and F-CUSE. This is because the center portion of thetransducer elements are transmitting three times, as shown inFig. 1(c). A 200 push duration was used for M-CUSE so that

SONG et al.: COMB-PUSH ULTRASOUND SHEAR ELASTOGRAPHY (CUSE) WITH VARIOUS ULTRASOUND PUSH BEAMS 1437

Fig. 2. Plots of particle axial velocity at different time steps for F-CUSE and M-CUSE in a homogeneous elastic phantom with shear wave speed of about 1.5 m/s.Shear waves from different push beams interfere with each other and eventually fill the entire FOV. Left column: F-CUSE: (a), (c), and (e) show that four shearwave sources were generated by the four focused push beams. Each push beam generates two shear wave fronts that propagate away from the push beam. Asindicated by the white arrow, the left-to-right shear wave from subgroup 1 appears in (c) and merges with the right-to-left shear wave from subgroup 2 in (e). Rightcolumn: M-CUSE. (b), (d), (f) show that four shear waves were generated by the four focused push beams. As indicated by the white arrow, the right-to-left shearwave from subgroup 4 appears in (d) and merges with the left-to-right wave from subgroup 3 in (f). Colorbar is in units of mm/s and the scale is different for eachtime step. “x” represents the lateral dimension. “z” represents the axial dimension.

the maximum push duration for these center elements wouldstill be 600 , to allow a fair comparison to both U-CUSE andF-CUSE in terms of transducer heating.For all CUSE methods, after comb-push transmis-

sion, the Verasonics system immediately switched toplane wave imaging mode using all transducer elementscenter frequency MHz). A plane wave imaging com-pounding method was used to improve the signal-to-noise-ratio(SNR) of shear wave tracking [17]. Three frames at threedifferent steering angles were used to obtain oneimaging frame, with a spatial resolution of one ultrasoundwave-length ( mm assuming ultrasound speed )and effective frame rate of 3.9 kHz.

B. Shear Wave Motion Estimation

The shear wave propagation induced axial particle velocitywas evaluated from in-phase/quadrature (IQ) data of con-

secutive frames tracked by the Verasonics system. The one-di-mensional autocorrelation method [18] was used to calculatefor each imaging pixel. The shear wave motion was obtainedfrom three pixels in space and two sampling points in the slowtime direction. Then a 3 3 pixel spatial median-filter (0.92mm0.92 mm) was used on each frame of the shear wave motion

image to remove noise spike points. Fig. 2 shows the snapshotsof shear wave motions at different time steps for F-CUSE andM-CUSE in a homogeneous phantom. Examples of shear wavemotions for U-CUSE are given in [15].


Fig. 3. Directional filtering. (a) Original shear wave field created by F-CUSEbefore directional filtering. (b) The 2-D Fourier transform of (a). (c) Directionalfilter masked out the second and fourth quadrants (corresponding to RL shearwaves) and preserved the first and third quadrants (corresponding to LR shearwaves). (d) Extracted LR shear waves by a 2-D inverse Fourier transform of (c).

C. Directional Filtering

To remove the shear wave interferences and achieve robustshear wave speed estimates, a directional filter similar to [19],[20] was used in this study to separate the left-to-right (LR)and right-to-left (RL) propagating shear waves. The shear wavefield data has two dimensions in space [lateral dimension (x)and axial dimension (z)] and one dimension in time [slow time(t)]. Fig. 3(a) shows a slice of the F-CUSE shear wave field datawith axes of lateral dimension (x) and slow time (t). The depthof the slice is at the focal plane of the focused push beams (25mm). A 2-D Fourier transform of the shear wave field yieldsa symmetric spectrum as shown in Fig. 3(b), with the first andthe third quadrants corresponding to the LR shear waves andthe second and the fourth quadrants corresponding to the RLshear waves. By designing a mask as shown in Fig. 3(c), onecan extract the LR shear waves [Fig. 3(d)] by preserving thefirst and third quadrants of the specturm while masking out thesecond and the fourth. A complementary mask to Fig. 3(c) willextract the RL shear waves by preserving the second and thefourth quadrants of the spectrum while masking out the first andthe third. The mask edges of the directional filters have beenapodized to minimize ripples [19].

D. Local Shear Wave Speed Recovery and 2-D Shear WaveSpeed Map Reconstruction

A time-of-flight algorithm based on cross-correlating shearwave motion profiles along the lateral direction was used inthis study to calculate shear wave propagation speed. Theshear wave speed of an imaging pixel was calculated fromtwo neighbor pixel points separated by eight ultrasound wave-lengths (eight imaging pixels) at the same depth [14]. Tofacilitate more robust cross-correlation, the shear wave motionprofiles were Tukey windowed (the ratio of tapered section to

constant section is 0.25) [21] so that both ends of the signalwere forced to be zero. The shear wave motion profiles werepre-interpolated by a factor of five before cross-correlation(using the “interp” function in MATLAB). Two-dimensionalshear wave speed maps can be obtained from both the LR shearwave field and RL shear wave field. The final shear wave speedmap is reconstructed using the same method as proposed in[15]. Fig. 4 shows an example of a final map reconstructionusing F-CUSE. Recon1 [Fig. 4(a)] is reconstructed from theLR shear wave field and Recon2 [Fig. 4(b)] is reconstructedfrom the RL shear wave field. The final map [Fig. 4(c)] has themiddle portion averaged from Recon 1 and Recon 2, part ofsubgroup 1 area from Recon 2 and part of subgroup 4 area fromRecon1. The same reconstruction principles apply to M-CUSE.

E. Ultrasound Safety Measurements

The same acoustic output measurements as in [15] were con-ducted in this study to measure the safety parameters for allCUSE methods. For succinctness, the experiment is not de-scribed here and one can refer to [15] for details. The focaldepth was 42 mm for F-CUSE and 45 mm for M-CUSE. Themechanical index , the spatial peak time average intensity

, the spatial peak pulse average intensity , andtemperature rise of single image acquisition were mea-sured. The pressures were derated by 0.3 dB/cm/MHz for thecalculations of , , and .

