WO2012035472A1

WO2012035472A1 - Quantification of tissue strain in ultrasonic elastography images

Info

Publication number: WO2012035472A1
Application number: PCT/IB2011/053938
Authority: WO
Inventors: Alwyn Patrick D'sa; Junzheng Man
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2010-09-16
Filing date: 2011-09-09
Publication date: 2012-03-22

Abstract

An ultrasonic diagnostic imaging system acquires a sequence of elastography images (elastograms) with a probe as the probe is used to vary the pressure and compress a region of interest of an anatomic mass within the body. A region of interest is identified in one of the images of the acquired sequence of elastograms. The mean change in strain from one image frame to the next within the region of interest is computed over the interval that the region of interest is variably compressed, e.g., from a starting minimum level of pressure to a final maximum level. This intermediate result is used to produce the strain rate. Total or cumulative strain is then calculated over the compression interval to produce a curve or profile of the total strain in the ROI. The final level of the total strain curve or profile is a measure of the stiffness of the tissue of the ROI.

Description

QUAN IFICA ION OF TISSUE STRAIN

IN ULTRASONIC ELASTOGRAPHY IMAGES

This invention relates to medical diagnostic ultrasound systems and, in particular, to ultrasound systems which assess the stiffness of tissue regions in the body by elastography .

Elastography is the assessment of the elastic properties of tissue in the body. It has been found that the stiffness of tissue in the body can give an indication of whether the tissue may be malignant or benign. The female breast, for instance, can contain a variety of different lumps, cysts, and other growths, some of which may be malignant and some of which may be benign. To spare the patient from needless biopsies, and perform them when needed, ultrasound is frequently used to assess tissue characteristics to determine whether to biopsy suspect tissue. Elastography can be performed to determine whether the breast contains softer or harder (stiffer) regions. Since stiffer tissue correlates more greatly with malignant masses, the identification of regions of stiffer tissue can indicate a need to make a definitive diagnosis by biopsy.

A problem posed by elastography is the need to accurately and reproducibly measure quantifiable properties of tissue non-invasively within the body. This means that the properties of the target tissue cannot be measured directly at the site of the tissue, but only through measurements made at the surface of the body through intervening tissues.

Accordingly, it is desirable to simplify the problem and make certain approximations and assumptions that will lead to valid data and analyses. One set of assumptions that is frequently made is that the tissue being examined is homogeneous and isotropic. These assumptions enable certain property of

materials equations to be applied to the problem, Poisson's ratio and Young's modulus. Poisson's ratio is the ratio, when a sample is stretched or

compressed in a given direction, of the expansion or contraction (strain) normal to the stretching or compressing force, to the expansion or contraction axially in the direction of the force. A related measure is Young's modulus, which is a measure of stiffness, and is defined as the ratio of the

uniaxial stress (pressure) applied to a sample over the resulting uniaxial strain (deformation) .

However, the stress component at target tissue is generally unknown and difficult to measure non- invasively .

Since each of these material properties can be expressed as a function of deformation (strain) , a number of researchers have concentrated on assessing the strain or deformation exhibited by tissue at different applied pressures. While strain has been shown to be a useful parameter, its shortcoming is that it varies with the applied pressure. It is thus technique-dependent, with different results obtained by researchers who apply different pressures to the tissue or use different techniques for applying pressure. A consequence is that strain measurements from patient to patient or from exam to exam with the same patient by different clinicians may not be directly comparable. A patient subjected to one type of pressure will exhibit different strain

(deformation) values as compared to another patient subjected to a different form or amount of pressure. To overcome this problem of pressure repeatability, efforts have been made in the past to use compressors which would apply controlled pressure to the tissue so that it is compressed (displaced) in known

increments. See, for example, US Pats. 5,293,870 (Ophir et al . ) , 5, 178, 147 (Ophir et al . ) , and

5,107,837 (Ophir et al . ) Such compression devices cause the elastography procedure to become cumbersome in a routine clinical setting and require mechanisms beyond just the ultrasound system. This has led to compression being performed manually with the

ultrasound probe to compress the tissue. However, using the probe alone as a compressor means that the compression and displacement cannot be precisely quantified .

To compensate for the pressure variation between patients and clinicians, strain of a reference point can be used to normalize strain of a target region. In addition, a strain ratio between normal tissue and tumor can indicate their relative stiffness, assuming similar pressure between two regions within a

patient. See, for example, US patent application serial number 61/239,984 (Bae et al . ) filed September 4, 2009.

