Contrast agent imaging with destruction pulses in diagnostic medical ultrasound

[0002] 1. Field of Invention

[0003] This invention relates generally to an improvement in diagnostic medical ultrasound imaging, and more specifically to more effective medical imaging of a human or animal due to improved utilization of an ultrasound contrast agent.

[0004] 2. Description of the Prior Art

[0005] A major goal of contrast agent imaging in diagnostic medical ultrasound is the measurement of physiological functions, such as perfusion of blood into tissue, by measuring the flow of contrast agent into tissue. In order to accomplish this, it is important to differentiate contrast agent from tissue.

[0006] One method for achieving the differentiation of contrast agent and tissue is the loss-of-correlation (LOC) method in which multiple pulses are fired into a body containing contrast agent. The first pulse produces a return signal from both contrast agent and tissue, and simultaneously destroys contrast agent. Since contrast agent is destroyed, the second pulse produces relatively more signal from tissue. By subtracting the two signals, the signal from the contrast agent is enhanced; while the signal from tissue, which is in both pulses, is suppressed. The two pulses can be acquired in the same frame (e.g. color Doppler processing) or in different frames. The more contrast agent that is destroyed between firings, the better the differentiation between contrast agent and tissue.

[0007] This method has drawbacks. First, a single pulse will rarely destroy all the contrast agent along a given ultrasound line. As a result, many pulses are needed to destroy the contrast agent. Firing many pulses along the same line will reduce the system frame rate, and thereby increase motion artifacts due to tissue motion, especially in cardiac applications. The system frame rate is defined as the actual frame rate at which images can be obtained by the ultrasound system. Secondly, the pulses which best destroy contrast agent are low frequency, long duration (low bandwidth) pulses. However, for better imaging resolution, high frequency, short duration (high bandwidth) pulses are needed. For second harmonic imaging, low frequency pulses are used, but these pulses are still short duration in order to provide good imaging resolution. Usually, a compromise is made between the requirements for imaging and destruction of contrast agent, further reducing the efficiency of contrast agent destruction by ultrasound pulses.

[0008] Another approach to LOC imaging, which overcomes this compromise, is to generate two types of pulses, one type for imaging and one type for destroying contrast agent. For example, a transmitter sends an imaging pulse into a region of interest in a body, followed by a pulse for destruction of contrast agent. A second imaging pulse is then sent into the region of interest in the body and the two images of the region before and after contrast agent destruction are compared.

[0009] However, this does not overcome the system frame rate problem, but in fact may make it worse. Furthermore, this approach to LOC imaging does not allow contrast agent to perfuse into a region of interest between firings of pulses for destroying contrast agent, so there is little contrast agent to image after destruction. Contrast agent perfusion imaging requires sending the pulses for destroying contrast agent only at specific intervals. Still further, a single pulse will rarely destroy all the contrast agent along a given ultrasound line.

[0010] Another approach discussed in the prior art is to use high power pulses simultaneously for imaging and destruction. In this case, low power pulses, or “locator frames” are used to image the body, and then at times determined by a timer or an ECG signal, a high power signal is sent to image the contrast agent as well as destroy it. The problem with this approach is that the same high power pulses are simultaneously used to both image and destroy the contrast agent. Therefore, trade-offs in the pulse parameters are made based on the desired amount of destruction, spatial resolution, and desired system frame rate.

[0011] Therefore, it would be desirable to improve the LOC method by enhancing the difference between contrast agent and tissue, without sacrificing the system frame rate or imaging resolution. Furthermore, an improved method for measuring physiology, such as perfusion, is needed.

[0012] One object of the invention is to enhance the difference between contrast agent and tissue without sacrificing the system frame rate or imaging resolution.

[0013] Another object of the invention is to improve the efficiency of contrast agent destruction.

[0014] Another object of the invention is to provide a better method for measuring and/or displaying physiology, such as perfusion.

[0015] A first aspect of the invention is directed to destroying contrast agent by use of high pulse repetition frequency (HPRF) destruction pulses. These pulses are transmitted at a rate faster than required to allow the pulses to propagate to the farthest boundary of the region of interest being imaged and return to the transducer.

[0016] A second aspect of the invention is directed to transmitting multiple destruction beams simultaneously in different directions to destroy contrast agent quickly over a region.

[0017] A third aspect of the invention is directed to triggering destruction frames at a fixed point of a physiological signal (e.g., every n beats of a cardiac cycle, where n is an integer), and displaying continuous imaging frames between the destruction frames.

[0018] A fourth aspect of the invention is directed to triggering destruction frames at a fixed point of a physiological signal, and displaying, between the destruction frames, imaging frames that are triggered at a different fixed point of the physiological signal.

[0019] A fifth aspect of the invention is directed to using triggered destruction frames and imaging frames and comparing imaging frames.

[0020] A sixth aspect of this invention relates to combining destruction pulses with subharmonic imaging.

[0021] These and other objects and advantages of the invention will become apparent to those skilled in the art from the following detailed description of the invention and the accompanying drawings.

[0022] FIG. 1 shows a timing diagram of a normal imaging situation.

[0023] FIG. 2 shows an HPRF timing diagram.