III. RESULTS

A. Homogeneous Phantom Experiments

Two homogeneous elasticity phantoms (CIRS Inc., Norfolk,VA, USA) with different shear moduli were used in this study totest the accuracy of U-CUSE, F-CUSE, and M-CUSE for shearwave speed measurements. The nominal Young’s modulus ofphantom 1 is 5.8 kPa [shear wave speed of 1.39 m/s calculatedfrom (1)], phantom 2 is 9.7 kPa (shear wave speed of 1.80 m/s).Both phantoms have ultrasound attenuation of 0.4 dB/cm/MHz,density of 1030 and sound speed of 1539.0 m/s. Theshear wave speeds were measured by magnetic resonance elas-tography (MRE) and 1-D transient elastography (1-D TE) andwere found to be in good agreement in a previous study [22].These are regarded as reference values for this study. The de-tails of MRE and 1-D TE experiments have been described in[22]. Fig. 5 shows the 2-D shear wave speed maps of phantom 1using U-CUSE, F-CUSE, and M-CUSE. No spatial smoothingfilter was applied to these maps. The lateral FOV is the sameas the transducer width (about 39 mm). The data acquisitionfor each map was less than 25 ms. A region-of-interest (ROI)from 5 to 35 mm in lateral direction and from 10 to 35 mm inaxial direction was selected on each shear wave speed map tomeasure the mean and standard deviation values of shear wavespeed, as shown by the red rectangular box in Fig. 5(a). Five ac-quisitions at five different locations in the phantoms were mea-sured by each CUSE method. The five measurements from dif-ferent CUSE methods are compared with values from MRE (30mm 60 mm ROIs from five different planes) and 1-D TE (25mm line ROI along axial direction and five measurements fromdifferent lateral locations). The final results are summarized in


Fig. 4. Schematic plots of 2-D shear wave speed map reconstruction in CUSE. (a) Shear wave speed map reconstructed using LR waves (indicated by blackarrows), (b) shear wave speed map reconstructed using RL waves, (c) final shear wave speed map combined by Recon1 and Recon2.

Fig. 5. 2-D shear wave speed maps of phantom 1 from different CUSE methods: (a) U-CUSE, (b) F-CUSE, and (c) M-CUSE. The red box in (a) indicatesthe measurement ROI for all three methods. The measured shear wave speeds within the ROI from (a) is , (b) is , (c) is

. All speed maps used the same color scale. No spatial smoothing filter was applied to these maps.

Table I. Themeasurements showed good agreements among dif-ferent methods. All CUSE methods produced consistent mea-surements with low variances among different locations.

B. Inclusion Phantom Experiment I

A CIRS breast elastography phantom (Model 059, CIRSInc., Norfolk, VA, USA) was tested in this study. This phantomhas a sound speed of 1540 m/s, ultrasound attenuation of 0.5dB/cm/MHz, density of 1030 and 13 spherical masseswith different sizes and locations. According to CIRS, thestiffness of the inclusions is about three times greater than thestiffness of the background [thus the shear wave speed of theinclusions is about 1.73 times greater than the backgroundaccording to (1)]. The nominal value of the moduli of thisphantom is not available from CIRS. An inclusion situatedabout 25 mm away from the phantom surface was locatedand imaged by U-CUSE, F-CUSE, and M-CUSE. F-CUSEand M-CUSE used beams focused at 25 mm. The initial pushduration was 600 for U-CUSE and F-CUSE, and was 200for M-CUSE as discussed in the Materials and Methods

session. The push duration was then gradually reduced to 450,300, 150, 120, 90, 60, 30, and 15 for U-CUSE and F-CUSEand 150, 100, 50, 40, 30, 20, 10, and 5 for M-CUSE. Foreach push duration, a 2-D shear wave speed map was recon-structed from the three CUSE methods. As shown in Fig. 6,2-D shear wave speed maps from a 600 push (U-CUSEand F-CUSE) and a 200 push (M-CUSE) are provided(no spatial smoothing filter applied). All three CUSE methodswere capable of providing smooth shear wave speed maps withgood contrast between the inclusion and the background. Theedges of the inclusions are sharp, and there are no significantartifacts throughout the speed maps. As shown in Fig. 6(a),ROIs of inclusion and background were selected to quanti-tatively measure the mean and standard deviation of shearwave speeds of the inclusion and background for all CUSEmethods with different push durations. The measured resultsare plotted in Fig. 7. All CUSE methods were able to providestable shear wave speed estimates of the phantom backgroundthroughout different push durations. F-CUSE and M-CUSEwere able to provide robust estimates of shear wave speedsof both the inclusion and background for all push durations,


TABLE ISHEAR WAVE SPEEDS OF PHANTOM 1 AND PHANTOM 2 MEASURED BY MRE, 1-D TE, U-CUSE, F-CUSE, AND M-CUSE

Fig. 6. 2-D shear wave speed maps of the breast inclusion phantom using (a) U-CUSE, (b) F-CUSE, and (c) M-CUSE. All shear wave speed maps usedthe same color scale and no spatial smoothing filter was applied. Push duration was 600 for U-CUSE and F-CUSE, and 200 for M-CUSE. ROIsshown in (a) were selected for all speed maps to measure the mean and standard deviation of shear wave speeds of the background (red dashed box) andthe inclusion (black dashed circle).

Fig. 7. Bar plots of mean and standard deviation measurements of shear wave speeds of the inclusion (a) and background (b) from different CUSE methods.Error bars are plotted from standard deviation. Error bar in (a) for U-CUSE at 15 reaches to about 11 m/s.

while U-CUSE failed below 30 , indicated by an increasedmeasurement of standard deviation in the inclusion.To provide a quantitative estimate of the performance of dif-

ferent CUSE methods, the SNR of shear wave speed measure-ments is calculated for the shear wave speed values given inFig. 7. SNR is given by [23]

(2)

where is the mean value of shear wave speed and is thestandard deviation. To achieve consistent estimate of SNR, the

same mean value of shear wave speed from 600 push ofU-CUSE and F-CUSE and 200 push of M-CUSE was usedto estimate SNR. The SNR measurements versus push durationof different CUSE methods are summarized in Fig. 8. Similarto the observations from Fig. 7, the SNR of inclusion measure-ments from U-CUSE started to decrease at 60 , and the back-ground started to decrease at 30 . F-CUSE and M-CUSE,however, provided consistent SNRs throughout all push dura-tion set-ups. This indicates that F-CUSE and M-CUSE requireshorter push durations to achieve shear wave speed maps withcomparable SNR to U-CUSE.