While the strain ratio approach of Bae et al . leads to the ability to assess regions of varying stiffness in ultrasound images, there remains a need to be able to quantify and compare the stiffness of different tissues in the body. In accordance with the principles of the present invention, tissue stiffness is quantified by acquiring and analyzing a sequence of strain (elastography) images as the compressive force applied to the body is varied. The sequence of strain images is used to compute a curve or profile of the strain rate variation over a time interval of compressive force variance. The strain rate data is then used to compute a curve or profile of total (cumulative) strain over the time interval of compressive force variance. As will be seen from the examples described below, the maximum or peak values of total (cumulative) strain curves are repeatable even though the technique used to apply the compressive force may vary from one procedure to another .

In the drawings:

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

invention .

FIGURE 2 illustrates the steps of a method for producing a quantified measure of manually induced strain .

FIGURE 3 illustrates an ultrasound image of a region of interest (ROI) of a target tissue mass in a tissue-mimicking phantom.

FIGURE 4 illustrates one of a plurality of elastography images of the ROI obtained during image acquisition as a variable pressure is applied to compress the target with an ultrasound probe.

FIGURE 5 illustrates instantaneous mean strain and strain rate versus time curves calculated from the elastogram data of an identified ROI. In FIGURES 5-8 the B mode ultrasound image of the ROI is shown on the left of the image display and the elastogram is displayed on the right.

FIGURE 6 illustrates total (cumulative) strain versus time curves for hard (stiff) tissue and softer background tissue.

FIGURE 7 illustrates strain rate versus time curves produced from image data acquired during rapid tissue compression. FIGURE 8 illustrates total (cumulative) strain versus time curves calculated using the rapidly acquired compression data of FIGURE 7.

Referring first to FIGURE 1, an ultrasound system constructed in accordance with the principles of the present invention is shown in block diagram form. An ultrasound probe 10 has an array transducer 12 for transmitting ultrasound waves to and receiving echoes from a region of the body. The array

transducer can be a one-dimensional array of

transducer elements or a two-dimensional array of transducer elements for scanning a two dimensional image field or a three dimensional image field in the body. The elements of the array transducer are driven by a transmit beamformer 16 which controls the steering, focusing and penetration of transmit beams from the array. A receive beamformer 18 receives echoes from the transducer elements and combines them to form coherent echo signals from points in the image field. The transmit and receive beamformers are coupled to the transducer array elements by transmit/receive switches 14 which protect sensitive receive circuitry during transmission. A beamformer controller 20 synchronizes and controls the operation of the beamformers.

The received echo signals are demodulated into quadrature (I and Q) samples by a quadrature bandpass (QBP) filter 22. The QBP filter can also provide band limiting and bandpass filtering of the received signals. The received signals may then undergo further signal processing such as harmonic separation and frequency or spatial compounding by a signal processor 24. The processed echo signals are applied to a detector 25 which performs amplitude detection of the echo signals by the equation ( I² + Q² )¹¹² for a B mode processor 26, and to a Doppler processor 28 for Doppler phase detection of tissue motion at points in the image field. The outputs of the B mode processor 26 and the Doppler processor 28 are coupled to a frame memory or buffer 30 for storage. The frame memory stores consecutive scans of the image field on a spatial basis for the calculation of strain by a strain estimator 32 from the frame-to-frame

displacement of particles in the image field; strain is calculated as a spatial derivative of

displacement. Strain may be calculated from

radiofrequency (r.f.) or baseband I and Q data, and may also be calculated from amplitude-detected (B mode) or tissue Doppler data. Any form of strain calculation such as strain, the ratio of lateral to axial strain, and strain velocity estimation may be employed. For instance, the echoes received at a common point in consecutive frames may be correlated to estimate displacement at the point. If no motion is present at the point, the echoes from consecutive frames will be the same. If motion is present, the echoes will be different and the motion vector indicates the displacement. US Pat. 6,558,324 (Von Behren et al . ) describes both amplitude and phase sensitive techniques for estimating strain and employs speckle tracking for strain estimation through block matching and correlation. US Pat.

5,524,636 (Sarvazyan et al . ) also uses speckle tracking to perform elastography and US Pat.