[0024] FIG. 3 shows focusing delay profiles for four beams without beam stagger.

[0025] FIG. 4 shows an example of using beam stagger in the preferred embodiment.

[0026] FIG. 5 is a block system diagram of a preferred embodiment for carrying out the invention.

[0027] FIG. 6 shows a timing diagram describing how the system of FIG. 5 implements a simple sequence of two destruction pulses followed by an imaging sequence.

[0028] FIG. 7 shows a timing diagram describing how the system of FIG. 5 changes parameters between destruction pulses.

[0029] FIG. 8 shows a flow chart explaining one mode of operation of the system controller of FIG. 5.

[0030] FIG. 9 shows a flow chart explaining another mode of operation of the system controller of FIG. 5.

[0031] FIG. 10 is a comparison of an image obtained just before a destruction frame and an image obtained just after a destruction frame.

[0032] FIG. 11 is a comparison of an image obtained just after a destruction frame and an image obtained from an nth imaging frame after the destruction frame.

[0033] Imaging pulses are pulses which are transmitted, received, processed and displayed. Imaging pulses in some cases can also be destructive to contrast agent. On the other hand, destruction pulses are herein defined as pulses which are transmitted into a body, but not used for imaging. Since destruction pulses are not used for imaging, there is no need to optimize destruction pulse parameters for imaging. These destruction pulses can be optimized for destruction of contrast agent by using low frequency, low bandwidth and high power pulses.

[0034] The first aspect of the invention is directed to improving the efficiency of destruction of contrast agent by using high pulse repetition frequency (HPRF) destruction pulses. Here, HPRF destruction pulses are defined as pulses which are transmitted at a rate faster than required to allow the pulses to propagate to the farthest boundary of a region of interest and return to the transducer. Since destruction pulses are not processed or displayed, it is not necessary to wait for the ultrasound signal to propagate to the farthest boundary of a region of interest being imaged and for signals from deep tissue interfaces to return to the transducer. Rather, multiple pulses are fired very rapidly into the body. Accordingly, in the preferred embodiment of the invention, the pulse repetition frequency (PRF) can be as high as allowed by the ability of the transmitter to reset and transmit another pulse after a pulse has been fired.

[0035] Multiple destruction pulses are normally needed to adequately destroy contrast agent in the body. Conventional systems transmit at a pulse repetition interval sufficient to allow the transmitted pulse to travel to the farthest boundary of a region of interest being imaged and return to the transducer, as well as to allow time for any reverberations of the pulse to die out. For example, if a system is being used to image 100 mm in depth, then in order to allow the ultrasound to travel 100 mm and return to the transducer, assuming a speed of sound of 1.54 mm/microsecond (us), the transmitter must wait about 130 us between transmitter firings (ignoring the time for the reverberations to die out). If four pulses are needed to destroy the contrast agent along one line, then the system must wait 390 us to destroy all the contrast agent along that line. This represents a significant problem if there is significant tissue motion during that time interval. However, destruction pulses do not have to be received and processed; their only purpose is to destroy the contrast agent. Therefore, instead of waiting 130 us between firings, pulses can be fired at a more rapid rate. Preferably, these pulses are fired at the maximum rate allowed by the transmitter. For example, if it requires 10 us to reset a transmitter after firing, then 4 pulses can be fired along one line in only 30 us, rather than 390 us. This represents a significant improvement in the time it takes to destroy contrast agent in the body. Of course, one can practice this invention by transmitting at any PRF faster than required to receive echoes from the farthest boundary of a region of interest, in this example, at any interval between 10 us and 130 us. However, at very shallow depths, HPRF pulses are less advantageous because the transmitter reset time is a large fraction of the pulse propagation time.

[0036] FIGS. 1 and 2 show simple timing diagrams for an example illustrating the difference between normal pulse transmission and HPRF pulse transmission. The timing diagrams depict the timing of actions for a transmitter and receiver. FIG. 1 shows a normal imaging situation. First the transmitter is set up, then a pulse is transmitted. Then the receiver waits for the received signals and processes them, before the whole sequence is repeated. FIG. 2 shows an HPRF timing diagram. In this example, the transmitter is set up, and then a destruction pulse is fired. The transmitter is then reset and another destruction pulse is fired. There is no pause to receive and process the returning signals, since only destruction is intended.

[0037] Another improvement can be made to HPRF destruction pulses to increase the efficiency of destruction by changing pulse parameters between the multiple firings of HPRF destruction pulses. For example, a maximum number of bubbles of contrast agent are destroyed at the point of peak intensity, which is very close to the transmit focus. In a preferred embodiment, the transmit focus is changed between successive pulse firings to uniformly destroy all the contrast agent along a given line.

[0038] For example, in order to destroy all the contrast agent along a given line, the system begins by firing one or more destruction pulses focused at a first depth. The focal depth is then changed, and mutiple pulses fired at a second depth. Optionally, the focal depth is then changed further, and one or more pulses are fired at one or more additional depths. In the preferred embodiment of the invention, this sequence is then repeated for several ultrasound lines covering a region. An alternative method involves firing one or more pulses focused at one depth for all the lines. Then, the focal depth is changed, and one or more pulses focused at this new depth are fired for all the lines, and this step is repeated as necessary.