Fig. 8. Plots of SNR of shear wave speed measurements of the inclusion andbackground from different CUSE methods.

C. Inclusion Phantom Experiment II

Another CIRS inclusion phantom (Elasticity QA Phantom,Model 049, CIRS Inc., Norfolk, VA, USA) was used in thisstudy to compare the penetration of different CUSE push beams.The sound speed of this phantom is and fre-quency dependent attenuation is MHz. Inorder to compare the robustness of different CUSE methods,the most challenging inclusion of this phantom was imaged,which has a nominal Young’s modulus of and whosebottom is about 50 mm away from the phantom surface, asshown by the B-mode image in Fig. 9(d). The nominal Young’smodulus of the phantom background is . Assumingthat the shear modulus is equal to one third of Young’s mod-ulus [24] and using (1), the nominal shear wave speed values forthe background and inclusion are and

, respectively. Both U-CUSE and F-CUSE transmittedpush beams with 600 duration and M-CUSE transmittedpush beams with 200 duration. The focal depth was 40 mmfor both F-CUSE and M-CUSE. The reconstructed shear wavespeed maps are shown in Fig. 9. All shear wave speed mapswere median-filtered with a 3 3 pixel spatial window. BothF-CUSE and M-CUSE provided smooth speed maps with goodcontrast between the inclusion and background. Compared withthe B-mode image of Fig. 9(d), the shape of the inclusion waswell preserved with sharp boundaries and no significant arti-facts. For U-CUSE, however, since the inclusion is very stiffand deep, it was difficult to generate sufficient shear wave mo-tion within the inclusion. Consequently, the shear wave estimateinside the inclusion was noisy and the inclusion is not delineatedwell. ROIs for the background and inclusion were selected toquantitatively measure the mean and standard deviation valuesof the shear wave speed of the inclusion and background, asshown in Fig. 9(a). The measured results are plotted in Fig. 10.Fig. 10 shows that all CUSE-methods provided robust shear

wave speed estimates of the background with excellent agree-ment to the nominal value. Moreover, F-CUSE and M-CUSEproduced accurate estimates of the inclusion shear wave speed

compared with the nominal value, while U-CUSE could notbecause of the noisy image of the inclusion. This indicatesthat both F-CUSE and M-CUSE have better penetration thanU-CUSE and thus should be used in elasticity imaging of deeptissues (e.g., liver and kidney).

D. Inclusion Phantom Experiment III

To evaluate the near field performances of different CUSEmethods, a shallow inclusion from the same CIRS breast elas-tography phantom as in “Inclusion Phantom Experiment I” wasused. The origin of the inclusion is about 15 mm away fromthe surface of the phantom. The same set-ups were used forall CUSE methods except that the focal depth was set to be 15mm for both F-CUSE and M-CUSE. The reconstructed shearwave speed maps are shown in Fig. 11. No spatial smoothingfilters were applied to these maps. All CUSE methods couldprovide good contrast between the inclusion and backgroundwith sharp boundaries. The measured shear wave speedvalues within the ROIs as shown in Fig. 11 of U-CUSE are:

for background, for inclusion;F-CUSE: for background,for inclusion; M-CUSE: for background,

for inclusion. To quantitatively evaluate thepreservation of the shape of the inclusion by different CUSEmethods, the diameter of the inclusion was measured usingboth B-mode images of the phantom and CUSE shear wavespeed maps. Five measurements were made for B-mode andeach CUSE method, as indicated by the five dashed lines inFig. 11(a). The mean and standard deviation values of thesemeasurements are summarized in Table II. Table III shows thevalues of Student’s -tests for the diameter measurements

among different methods. The results showed that the diametermeasurements were statistically different between M-CUSEand the other methods. This is in accordance with the ob-servation from Fig. 11 that both U-CUSE and F-CUSE wellpreserved the inclusion shape while M-CUSE generated a moresquare-shaped inclusion.

E. Ultrasound Safety Measurements

The measured ultrasound safety parameters of U-CUSE,F-CUSE, and M-CUSE as well as the FDA regulatory limitsare summarized in Table IV. All safety parameters are belowFDA regulatory limits [25] for all CUSE methods.

IV. DISCUSSION

Homogeneous phantom experiments showed that all threeCUSE methods were capable of providing smooth 2-D shearwave speed maps (Fig. 5) with accurate shear wave speed es-timates compared with MRE and 1-D TE (Table I). The firstinclusion phantom experiment showed that all CUSE methodswere capable of producing smooth shear wave speed maps withgood contrast between the inclusion and background as well assharp inclusion boundaries. There is no significant artifact in thespeed maps. All CUSE methods were able to provide consistentestimates of shear wave speeds for both inclusion and back-ground with decreased push durations (Fig. 7). Both F-CUSEand M-CUSE could sustain decreased push duration better thanU-CUSE and were able to keep consistent SNR output (Fig. 8).


Fig. 9. Reconstructed shear wave speed maps and the B-mode image for the inclusion in the CIRS phantom. (a) Shear wave speed map reconstructed usingU-CUSE. The red dashed rectangular box is the ROI for background shear wave speed measurement; the black dashed circular ROI is for inclusion shear wavespeed measurement. Same ROIs were used for both F-CUSE and M-CUSE. (b) Shear wave speed map reconstructed using F-CUSE. (c) Shear wave speed mapreconstructed using M-CUSE. (d) B-mode image of the inclusion phantom.