6,527,717 (Jackson et al . ) determines tissue movement by correlating speckle. US Pat. 5,474,070 (Ophir et al . ) estimates tissue compression from time-shifted differences among segments of A-line pairs. US Pat. 6,099,471 (Torp et al . ) describes the estimation of strain velocity calculated as a gradient of tissue velocity. US Pat. 5,800,356 (Criton et al . )

describes the use of the Doppler vector to select points for strain estimation in the direction of the motion produced by the applied force. Preferably a phase-sensitive technique is used since, as

recognized by Von Behren et al . , r.f. data will typically yield the most accurate estimates of strain. Frame-to-frame comparisons of the Doppler phase shift at a point is a technique that is

sensitive to small displacements of tissue.

The tissue displacement is caused by varying the pressure applied to the body by the probe. Any other type of compression source can alternatively be used, including mechanical vibration or acoustic radiation force impulse. However compression applied by varying the force of the probe against the body is preferred as it requires no additional devices or special pulse transmissions. Another reason for the preference of strain estimation with phase-sensitive techniques is that the slight motion produced by small, virtually imperceptible motion occurring while holding a probe against the body can be sensed and used to estimate strain by the strain estimator 32.

The strain estimator 32 produces an estimated strain value at each point in the image field which are spatially arranged to form a strain image or elastogram. A sequence of such elastograms are produced from successive images acquired during probe pressure variation and stored as time-sequential strain images of the image field at 34. In

accordance with the principles of the present

invention, the strain image data is coupled to a strain rate curve calculator 36. In a preferred implementation the calculator 36 produces a strain rate curve in a two-step process, first producing an instantaneous mean strain curve for an ROI, then scaling this data by the system pulse repetition frequency (PRF) to produce a strain rate versus time curve. The strain rate data is applied to a total (cumulative) strain curve calculator 38, which produces a total strain curve that yields a

quantified measure of the stiffness of the tissue of the ROI . The B mode and strain image are coupled to an image processor 42, as well as the calculated curves, for production of images on a display 50.

FIGURE 2 illustrates a method for producing a quantified measure of tissue stiffness in accordance with the present invention using the ultrasound system of FIGURE 1. In step 60 the ultrasound probe 10 is placed against the body over an ROI and pressed against the body with a varying force to cause deformation of the tissue of the underlying ROI. The pressure of the probe can be gradually increased so that the tissue within the ROI is fully deformed with the application of maximum reasonable force, or a maximum force can be applied, then decreased

gradually. As the applied force is varied, the ultrasound system is triggered to store a sequence of elastography images in the frame buffer 30.

Triggering can be done by the equivalent techniques of triggering the start of storage in the buffer, or running the images through the buffer continuously as a first-in, first-out buffer, then triggering the buffer to stop running and retain the immediately preceding sequence of images in the buffer. In a constructed embodiment the frame buffer can store up to twenty seconds of continuous image frames as the tissue is variably compressed with the probe. It is customary to acquire and store the images in a side- by-side format where the ultrasound image is displayed on the left of the screen and the

elastography image is displayed on the right or in a single image display format where the elastography image is overlaid on top of the ultrasound image.

At step 62 the clinician may review the image frames stored in the buffer and delete (trim) frames acquired prior to or after the probe pressure was variably applied. For example, if several seconds elapsed during triggered acquisition before the clinician started to increase the probe pressure, frames acquired during that time may be eliminated from storage. At step 64 elastogram images are produced from the sequence of images stored in the buffer. From the elastogram images and also the B mode images, the clinician identifies an ROI at 66 for which stiffness is to be quantified. This may be done by placing a ROI delineator of the desired size and shape, such as a circle, over a tissue mass seen in the elastogram image. Alternatively the ROI may be identified by tracing its outline by freehand tracing. The outline of the ROI may be placed slightly inside the apparent boundary of the mass to capture only suspect tissue and no normal tissue in the ROI for the subsequent stiffness measurement. During image analysis, the ROI tracks the motion of the tissue mass from frame-to-frame to compensate for tissue motion resulting from patient breathing, probe movement, etc. during image acquisition.

At step 68 a mean instantaneous strain curve is calculated from the strain data encompassed within the identified ROI. At step 70 a strain rate curve is calculated from the mean strain data. At step 72 the total (cumulative) strain curve is computed and displayed for the duration of the tissue pressure variation. The steps of FIGURE 2 may be more fully

understood by reference to the following figures which show the results of each step. FIGURE 3 shows a two-dimensional B mode ultrasound image of a region of a phantom which contains a hard (stiff) tissue- mimicking target 80. The circular hard target 80 is barely visible against the simulated normal

