This invention relates to medical diagnostic ultrasound systems and, in particular, to ultrasound systems which perform imaging and therapy for

ultrasound therapy and drug delivery.

Ischemic stroke is one of the most debilitating disorders known to medicine. The blockage of the flow of blood to the brain can rapidly result in paralysis or death. Attempts to achieve

recanalization through thrombolytic drug therapy such as treatment with tissue plasminogen activator (tPA) has been reported to cause symptomatic intracerebral hemorrhage in a number of cases. Advances in the diagnosis and treatment of this crippling affliction are the subject of continuing medical research.

International patent publication WO 2008/017997 (Browning et al . ) describes an ultrasound system which provides microbubble-mediated therapy to a thrombus such as one causing ischemic stroke.

Microbubbles are infused or developed into the bloodstream in the vicinity of a thrombus.

Ultrasound energy is delivered to the microbubbles at the thrombus to disrupt or rupture the microbubbles.

This microbubble activity can in many instances aid in dissolving or breaking up the blood clot and return a nourishing flow of blood to the brain and other organs. Such microbubble activity can be used to deliver drugs encapsulated in microbubble shells, and well as microbubble-mediated sonothrombolysis.

High mechanical index (MI) impulses from a diagnostic ultrasound system have been utilized in small animal models to target drug delivery and enhance thrombolysis in the presence of intravenously infused microbubbles . These high MI impulses can induce inertial cavitation (IC) of the microbubbles, which may also cause unwanted bioeffects such as hemorrhage, cell death, and cardiac arrhythmias when using transthoracic or transcranial impulses. At an intermediate MI, stable cavitation (SC) of

microbubbles is induced which may produce equivalent or higher levels of thrombus dissolution or drug delivery as that achieved with IC but without

unwanted bioeffects. Cavitation monitoring and control techniques are required to reproducibly induce the desired state and level of cavitation to achieve the desired therapeutic bioeffects without the harmful ones. Cavitation monitoring and control are also important for thermal HIFU where both bubble generation/distribution and temperature distribution are important.

A large variety of temporal and spectral based cavitation detection techniques have been developed and used by different groups based on various bubble acoustic signatures, including detection of "spikes" in received signals; post-excitation acoustic

emissions; sub- and ultra- harmonic generation;

broadband noise emission; increased spectral noise between harmonics; Doppler spectral broadening; and reduced cross correlation of consecutive scattered signals. See, e.g., M. Hodnett et al . , "A detector for monitoring the onset of cavitation during

therapy-level measurements of ultrasonic power, " Journal of Physics : Conference series 1, at pp 112-17

(2004) . These techniques are often restricted to laboratory use because of limited sensing volume associated with the single-element transducer

generally used as a cavitation detector.

Array-based imaging systems can be used as cavitation detectors over a large control volume, i.e., the entire field of view of the ultrasound array. Imaging systems, however, suffer from limited frequency bandwidth. Additionally, in vivo,

frequency-dependent tissue attenuation further limits the available bandwidth. In transcranial treatment, the ultrasonic energy is delivered through the skull, which will apply an unknown amount of attenuation to the ultrasonic energy. The amount of energy actually delivered to the therapy site thus cannot be

accurately known from the energy setting of the ultrasound system. Therefore, new cavitation

monitoring/control techniques are needed for human clinical applications.

In accordance with the principles of the present invention, a diagnostic ultrasound system and method are described which provide an anatomical indication of the locations of inertial and stable cavitation during ultrasonic therapy or drug delivery. A cavitation image is developed from cavitation indicia of echo signals at the site of treatment. The cavitation image provides a visual indication of locations of stable and inertial cavitation in the body. The clinician can then adjust the transmit energy of the ultrasound system to provide the desired therapeutic effect. In an automated control system the adjustment can be performed automatically to maintain a desired level of cavitation activity. An alert can be produced if the level of cavitation varies from a desired level.

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 illustrate the delivery of ultrasonic therapy in a two dimensional (2D) imaging plane

FIGURE 3 illustrates the delivery of ultrasonic therapy in a three dimensionally image volume.

FIGURE 4 illustrates the harmonic content of a cavitation echo which is used in an embodiment of the present invention.

FIGURE 5 is a flowchart illustrating the

development of a cavitation image in accordance with the present invention.

FIGURE 6 is a flowchart illustrating treatment with cavitation monitoring and control in accordance with the principles of the present invention.