Fig. 10. Bar-plots of the mean and standard deviation values of the measuredshear wave speeds of the inclusion and the background. Error-bars are plottedfrom the standard deviation values.

TABLE IIINCLUSION DIAMETER MEASUREMENTS BY B-MODE AND

CUSE SHEAR WAVE SPEED MAPS

Since no focusing occurs in the unfocused push, significant pushbeam energy is dissipated in the near-field and thus unfocused

TABLE IIIVALUES OF STUDENT’S T-TESTS FOR THE DIAMETER MEASUREMENTS

TABLE IVSUMMARY OF MEASURED ULTRASOUND SAFETY PARAMETERS OF ALL

CUSE METHODS COMPARED WITH FDA REGULATORY LIMIT

pushes usually require push beams with longer durations to pro-duce comparable shear wave amplitudes to focused push. There-fore, when reducing the push duration, U-CUSE suffered fromlack of shear wavemotion first and failed to provide robust shearwave speed maps below 60 push duration.


Fig. 11. 2-D shear wave speed maps of a shallow inclusion from the breast inclusion phantom using (a) U-CUSE, (b) F-CUSE, and (c) M-CUSE. Black dashedcircle and the red dashed rectangle indicate the ROIs that were selected to measure the shear wave speed of the inclusion and the background. The same ROIs wereused for all maps. All shear wave speed maps used the same color scale and no spatial smoothing filter was applied. Five measurements of the inclusion diameterwere made on each map, as indicated by the five dashed lines in (a).

The second inclusion phantom experiment demonstrated thatgreater push beam penetration and deeper imaging depths canbe gained by using F-CUSE and M-CUSE. It is noticeable thatthis inclusion phantom is very challenging for elasticity imagingbecause the background shear modulus is close to 9 kPa, whilethe inclusion is close to 26 kPa and is about 5 cm from thetransducer surface. Both F-CUSE and M-CUSE were able topreserve the shape of the inclusion and accurately estimate theshear wave speed values (Figs. 9 and 10), although the lowerboundary of the inclusion was not well separated from back-ground in Fig. 9(b) and (c) because of weak shear wave motionin this region. U-CUSE was not able to produce shear waveswith sufficient amplitude in the inclusion and was only able torecover the shallow upper part of the inclusion.The final shallow inclusion test showed that U-CUSE and

F-CUSE better preserved the shape of the inclusion thanM-CUSE in shallow imaging. One possible reason for this isthat the push beam out of the focal plane is greatly divergedwhen a large aperture for push is used [22]. The diverged pushbeam generates shear waves with long wavelengths with lowspatial resolutions. This results in smearing of the inclusioncorners and distortion of the inclusion shapes. For F-CUSE,however, since only 32 elements were used for each push beam,the divergence of the push beam is less than in M-CUSE andtherefore less distortion of the inclusion shape occurred. Onemay have to further reduce the number of elements for eachpush beam for F-CUSE to image an even shallower inclusion,i.e., 20 elements for each push beam. The decreased shear waveenergy due to reduced number of elements can be compensatedfor by using a larger number of push beams, i.e., six pushbeams with 20 elements for each push beam. For U-CUSE, thepush beam divergence is minimal because the push beams areunfocused. Since shear waves generated by unfocused pushbeams are more planar and have stronger intensity in near field,U-CUSE is ideal for shallow tissue imaging. One can alsochange the number of elements for each push beam as well asthe total number of push beams for U-CUSE. Theoretically,a smaller aperture for the unfocused push will give narrowershear waveforms with shorter shear wavelengths and thushigher spatial resolution up to a point. Although shear waveamplitude will be lower with the smaller aperture of unfocusedpush, less attenuation occurs in near field and therefore it isfeasible to decrease the aperture size. Meanwhile, reduced

aperture size allows a larger number of push beams whichcan compensate for the loss of shear wave amplitude. Futurestudy is needed to systematically optimize the comb-push beamset-ups for different tissue applications.The focused comb-push and marching comb-push described

here extend the imaging range of depth and flexibility of theCUSE method. The U-CUSE method has no control of ultra-sound intensity and shear wave amplitude along the depth direc-tion. Typically with unfocused comb-push, shear wave ampli-tude decreases with depth, with maximum amplitude occurringclose to the transducer surface. F-CUSE and M-CUSE providethe flexibility to control the distribution of shear wave energyalong the depth direction. This change, although conceptuallynot complicated, has important practical value. As shown in thesecond inclusion phantom experiment, they allow effective de-livery of comb-push energy to a depth larger thanwhat is achiev-able with U-CUSE. This beneficial feature will be important forapplications such as liver imaging.

A. Comparison With SSI and ARFI Shear Wave Imaging

The marching comb-push in M-CUSE divides the transducerelements into several overlapping sub-apertures to transmitmultiple focused push beams sequentially for shear wavegeneration. While this is similar to the push beams used inSSI [5]–[7], [26]–[28] which is implemented in the Aixplorerultrasound system (SuperSonic Imagine, Aix-en-Provence,France) and ARFI shear wave imaging [4], [29], there areseveral significant differences. 1) M-CUSE does not requireshear wave detection between push beams, unlike ARFI shearwave imaging. The detection starts after the transmission ofthe last push beam and it is the comb-shaped shear wave fieldthat marching comb-push is aiming to produce, not individualshear waves from each push beam. 2) M-CUSE distributesconsecutive focused push beams in the lateral direction, unlikeSSI in which consecutive focused push beams are distributedalong the axial direction. 3) M-CUSE does not require a prioriknowledge of the medium shear wave speed to chase down orgo faster than the shear wave front to enhance the shear waveas introduced in [7]. In fact the marching speed of consecutivefocused push beams was set to be as fast as possible in orderto detect the comb-shaped shear wave field as soon as possible.4) M-CUSE can arbitrarily alter the push beam sequences, e.g.,the order of the subgroup activations as in Fig. 1(c), because the