background tissue surrounding the target, as is characteristic of many masses in the body. In this example the clinician has placed small calipers at the top and bottom of the hard target to measure its size which, in this example is 1.12 cm as shown below the image. When the target 80 is beneath the probe and seen in the image, the clinician triggers acquisition to store a sequence of elastography images as pressure is gradually applied with the probe. As the probe pressure is continually

increased (or decreased) , the white bar 82 below the image begins to fill with color or extend to the right as the acquired elastography images fill the image buffer. The buffer bar 82 indicates the remaining capacity of the buffer and hence the time remaining during which images can be acquired and stored as pressure is varied. When the clinician has applied a maximum pressure to compress the tissue of the ROI, generally when the tissue can be compressed no further with a moderate pressure, image

acquisition is terminated. The clinician can then review and trim the images in the buffer by replaying them as a repeating loop. As the images are

replayed, a marker 84 moves along the buffer bar 82 to indicate the location in the image sequence of the image currently displayed on the screen.

With reference to the acquired sequence of strain images (elastograms ) in the image buffer, each strain image indicates the net deformation of tissue at each point in the image from one frame to the next. As explained above, this can be done in a variety of ways, such as by integrating the velocity (Doppler phase shift) values for common points between successive images or identifying the

displacement of the speckle pattern of image points from one image to the next. An elastogram 86, overlaid over the B mode image, is shown in FIGURE 4 with the hard target 88 shown as a different color than that of the background tissue surrounding it.

FIGURE 5 shows two circles 88' and 90 which identify ROIs in the elastogram 86 at the top of the screen. To the left of the elastogram 86 is the corresponding B mode image. The circle 88'

identifies the hard tissue mass in the image and encompasses most of the pixels of the mass. The circle 90 identifies an ROI of normal tissue to the left of the mass. Mean instantaneous strain curves 92 and 94 are shown for the pixel values encompassed in the ROI circles for the sequence of elastograms. This is done by summing and averaging all of the strain values (pixel color values) in the ROI for each elastogram in the sequence, then expressing the change from one frame to the next as a percentage.

For example, if the tissue points in a ROI had a mean location of 1.00 cm in one frame and a mean location of 1.04 cm in the next frame, the percentage change from one frame to the next would be 0.04 cm divided by 1.00 cm times 100% which yields a 4% change.

These incremental changes from one frame to another are plotted as curves 92 and 94 in FIGURE 5, with curve 92 for the hard tissue ROI 88' and curve 94 for the normal (softer) tissue ROI 90. These values of curves 92 and 94 can be used for the next step but in actual practice these incremental values can be very small. For example, the gradations along the y-axis in this example are increments of 0.005%. Hence it is generally preferable to scale these increments to more sizeable values. In an implementation of the present invention, the mean instantaneous strain values are scaled upward by multiplying them by the system PRF, which can be 50, 60, or 70 Hz or more. The scaled curves are of the same shape as those of FIGURE 5 but of larger values of the percentage change in strain per second, or the strain rate as a function of time as shown by the x-axis in FIGURE 5. The curves 92 and 94 for the ROIs show peak values at the time of maximum applied probe pressure, and decline thereafter as the probe pressure is

maintained at its maximum value because there is no further deformation of the tissue.

From the scaled strain rate values of FIGURE 5 cumulative or total strain curves 96 and 98 are formed as shown in FIGURE 6. These curves are produced by calculating, for each point in time along the x-axis, the cumulative area under the strain rate curve of FIGURE 5 up to that point in time. This results in curves such as those of 96 and 98, which increase in amplitude fairly rapidly as the tissue is compressed (probe pressure increases) , and reach a fairly steady plateau level after the time of maximum applied pressure. The plateau level is the maximum value of the total strain curve and is a reproducible indicator of the stiffness of the tissue of the ROI .

In the example of FIGURE 6, the normal tissue total strain curve 98 reaches a plateau level of around 7.8%, and the hard tissue total strain curve reaches a plateau level of around 3.4%. The total

deformation of the hard ROI and the normal tissue is dependent upon the relative elasticity of their respective materials and is quantified by these measures .