Referring to FIGURE 1, an ultrasound system constructed in accordance with the principles of the present invention is shown in block diagram form. Two transducer arrays 10a and 10b are provided for transmitting ultrasonic waves and receiving echo information. In this example the arrays shown are two dimensional arrays of transducer elements capable of providing 3D image information although an

implementation of the present invention may also use two dimensional arrays of transducer element which produce 2D (planar) images. The transducer arrays are coupled to microbeamformers 12a and 12b which control transmission and reception of signals by the array elements. Microbeamformers are also capable of at least partial beamforming of the signals received by groups or "patches" of transducer elements as described in US Pats. 5,997,479 (Savord et al . ) ,

6,013,032 (Savord), and 6,623,432 (Powers et al . ) In this example the two arrays and microbeamformers are part of a headset which locates the arrays on the temples on both sides of the head for transcranial delivery. In other implementations such as cardiac delivery a probe with a single 2D array and

microbeamformer may be used. Signals are routed to and from the microbeamformers by a multiplexer 14. The multiplexer is coupled to a transmit/receive (T/R) switch 16 which switches between transmission and reception and protects the system beamformer 20 from high energy transmit signals. The transmission of ultrasonic pulses from the transducer arrays 10a and 10b under control of the microbeamformers 12a and 12b is directed by the transmit controller 18 coupled to the T/R switch, which receives input from the user's operation of the user interface or control panel 38.

The partially beamformed signals produced by the microbeamformers 12a, 12b are coupled to the system beamformer 20 where partially beamformed signals from the individual patches of elements of an array are combined into a fully beamformed signal. For example, the system beamformer 20 may have 128 channels, each of which receives a partially

beamformed signal from a patch of 12 transducer elements. In this way the signals received by over 1500 transducer elements of a two dimensional array can contribute efficiently to a single beamformed signal.

The beamformed signals are coupled to a

fundamental/harmonic signal separator 22. The separator 22 acts to separate linear and nonlinear signals so as to enable the identification of the strongly nonlinear echo signals returned from

microbubbles . The separator 22 may operate in a variety of ways such as by bandpass filtering the received signals in fundamental frequency and

harmonic frequency bands, or by a process known as pulse inversion harmonic separation. A suitable fundamental/harmonic signal separator is shown and described in international patent publication WO 2005/074805 (Bruce et al . ) The separated signals are coupled to a signal processor 24 where they may undergo additional enhancement such as speckle removal, signal compounding, and noise elimination.

The processed signals are coupled to a B mode processor 26 and a cavitation processor 28. The B mode processor 26 employs amplitude detection for the imaging of structures in the body such as muscle, tissue, and blood cells. B mode images of structure of the body may be formed in either the harmonic mode or the fundamental mode. Tissues in the body and microbubbles both return both types of signals and the harmonic returns of microbubbles enable

microbubbles to be clearly segmented in an image in most applications. The processor detects the signal characteristics of cavitation and produces cavitation image signals as described below. The system may also include a Doppler processor which processes temporally distinct signals from tissue and blood flow for the detection of motion of substances in the image field including microbubbles. The anatomic and cavitation signals produced by these processors are coupled to a scan converter 32 and a volume renderer

34, which produce image data of tissue structure, flow, cavitation, or a combined image of several of these characteristics. The scan converter converts echo signals with polar coordinates into image signals of the desired image format such as a sector image in Cartesian coordinates. The volume renderer 34 converts a 3D data set into a projected 3D image as viewed from a given reference point as described in US Pat. 6,530,885 (Entrekin et al . ) As described therein, when the reference point of the rendering is changed the 3D image can appear to rotate in what is known as kinetic parallax. This image manipulation is controlled by the user as indicated by the Display Control line between the user interface 38 and the volume renderer 34. Also described is the

representation of a 3D volume by planar images of different image planes, a technique known as

multiplanar reformatting. The volume renderer 34 can operate on image data in either rectilinear or polar coordinates as described in US Pat. 6,723,050 (Dow et al . ) The 2D or 3D images are coupled from the scan converter and volume renderer to an image processor 30 for further enhancement, buffering and temporary storage for display on an image display 40.

A graphics processor 36 is also coupled to the image processor 30 which generates graphic overlays for displaying with the ultrasound images. These graphic overlays can contain standard identifying information such as patient name, date and time of the image, imaging parameters, and the like, and can also produce a graphic overlay of a beam vector steered by the user as described below. For this purpose the graphics processor received input from the user interface 38. In an embodiment of the present invention the graphics processor can be used to overlay a cavitation image over a corresponding anatomical B mode image as described below. The user interface is also coupled to the transmit controller 18 to control the generation of ultrasound signals from the transducer arrays 10a and 10b and hence the images produced by and therapy applied by the

transducer arrays. The transmit parameters

controlled in response to user adjustment include the MI (Mechanical Index) which controls the peak intensity of the transmitted waves, which is related to cavitational effects of the ultrasound, steering of the transmitted beams for image positioning and/or positioning (steering) of a therapy beam as discussed below .