order of the push beam transmissions is irrelevant to producinga comb-shaped shear wave field with multiple shear waves.This is different from SSI in which the consecutive focusedpush beams must be excited from top to bottom or from bottomto top, to construct a single shear wave with specific Machnumbers.Note that all three CUSE methods produce a complex shear

wave field with multiple shear waves to reconstruct a full FOVshear elasticity map with only one push-detection acquisition,while SSI and ARFI shear wave imaging require multiple push-detection acquisitions to reconstruct a full FOV map because of1) absence of shear waves at the push beam region and 2) signif-icant shear wave attenuation in areas that are far from the pushbeam region. Therefore, for full FOV 2-D shear wave imaging,the frame rate of the CUSE methods will be higher than SSI andARFI shear wave imaging. This higher frame rate will be impor-tant for imaging a beating heart or contracting skeletal muscle.SSI and ARFI shear wave imaging produce and analyze a singleshear wave front at a time, while the CUSE methods produceand analyzemultiple shear wave fronts simultaneously thanks tothe directional filter which can differentiate shear waves propa-gating at different directions. The cross-correlation of multipleshear wave fronts (as shown below), also gives a narrower cor-relation peak for the CUSE methods than SSI and ARFI shearwave imaging, which suggests a more robust shear wave speedestimate from the CUSE methods.To illustrate the differences described above, a direct com-

parison of reconstructed shear wave speed maps from eachimaging sequence, imaging speed, and shear wave signalbetween M-CUSE, SSI, and ARFI shear wave imaging (onthe Verasonics system used for this study) was conducted onphantom 1. For M-CUSE, the same sequence introduced inSection II was used to image the phantom. The single dataacquisition was about 25 ms. The reconstructed shear wavespeed map is shown in Fig. 12(a). Fig. 12(b) also shows atypical pair of shear wave particle velocity waveforms afterdirectional filtering that were used to recover local shear wavespeed. Note that the four shear waves from the four push beamswere used simultaneously for the cross-correlation calculation.The frame rate of the 2-D shear wave speed map reconstructedby M-CUSE can reach approximately ms inreal-time imaging in principle (ignoring computational cost,thermal safety and transducer heating).For SSI, four consecutive focused push beams with focal

depths of 6, 14, 22, and 30 mm were transmitted. The F-numberof each push beam was fixed to be 1. The push duration of eachpush beam was 150 . The time interval between consecutivepush beams was 15 . Data acquisition started after the trans-mission of the last push beam and lasted for about 25 ms. Thefirst set of SSI push (SSI Sequence 1) was positioned on theleft side of the FOV, followed by a 25 ms data acquisition, andthe reconstructed shear wave speed map is shown in Fig. 12(c).The second set of SSI push (SSI Sequence 2) was positionedon the right side of the FOV, followed by a 25 ms data acquisi-tion, and the reconstructed shear wave speed map is shown inFig. 12(d). Note that the push beam area in Fig. 12(c) and (d)could not be properly recovered within each SSI sequence dueto absence of shear waves in the push beam region. A final

full FOV shear wave speed map [Fig. 12(e)] was then recon-structed by concatenating Fig. 12(c) and (d). The concatenationwas done by averaging the areas of Sequences 1 and 2 withoutthe push beams, filling the push beam area of Sequence 1 withSequence 2 data, and filling the push beam area of Sequence 2with Sequence 1 data. Since two SSI sequences were neededto reconstruct the full FOV shear wave speed map, the totaldata acquisition time needed was at least ms.Consequently, the real-time imaging frame rate of SSI wouldbe 20 Hz, half that of M-CUSE. According to [14], three SSIpush sequences are typically needed to reconstruct a full FOVshear wave speed map—bringing the total data acquisition timeto ms, further reducing the frame rate.Also, as shown in Fig. 12(f), SSI processes a single shear wavefront from each SSI sequence for cross-correlation, as opposedto M-CUSE [Fig. 12(b)].For ARFI shear wave imaging, a single focused push beam

with aperture size of 64 elements, focal depth of 25 mm (sameas the individual focused push in M-CUSE) and push durationof 600 was used for each imaging sequence. The push du-ration was longer than in M-CUSE and SSI because only onepush beam was transmitted. Data acquisition started immedi-ately after the transmission of the push and lasted for 25 ms.The first ARFI push (ARFI Sequence 1) was transmitted on theleft side of the FOV, followed by a 25 ms data acquisition, andthe reconstructed shear wave speed map is shown in Fig. 12(g);the second ARFI push (ARFI Sequence 2) was transmitted onthe right side of the FOV, followed by another 25 ms data ac-quisition, and the reconstructed shear wave speed map is shownin Fig. 12(h). Similar to SSI, ARFI shear wave imaging cannotreconstruct a full FOV shear wave speed map with only onepush-detection acquisition. A similar concatenation method asused for SSI above was used to combine Fig. 12(g) and (h) intoa final full FOV shear wave speed map [Fig. 12(i)], bringingthe total data acquisition time to 50 ms, and a single shear wavefront [Fig. 12(j)] was used for cross-correlation, as opposed toM-CUSE [Fig. 12(b)].A major difference between M-CUSE and SSI and ARFI

shear wave imaging is that multiple shear waves are pro-duced and processed simultaneously by M-CUSE, as shownin Fig. 12(b), (f), and (j). This allows more robust cross-cor-relation calculations of shear wave speed. Fig. 13(a) showsa direct comparison of the normalized correlation coefficientplots from the shear wave signals of M-CUSE [Fig. 12(b)],SSI [Fig. 12(f)] and ARFI shear wave imaging [Fig. 12(j)].M-CUSE has a narrower correlation peak than SSI and ARFI,which suggests a more robust shear wave speed estimate fromM-CUSE. To show that this relationship holds at differentspatial locations, the cross-correlation function width of allimaging pixels within the black ROIs in Fig. 12 were calculatedfor M-CUSE, SSI, and ARFI shear wave imaging. The width isgiven by measuring the full-width at half-maximum (FWHM)of the cross-correlation function. Fig. 13(b) shows the meanand standard deviation values of the measured cross-corre-lation function widths among different modalities. Note thatsimilar to the result in Fig. 13(a), M-CUSE has a narrowercross-correlation function width than SSI and ARFI shear waveimaging. One possible reason is that M-CUSE cross-correlates