FIGURES 7 and 8 illustrate that the total

(cumulative) strain measures are relatively

independent of the technique for applying pressure with the probe and will produce repeatable

measurements. FIGURE 7 shows the pressure applied with the probe for a very short duration and

thereafter held at its maximum applied pressure until the end of image acquisition. In FIGURE 7 the probe pressure is applied in a span of about one second and then maintained, compared with the five seconds of pressure increase seen in FIGURE 5. The strain rate curves 192 and 194 in FIGURE 7 are seen to rapidly rise, then fall, as the probe pressure is applied and maintained at its maximum level. When the resultant total strain curves 196 and 198 are calculated for the hard and normal tissues as shown in FIGURE 8, it is seen that each curve quickly rises as pressure is applied and thereafter maintains its maximum

(plateau) level until the end of image acquisition. As seen in FIGURE 8, the maximum (plateau) levels of the total strain measurement for the hard and normal tissue regions are virtually the same as those shown previously in FIGURE 6, at 3.2% and 7.8%,

respectively. Thus, the system and technique of the present invention produce relatively constant

measures of tissue stiffness in the presence of different techniques for applying pressure for deformation with the ultrasound probe.

The present invention may be implemented on an ultrasound cart as part of the ultrasound image acquisition and processing system, or may be

implemented for post-processing analysis in an image workstation, for instance.

Claims

WHAT IS CLAIMED IS:

1. An ultrasonic diagnostic imaging system for producing quantified measures of tissue stiffness comprising:

an ultrasound probe with an array transducer operable for acquiring a sequence of frames of elastography image data of an image field as a varying pressure is applied to the image field region with the probe;

a strain estimator, responsive to the

elastography image data for producing estimated strain values at a plurality of points in the image field of the sequence of elastography image frames; a region of interest (ROI) selector operable to indicate a location of target tissue in the image field;

a strain rate calculator, responsive to strain values from the time sequence of image frames, to produce a curve or profile of the change in strain in the ROI as the probe pressure is varied; and

a total strain calculator, responsive to the change in strain in the ROI, which produces a measure of the cumulative strain of the ROI as the probe pressure is varied.

2. The ultrasonic diagnostic imaging system of Claim 1, wherein the strain rate calculator further comprises a mean strain calculator, responsive to strain values from the time sequence of elastography image frames, to produce a curve or profile of the change in the mean instantaneous strain in the ROI from one image frame to the next as the probe

pressure is varied.

3. The ultrasonic diagnostic imaging system of Claim 2, wherein the strain rate calculator further is operable to scale mean strain values to produce a curve or profile of strain rate versus time in the ROI as the probe pressure is varied.

4. The ultrasonic diagnostic imaging system of Claim 3, wherein the strain rate calculator is further operable to scale mean instantaneous strain values as a function of the pulse repetition

frequency of the ultrasound system.

5. The ultrasonic diagnostic imaging system of Claim 1, wherein the ROI selector further comprises a user control by which a user can manually identify an

ROI in an image.

6. The ultrasonic diagnostic imaging system of Claim 1, wherein the ROI selector is further operable to track the location of target tissue from frame to frame .

7. The ultrasonic diagnostic imaging system of Claim 6, further comprising a display responsive to the strain image processor and the total strain calculator for producing a display of an elastogram and a total strain curve,

wherein the region of interest is shown on the elastogram.

8. The ultrasonic diagnostic imaging system of Claim 1, further comprising a frame buffer which is operable to store a sequence of image frames as the pressure applied to the image field region with the probe is varied.

9. The ultrasonic diagnostic imaging system of Claim 8, wherein the frame buffer is triggered by a user to store sequentially acquired image frames.

10. The ultrasonic diagnostic imaging system of Claim 8, further comprising a display responsive to the storage of image frames by the frame buffer which displays an indication of the filling of the frame buffer with stored image frames.

11. A method for producing a quantified measure of tissue stiffness in a region of interest

comprising :

acquiring a sequence of ultrasound images with an ultrasound probe as the region of interest is variably compressed by the probe;

producing a sequence of strain images from data of the sequence of ultrasound images;

identifying a region of interest (ROI) in one of the ultrasound images;

computing a curve or profile of the change in strain in the ROI as the ROI is variably compressed by the probe; and

computing a curve or profile of total

(cumulative) strain in the ROI from the change in strain in the ROI as the ROI is variably compressed by the probe .

12. The method of Claim 11 wherein computing a curve or profile of the change in strain in the ROI further comprises:

computing a curve or profile of mean

instantaneous strain of the ROI for a time interval during which the ROI is variably compress by the probe; and

using the mean instantaneous strain values to compute a strain rate curve or profile.

13. The method of Claim 12, wherein using the mean strain values to compute a strain rate curve or profile further comprises scaling the mean

instantaneous strain values.

14. The method of Claim 11, further comprising triggering the storage of a sequence of acquired images in a frame buffer.

15. The method of Claim 14, further comprising trimming the number of images stored in the frame buffer .