The transducer arrays 10a and 10b in the

transcranial headset implementation transmit

ultrasonic waves into the cranium of a patient from opposite sides of the head, although other locations may also or alternately be employed such as the front of the head or the sub-occipital acoustic window at the back of the skull. The sides of the head of most patients advantageously provide suitable acoustic windows for transcranial ultrasound at the temporal bones around and above the ears on either side of the head. In order to transmit and receive echoes through these acoustic windows the transducer arrays must be in good acoustic contact at these locations which may be done by holding the transducer arrays against the head with the headset.

FIGURE 2 illustrates a two dimensional imaging implementation of the present invention. In this example the transducer array 122 is a one or two dimensional array which performed 2D imaging. This transducer array, like the other arrays described herein, is covered with a lens 124 which electrically insulates the patient from the transducer array and in the case of a one dimensional array may also provide focusing in the elevation (out-of-plane) dimension. The lens is pressed against the skinline 100 for acoustic coupling to the patient. The transducer array 122 is backed with acoustic damping material 126 which attenuates acoustic waves

emanating from the back of the array to prevent their reflection back into the transducer elements. Behind this transducer stack is a device 130 for rotating the image plane 140 of the array. The device 130 may be a simple knob or tab which may be grasped by the clinician to manually rotate the circular array transducer in its rotatable transducer mount (not shown) . The device 130 may also be a motor which is energized through a conductor 132 to mechanically rotate the transducer as discussed in US Pat.

5,181,514 (Solomon et al . ) When a two dimensional array is used, the image plane can be rotated

electronically with no physical movement of the transducer array. The image plane can also be steered in elevation as described in US Pat.

7,037,264. Rotating the one dimensional array transducer 122 as indicated by arrow 144 will cause its image plane 140 to pivot around its central axis, enabling the repositioning of the image plane for full examination of the vasculature in front of the transducer array. As discussed in the '514 patent, the planes acquired during at least a 180° rotation of the array will occupy a conical volume in front of the transducer array, which may be rendered into a 3D image of that volumetric region. Other planes outside this volumetric region may be imaged by repositioning, rocking or tilting the transducer array in its headset in relation to the skull 100.

If a stenosis is found in the image of the plane being imaged, the therapeutic beam vector graphic 142 can be steered by the clinician to aim and focus the beam at the stenosis 144 and therapeutic pulses applied to disrupt the microbubbles at the site of the stenosis.

FIGURE 3 illustrates a 3D imaging implementation of the present invention which uses a 2D matrix array transducer 10a. In this illustration the transducer array 10 is held against the skinline 100 of the patient with the volume 102 being imaged projected into the body. The user will see a 3D image of the volume 102 on the display of the ultrasound system in either a multiplanar or volume rendered 3D

projection. The user can manipulate the kinetic parallax control to observe the volume rendered 3D image from different orientations. The user can adjust the relative opacity of the tissue and flow components of the 3D image to better visualize the vascular structure inside the brain tissue as

described in US Pat. 5,720,291 (Schwartz) or can turn off the B mode (tissue) portion of the display entirely and just visualize the flow of the vascular structure inside the 3D image volume 102.

When the site of the treatment such as a

thrombus 144 is being imaged in the volume 102, a microbubble contrast agent is introduced into the patient's bloodstream. In a short time the

microbubbles in the bloodstream will be pumped through to the vasculature of the treatment site and appear in the 3D image. Therapy can then be applied by agitating or breaking microbubbles at the site of the stenosis in an effort to dissolve the blood clot. The clinician activates the "therapy" mode, and a therapy graphic 110 appears in the image field 102, depicting the vector path of a therapeutic ultrasound beam with a graphic thereon which may be set to the depth of the thrombus. The therapeutic ultrasound beam is manipulated by a control on the user

interface 38 until the vector graphic 110 is focused at the site of the blockage. The energy produced for the therapeutic beam can be within the energy limits of diagnostic ultrasound or in excess of the

ultrasound levels permitted for diagnostic

ultrasound, in which case the microbubbles at the site of the blood clot will be sharply broken. The energy of the resulting microbubble ruptures will strongly agitate a blood clot, tending to break up the clot and dissolve it in the bloodstream. In many instances insonification of the microbubbles at diagnostic energy levels will be sufficient to dissolve the clot. Rather than breaking in a single event, the microbubbles may be vibrated and

oscillated, and the energy from such extended

oscillation prior to dissolution of the microbubbles can be sufficient to break up the clot.