Fig. 12. Comparisons between M-CUSE and SSI/ARFI shear wave imaging. (a) 2-D shear wave speed map of phantom 1 reconstructed by M-CUSE. The blackdashed box indicates the ROI used for shear wave speed measurements. Same ROIs were used for SSI and ARFI shear wave imaging. Shear wave speed withinthe ROI is mean std . A and B indicate the spatial points from which the shear wave particle velocity signals were plotted, as shown in (b).(c) and (d) 2-D shear wave speed maps of phantom 1 reconstructed from the first (c) and the second (d) SSI sequence. Red dashed box indicates the push beam areawhere the local shear wave speed could not be recovered properly. (e) Final full FOV shear wave speed map reconstructed by combining maps (c) and (d). Shearwave speed within the ROI is . (f) The shear wave particle velocity signals from points A and B that were used in cross-correlationcalculation for local shear wave speed recovery in SSI. (g) and (h) 2-D shear wave speed maps of phantom 1 reconstructed from the first (g) and the second (h) ARFIshear wave imaging sequence. The red dashed box indicates the push beam area where the local shear wave speed could not be recovered properly. (i) Final fullFOV shear wave speed map reconstructed by combining maps (g) and (h). Shear wave speed within the ROI is mean std (j) Shear waveparticle velocity signals from points A and B that were used in cross-correlation calculation for local shear wave speed recovery in ARFI shear wave imaging.

Fig. 13. (a) Cross-correlation coefficient plots of the shear wave signals pro-duced by M-CUSE [Fig. 12(b)], SSI [Fig. 12(f)], and ARFI shear wave imaging[Fig. 12(j)]. (b) Bar-plots of the mean and standard deviation values of the mea-sured cross-correlation function width of M-CUSE, SSI, and ARFI shear waveimaging within the ROI in Fig. 12.

multiple shear wave fronts, which should narrow the cross-cor-relation width. Another possible reason is that as the shearwave propagates away from the push beam, the high frequency

component attenuates faster than the low frequency component,which widens the shear wave motion signal and consequentlybroadens the cross-correlation function. Because M-CUSE dis-tributes multiple focused push beams along the lateral directionso that each imaging pixel always has a shear wave source closeby, the dominant shear wave front has a narrower time profileand consequently narrows the final cross-correlation function.This brief comparison study is not comprehensive, and fu-

ture study is needed to fully investigate the differences amongthese modalities. Table V summarizes the differences betweenM-CUSE, SSI and ARFI shear wave imaging (F-CUSE is in-cluded as well, since the conclusions for M-CUSE discussedabove also apply to F-CUSE).

B. Limitations

One potential drawback of M-CUSE is that a shorter push du-ration (200 instead of 600 ) has to be used, if maximum


TABLE VSUMMARY OF THE DIFFERENCES BETWEEN F-CUSE/M-CUSE AND SSI AND ARFI SHEAR WAVE IMAGING

transducer heating is to be controlled at the same level for allCUSE configurations, because the center transducer elementsare excited three times during the M-CUSE sequence. Shorterpush duration suggests a weaker shear wave signal from eachpush beam, which will reduce the SNR of the shear wave signaland the robustness of the shear wave speed estimate. One can in-crease the push duration by reducing the number of push beams,i.e., having fewer overlapping beams. This will increase the am-plitude of the shear wave signal from each push beam but reducethe total number of “teeth” in the comb-push beams. A futurestudy is needed to balance the trade-off between the single shearwave amplitude and the total number of push beams.One potential drawback of F-CUSE is that the total amount of

acoustic energy is spread over all pushes that are transmitted si-multaneously, which may cause the energy of individual pushbeam in F-CUSE to be lower than the push in conventionalshear wave imaging methods [1], [4], [5]. However, the advan-tage of the simultaneous laterally distributed push beams is thesignificant increase in the frame rate of shear wave elasticityimaging. There is a trade-off between frame rate and the qualityof the shear wave speed maps (i.e., the SNR of the shear wavewhich is related to the acoustic energy of the push beam). Con-ventional shear wave imaging methods like SSI have higheracoustic energy for the push beam but lower frame rate, whileCUSE has higher frame rate but lower acoustic energy for eachpush beam. One can decrease the number of simultaneous fo-cused push beams to increase the aperture size of each pushbeam in F-CUSE to increase the acoustic energy of individualpush beams.Because CUSE is not creating a constructively interfering

diffraction pattern as in SSI, the individual shear wave SNRfrom CUSEmay be lower than SSI. However, CUSE effectivelyincreases the frame rate and overall SNR of shear wave imagingby distributing multiple push beam sources along the lateral di-rection to create multiple shear waves so that each imaging pixelwill always have one or more push beam sources close by.As shown in Figs. 6 and 7, the shear wave speed estimate for

the inclusion is high for U-CUSE. This may be caused by thedirect push beam on the inclusion: the inclusion was positioneddirectly under the center push beam. Different modes and com-plex interferences of shear waves with higher velocities may beexcited within the inclusion with direct pushing, which may re-sult in the overestimation. Also, although the directional filterused in this study can remove the artifacts caused by horizon-tally reflected shear waves as shown in [19], shear wave fronts

can still be disturbed by the boundaries of the inclusions andthus propagate in various directions. This may result in artifactsbecause a lateral propagation direction of shear waves was as-sumed in this study when calculating local shear wave speeds.Future work including a finite element modeling (FEM) studyis needed to address these issues. Also, all experiments in thisstudy were conducted in phantoms. In vitro and in vivo tests onvarious soft tissues are needed in the future to optimize CUSEfor different types of applications.For safety measurements, the temperature rise and