In accordance with the principles of the present invention, the energy level of transmitted ultrasound is monitored and controlled to produce a desired level of cavitation at the site of the therapy.

Stable cavitation is indicated by the occurrence of combined sub- and ultra-harmonics, while inertial cavitation is indicated by a combination of the following features: (1) elevated noise levels between all (sub, 1st, ultra- and 2nd and higher) harmonic components (or spectral peaks) in the transducer' s lower frequency band is indirectly indicative of the high-frequency noise directly associated with fast collapsing of inertially

cavitating bubbles. The noise increase at the lower frequency band is not a direct indicator of inertial cavitation; (2) Doppler spectrum changes includes both bubble destruction (associated with inertial and stable cavitation) and flow velocity (speed/

direction) . The difference between the broadened spectrogram at a high MI and the "true" spectrogram (flow waveform) at a very low MI is an accurate indicator of bubble destruction; (3) RF echo traces and their cross-correlations are also good indicators for when, how often, and most importantly, how fast the bubble destruction occurs. Fast destruction is indicative of instant bubble disintegration that most often leads to inertial cavitation.

The degree of cavitation caused by transmitted ultrasound pulses or waves is related to the MI setting of the ultrasound transmitter. At a low MI of 0.1 and below, microbubbles are agitated or oscillated but generally are not ruptured or

dissolved. They remain stable and intact in the presence of low MI ultrasound. At around an MI of

0.3, stable cavitation can occur. Stable cavitation results in significant mechanical effects of

disrupted and rupturing microbubbles but without deleterious effects which can damage cells in the body. Above around an MI of 0.6, unwanted bioeffects can occur.

FIGURE 4 illustrates a typical spectrum of an echo returned from cavitation activity at an MI of 0.3. The fundamental frequency of the transmitted ultrasound is f ₀ , the major peak in the spectrum of the returning echo. Second (f ₂ ) , third (f ₄ ) , and fourth (f ₆ ) harmonic peaks are also seen in the spectrum. Between the fundamental (f ₀ ) and second harmonic (f ₂ ) peak is an ultra-harmonic peak at fi . s , which is a peak between the integer harmonics. These peaks are detected by filtering and combined in the cavitation processor 28 as shown in FIGURE 5 to create a cavitation image indicating the presence, type, and location of cavitation in the ultrasound field.

In an implementation of the present invention a series of pulses or beams are transmitted across the image field to form an anatomical image of the region of the therapy. Interleaved with this standard imaging is the transmission of cavitation pulses or beams to produce echoes for the detection of

cavitation at spatial locations in the image field. The cavitation pulses or beams may scan the same area or volume as the imaging pulses or beams, or only a portion thereof around the therapy site. The imaging pulses are transmitted at a low MI of 0.1 or less, for instance, to image the tissue structure in the image field and microbubbles in the vasculature of the tissue. The cavitation pulses are transmitted at a higher MI of 0.3 or greater or at a level where the onset of cavitation is anticipated. The echoes from the imaging and cavitation pulse sequences are processed as shown by the flowchart of FIGURE 5.

This implementation of the present invention produces an anatomical image of the location of cavitation in the image field by combining a B mode image with detected cavitation at spatial locations in the anatomy of the B mode image. The RF echo data from the B mode pulses is processed starting at 60. The echo data is filtered at 62 (generally by the signal processor 24) and envelope detected at 64 to produce a B mode image at 66, the latter being performed by the B mode processor 26. The RF data may all be at the second harmonic, or at the fundamental frequency for tissue with an overlay of second harmonic signals from microbubbles. Processing of the RF echo data from the cavitation pulse sequence begins at 70.

Filtering is performed to detect the signal content at three spectral locations: the fundamental

frequency (f ₀ ) at 72, between the first ultra-harmonic frequency and the second harmonic ( fi . ₇₅ ) at 74, and the first ultraharmonic frequency ( fi . s ) at 76. The filtered RF signals are envelope detected to produce Fundamental data at 82, Noise data at 84 and

Ultraharmonic data at 86. These detected signals are then combined and analyzed as shown at 92 and 94 to detect the presence of inertial and stable

cavitation. If the ratio of the detected

Ultraharmonic data at a location in the image field to the Noise data at that location is greater than

6dB, stable cavitation is determined to be present at that location (94) . This determination is

essentially based on the assessment that the

Ultraharmonic data being much greater than the Noise data at a spectral location that should be low in harmonic content indicates a strong echo return from microbubble activity, but insufficient Noise data to indicate strong bubble ruptures. If this ratio is less than 6dB and the ratio of the Noise data to the Fundamental data at the location is greater than or equal to 25dB, then inertial cavitation is determined to be present at that location. This determination is essentially based on the assessment that

Ultraharmonic data from microbubble activity is much less than the Noise data and that the Noise data level is significantly in excess of the strong

Fundamental data of the echo, indicating excessive bubble rupturing. When stable cavitation is

determined at a location in the image field, the picture element at that location is colored green in this example as indicated at 154. When inertial cavitation is determined at a location, the picture element at that location is colored red as indicated at 152. The picture elements of the B mode image are all nominally gray in this example as indicated at

156. The picture elements in the B mode image may be replaced by red and green picture elements at

locations where inertial and stable cavitation are determined in the image field. Alternatively, an inertial cavitation image is formed of just the red picture elements, a stable cavitation image is formed of just the green picture elements and the inertial and stable cavitation images overlaid over the B mode image to spatially present the cavitation locations to the viewer. Another alternative is to combine the inertial and stable cavitation images into one cavitation image for overlay over the B mode image. Each approach shows the user where cavitation has been detected in the image field, and the type of cavitation detected by the color coding and

brightness .

FIGURE 6 is an exemplary flowchart of a therapy procedure which may be performed in accordance with the present invention. A microbubble contrast agent is introduced into the bloodstream of the patient and at 160 low MI harmonic imaging is performed to visualize the microbubbles in the bloodstream and to locate the target, the thrombus, blood clot, or site where drug delivery is to take place. The probe and/or the image field may be scanned and moved until the site of the procedure is visualized and the clinician sees that microbubbles are present at the site. At 163 the clinician turns up the power of the ultrasound system to a higher MI and performs

cavitation imaging. Cavitation imaging may also use a different pulse than the initial harmonic imaging. For example, the initial survey imaging may use a short pulse for good spatial resolution, whereas cavitation imaging may use a longer pulse (greater number of cycles) to provide good spectral resolution for the spectral measurements performed by the filtering in FIGURE 5. Coded pulses and chirp

(variable frequency) pulses accompanied by

deconvolution filtering may also be suitable.

Cavitation imaging is performed while the MI is adjusted until at least some stable cavitation is seen by green colorizing in the B mode image. When some cavitation is seen, the clinician increases the power further until stable cavitation is seen where microbubbles are present and little or no inertial cavitation is seen (red color) . If the power is turned up too far and a significant amount of red inertial cavitation is seen, the clinician turns down the power so that most or all of the cavitation seen is stable cavitation. Therapy may then continue at this setting at 166. Another possibility is that the moderate, diagnostic level setting for cavitation imaging is insufficient for the needed therapy and more power is needed, or different pulses or pulse sequences are used for the therapy. The vector graphic 110, 142 may then be turned on and set for the therapy site and therapy is enabled at 166 to direct the requisite ultrasonic therapy pulses to the indicated therapy site. Echoes returning from the higher level therapy pulses may also be processed to form a cavitation image in response to the higher level therapy pulses. Anatomical imaging, cavitation imaging, and therapy transmission are time- interleaved so that the progress of the therapy can be monitored by visualization in real time and the MI setting adjusted as needed. As previously mentioned, the cavitation imaging data can be used as feedback in an automated implementation, automatically keeping the cavitation in the stable cavitation regime should the detected cavitation stray into inertial

cavitation where undesired bioeffects may occur.

There may be periods where the microbubbles have become substantially completely ruptured or

dissolved, in which case therapy is paused to allow blood flow to replenish the supply of microbubbles at the therapy site. When the desired result of the therapy, e.g., release of the desired amount of a drug or dissipation of a thrombus, is seen, the therapy is terminated.

The ultrasound system of FIGURE 1 also includes an inertial cavitation detector 50 and a loudspeaker 42. These features are useful in an automated or monitoring implementation to alert a user when undesired inertial cavitation is occurring. The inertial cavitation data, e.g., red picture elements indicative of inertial cavitation, are applied to the inertial cavitation detector. When this data such as the number of red pixels in a cavitation image is greater that a preset level, preferably one that is user-adjustable, an alert is sounded by the

loudspeaker 42, is displayed on the display screen 40, or both. This alerts the clinician that an undesired level of inertial cavitation has been sensed, giving the clinician the immediate

opportunity to correct the situation by adjusting the transmit level setting of the ultrasound system.