were measured from only one push-detection acquisition. Onemay expect an increased and when repeating CUSEmeasurements in real time. The temperature rise measure-ment of M-CUSE did not consider the fact that when transmit-ting multiple overlapping push beams, one can get peak heatingin the near field of overlapping and adjacent acoustic radiationforce excitations, especially in highly-attenuating media [30].Since the measurement of M-CUSE from a single pushwas only 0.006 and only four overlapping focused pushbeams spaced out over a relatively large lateral range were ex-cited per image acquisition, one may still expect a low tem-perature rise even if the peak heating location would occur inthe near field. Future study is needed to investigate thein continuous M-CUSE imaging where maximum heating ismore likely to occur in near field. Finally, the safety measure-ments in this paper did not account for the effect of transducersurface heating. For the current set-up where the accumulated“ON” time for each element per acquisition was kept below afixed duration ( was equal to 600 in this paper),F-CUSE should produce the highest amount of transducer sur-face heating in continuous imaging because all the transducerelements are excited for for every acquisition. M-CUSEhas lower transducer surface heating than F-CUSE because onlysome of the overlapping elements are excited for whilethe other elements are excited for less than . U-CUSEshould also have lower transducer surface heating than F-CUSEbecause only part of the elements are excited per unfocusedcomb-push. Moreover, because only 60 elements are excitedper comb-push, one can alternate different combinations of un-focused push beams that are distributed at different lateral lo-cations, which should reduce the transducer surface heating ofU-CUSE. Future study will be needed to fully understand thetransducer surface heating during continuous imaging.Based on this study, U-CUSE and F-CUSE can be optimized

for shallow tissue elasticity imaging, such as breast, skeletal


muscle like biceps, and thyroid; F-CUSE and M-CUSE can beoptimized for deeper tissue elasticity imaging, such as liver,kidney, spleen and heart. One can combine different CUSEmethods to obtain strong shear wave signals in both near andfar fields. Moreover, CUSE methods can also be implementedon other types of ultrasound transducers like curved and phasedarrays, which have lower ultrasound center frequencies forbetter penetration.

V. CONCLUSION

This paper introduces two new versions of comb-push ul-trasound shear elastography (CUSE) using focused ultrasoundpush beams: focused CUSE (F-CUSE) and marching CUSE(M-CUSE). Both new techniques, like the original unfocusedCUSE (U-CUSE), can acquire a full FOV 2-D shear wave speedmap with rapid data acquisition. Homogeneous and inclusionphantom results indicate that all CUSE methods were ableto provide smooth 2-D shear wave speed maps with accurateshear wave speed measurements and good contrast between theinclusion and background. This paper shows that both F-CUSEand M-CUSE achieved better shear wave penetration thanU-CUSE and require shorter push durations to achieve shearwave speed maps with comparable SNR to U-CUSE, whileU-CUSE and F-CUSE preserved the shape of the inclusionbetter than M-CUSE in shallow imaging. These promisingresults indicate that CUSE with various push beams can beused in a wide range of depths for soft tissue elasticity imaging.Future work includes application-oriented optimizations ofCUSE based on in vitro and in vivo tissue experiments as wellas implementation of CUSE on different types of transducerssuch as curved arrays and phased arrays.

ACKNOWLEDGMENT

The authors would like to thank R. Kinnick for his experi-mental assistance.

REFERENCES[1] A. P. Sarvazyan, O. V. Rudenko, S. D. Swanson, J. B. Fowlkes, and

S. Y. Emelianov, “Shear wave elasticity imaging: A new ultrasonictechnology of medical diagnostics,” Ultrasound Med. Biol., vol. 24,pp. 1419–1435, Nov. 1998.

[2] Y. Yamakoshi, J. Sato, and T. Sato, “Ultrasonic imaging of internalvibration of soft tissue under forced vibration,” IEEE Trans. Ultrason.Ferroelectr., Freq. Control, vol. 37, no. 2, pp. 45–53, Mar. 1990.

[3] K. R. Nightingale, M. L. Palmeri, R. W. Nightingale, and G. E. Trahey,“On the feasibility of remote palpation using acoustic radiation force,”J. Acoust. Soc. Am., vol. 110, pp. 625–634, Jul. 2001.

[4] K. Nightingale, S. McAleavey, and G. Trahey, “Shear-wave generationusing acoustic radiation force: In vivo and ex vivo results,” UltrasoundMed. Biol., vol. 29, pp. 1715–1723, Dec. 2003.

[5] J. Bercoff, M. Tanter, and M. Fink, “Supersonic shear imaging: A newtechnique for soft tissue elasticity mapping,” IEEE Trans. Ultrason.Ferroelectr., Freq. Control, vol. 51, no. 4, pp. 396–409, Apr. 2004.

[6] J. Bercoff, M. Tanter, S. Chaffai, and M. Fink, “Ultrafast imagingof beamformed shear waves induced by the acoustic radiation force.Application to transient elastography,” in Proc. IEEE Int. UltrasonicsSymp., Munich, Germany, 2002, pp. 1899–1902.

[7] J. Bercoff, “L’imagerie echographique ultrarapide et son appli-cation—A l’etude de la viscoelasticite du corps humain,” Ph.D.dissertation, Univ. Paris 7—Denis Diderot, Paris, France, 2004.

[8] S. A. McAleavey, M.Menon, and J. Orszulak, “Shear-modulus estima-tion by application of spatially-modulated impulsive acoustic radiationforce,” Ultrason. Imag., vol. 29, pp. 87–104, Apr. 2007.

[9] S. Chen, M. W. Urban, C. Pislaru, R. Kinnick, Y. Zheng, A. Yao, andJ. F. Greenleaf, “Shearwave dispersion ultrasound vibrometry (SDUV)for measuring tissue elasticity and viscosity,” IEEE Trans. Ultrason.Ferroelectr., Freq. Control, vol. 56, no. 1, pp. 55–62, Jan. 2009.

[10] S. Chen, M. Fatemi, and J. F. Greenleaf, “Quantifying elasticity andviscosity from measurement of shear wave speed dispersion,” J.Acoust. Soc. Am., vol. 115, pp. 2781–2785, Jun. 2004.

[11] Z. Hah, C. Hazard, B. Mills, C. Barry, D. Rubens, and K. Parker, “In-tegration of crawling waves in an ultrasound imaging system. Part 2:Signal processing and applications,” Ultrasound Med. Biol., vol. 38,pp. 312–323, Feb. 2012.

[12] C. Hazard, Z. Hah, D. Rubens, and K. Parker, “Integration of crawlingwaves in an ultrasound imaging system. Part 1: System and design con-siderations,” Ultrasound Med. Biol., vol. 38, pp. 296–311, Feb. 2012.

[13] H. Zhao, P. Song, M. W. Urban, J. F. Greenleaf, and S. Chen, “Shearwave speed measurement using an unfocused ultrasound beam,” Ul-trasound Med. Biol., vol. 38, pp. 1646–1655, Sep. 2012.

[14] M. Tanter, J. Bercoff, A. Athanasiou, T. Deffieux, J. L. Gennisson, G.Montaldo, M. Muller, A. Tardivon, and M. Fink, “Quantitative assess-ment of breast lesion viscoelasticity: Initial clinical results using super-sonic shear imaging,” Ultrasound Med. Biol., vol. 34, pp. 1373–1386,Sep. 2008.

[15] P. Song, H. Zhao, A. Manduca, M. W. Urban, J. F. Greenleaf, and S.Chen, “Comb-Push ultrasound shear elastography (CUSE): A novelmethod for two-dimensional shear elasticity imaging of soft tissues,”IEEE Trans. Med. Imag., vol. 31, no. 9, pp. 1821–1832, Sep. 2012.

[16] P. Song, M. W. Urban, A. Manduca, H. Zhao, J. F. Greenleaf, andS. Chen, “Comb-push ultrasound shear elastography (CUSE): A noveland fast technique for shear elasticity imaging,” presented at the IEEEInt. Ultrason. Symp., Dresden, Germany, 2012.

[17] M. Tanter, J. Bercoff, L. Sandrin, and M. Fink, “Ultrafast compoundimaging for 2-D motion vector estimation: Application to transientelastography,” IEEE Trans. Ultrason. Ferroelectr., Freq. Control, vol.49, no. 10, pp. 1363–1374, Oct. 2002.

[18] C. Kasai, K. Namekawa, A. Koyano, and R. Omoto, “Real-time two-dimensional blood flow imaging using an autocorrelation technique,”IEEE Trans. Sonics Ultrason., vol. 32, no. 3, pp. 7–7, May 1985.

[19] T. Deffieux, J. L. Gennisson, J. Bercoff, and M. Tanter, “On the effectsof reflected waves in transient shear wave elastography,” IEEE Trans.Ultrason. Ferroelectr., Freq. Control, vol. 58, no. 10, pp. 2032–2035,Oct. 2011.

[20] A. Manduca, D. S. Lake, S. A. Kruse, and R. L. Ehman, “Spatio-tem-poral directional filtering for improved inversion of MR elastographyimages,” Med. Image. Anal., vol. 7, pp. 465–473, Dec. 2003.

[21] F. J. Harris, “Use of windows for harmonic-analysis with discreteFourier-transform,” Proc. IEEE, vol. 66, no. 1, pp. 51–83, Jan. 1978.

[22] H. Zhao, P. Song, M.W. Urban, R. R. Kinnick, M. Yin, J. F. Greenleaf,and S. Chen, “Bias observed in time-of-flight shear wave speed mea-surements using radiation force of a focused ultrasound beam,” Ultra-sound Med. Biol., vol. 37, pp. 1884–1892, Nov. 2011.

[23] B. J. Oosterveld, J. M. Thijssen, and W. A. Verhoef, “Texture ofB-mode echograms: 3-D simulations and experiments of the effects ofdiffraction and scatterer density,”Ultrason. Imag., vol. 7, pp. 142–160,Apr. 1985.

[24] T. E. Oliphant, A. Manduca, R. L. Ehman, and J. F. Greenleaf,“Complex-valued stiffness reconstruction for magnetic resonanceelastography by algebraic inversion of the differential equation,”Magn. Reson. Med., vol. 45, pp. 299–310, Feb. 2001.

[25] B. A. Herman and G. R. Harris, “Models and regulatory considerationsfor transient temperature rise during diagnostic ultrasound pulses,”Ul-trasound Med. Biol., vol. 28, pp. 1217–1224, Sep. 2002.

[26] M. Fink and M. Tanter, “Procede et dispositif d’imagerie utilisant desondes de cisaillement,” France Patent 2844058, 2002.

[27] M. Muller, J. L. Gennisson, T. Deffieux, M. Tanter, and M. Fink,“Quantitative viscoelasticity mapping of human liver using supersonicshear imaging: Preliminary in vivo feasibility study,”Ultrasound Med.Biol., vol. 35, pp. 219–229, Feb. 2009.

[28] E. Bavu, J. L. Gennisson, M. Couade, J. Bercoff, V.Mallet, M. Fink, A.Badel, A. Vallet-Pichard, B. Nalpas, M. Tanter, and S. Pol, “Noninva-sive in vivo liver fibrosis evaluation using supersonic shear imaging:A clinical study on 113 hepatitis C virus patients,” Ultrasound Med.Biol., vol. 37, pp. 1361–1373, Sep. 2011.

[29] N. C. Rouze, M. H. Wang, M. L. Palmeri, and K. R. Nightingale, “Pa-rameters affecting the resolution and accuracy of 2-D quantitative shearwave images,” IEEE Trans. Ultrason. Ferroelectr., Freq. Control, vol.59, no. 8, pp. 1729–1740, Aug. 2012.

[30] M. L. Palmeri and K. R. Nightingale, “On the thermal effects associ-ated with radiation force imaging of soft tissue,” IEEE Trans. Ultrason.Ferroelectr., Freq. Contr., vol. 51, no. 5, pp. 551–565, May 2004.

Documents

IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 32, NO. 8 ......comb-push beams and track shear wave motions with a linear array transducer L7-4 (Philips Healthcare, Andover, MA, USA)