[0039] Other parameters, such as transmit frequency, can also be varied to achieve substantial destruction along a line. Transmit frequency is important because lower frequency pulses have less attenuation and are able to penetrate deeper into the body. Moreover, destruction pulses can be fired at various frequencies in sequence to increase the efficiency of contrast agent bubble destruction. Since a population of bubbles consists of bubbles of varying sizes, which will preferentially burst at different resonating frequencies, destruction with pulses of different frequencies increases the efficiency of bubble destruction.

[0040] Pulse power can also be increased for better destruction coverage throughout the field of view. For example, different destruction beams focused at different focal depths can have different transmitted power levels, such that destruction beams focused at deeper depths are fired at a greater transmitted power level than destruction beams focused at shallower depths to achieve uniform destruction along a line. The government regulatory (e.g., the Food and Drug Administration) limits on transmit pulse power and the benefits of using a maximum amount of power are discussed more filly in co-pending U.S. Pat. No. 6,045,506, entitled “Ultrasonic Imaging Method and Apparatus for Adjusting Transmitted Power Levels,” filed on Aug. 31, 1998, which is assigned to the assignee of the present invention and hereby incorporated by reference. The teachings and advantages disclosed therein can be combined with any aspect of this invention. Accordingly, transmit power can also be varied as a function of angle from the center of the array; or as a combination of depth and such lateral angle.

[0041] In a second aspect of the invention, multiple transmit beam techniques are used to increase the speed and efficiency of contrast agent destruction without a reduction in the system frame rate. Cole, et al. disclose an ultrasound system capable of transmitting multiple beams in different directions simultaneously in U.S. Pat. No. 5,675,554. Other ultrasound systems capable of firing multiple beams may also be used with this invention.

[0042] In one embodiment of the invention, several closely spaced beams are sent out simultaneously. Preferably, the beams are staggered so that the different beams do not overlap as they are transmitted. FIG. 3 shows focusing delay profiles for four beams without beam stagger. In this example, the focusing delays shown are for closely spaced beams steered at slightly different angles about the center of the array. Here, negative numbers represent advance (pulses are transmitted earlier) and positive numbers represent delay (pulses are transmitted later). As shown, the delay curves cross over each other, indicating that pulses for the different beams are transmitted at the same time from some element. In this case, the pulse at that element is the superposition of the pulses required for each beam. This can increase the peak amplitude required to transmit the multiple beams correctly; and because a transmit beamformer typically has a maximum peak transmit amplitude, either the amplitude of all four beams must be decreased, or the transmit signal for some elements will be saturated. Furthermore, ultrasound pulses have a finite duration, so that interference and subsequent signal strength limitation can occur even if the pulses do not exactly overlap.

[0043] FIG. 4 shows an example of using beam stagger in the preferred embodiment. Here, a delay of slightly more than 0.8 us was added in between each set of adjacent beams. This enables the four beams to be transmitted without interference at the transducer. Hence all four beams can be transmitted without reducing the transmit amplitude and without incurring saturation at some transmitter elements.

[0044] Of course, beneficial results can also be obtained using this aspect of the invention (the use of multiple transmit beams with destruction pulses) without beam stagger. Because the destruction pulses are not used for imaging, a level of saturation or interference between the beams may be tolerated which is not acceptable in a prior art diagnostic medical ultrasound imaging system.

[0045] While the example described above was directed to closely spaced multiple transmit beams, this aspect of the invention can also be practiced with widely spaced multiple transmit beams.

[0046] In an alternate embodiment, multiple beam transmit technology is used to implement either HPRF destruction pulses and/or to vary pulse parameters between destruction pulses fired along the same line. For example, a sequence of two or more identical pulses is fired in the same direction, staggered as described above, while the receive beamformer is disabled. This allows destruction pulses to be fired at an extremely high rate, as the transmit beamformer does not need to be reset between every firing. In addition, on some beamformers, the invention may be easier to implement this way than by changing the sequence and repetition frequency of pulse firing. Alternatively, the pulses fired in the same line are focused at different depths, and again, are preferably staggered to prevent overlap.

[0047] A third aspect of the invention is directed to triggering destruction frames and displaying continuous imaging frames between the destruction frames. Here, continuous imaging frames are defined as imaging frames that are fired one after the other, with only enough time between frames for the last pulse of the previous frame to propagate to the farthest boundary of the region imaged and return to the transducer, with some overhead for the system to complete the frame. This overhead time between frames includes time for the system to remove any reverberations, process and display the image, and complete one imaging frame before beginning another imaging frame. A destruction frame consists of a series of destruction pulses, preferably swept over the entire field of view and at a variety of focal depths in order to achieve uniformity of contrast agent destruction throughout the field of view. A HPRF destruction frame consists of a series of HPRF destruction pulses. Triggered destruction frames are destruction frames fired at some time triggered from a timer or physiological signal, such as a cardiac signal (e.g., an ECG signal) or a respiratory signal. Triggering can occur at some fixed or variable time after the physiological signal. Triggered destruction frames are alternated with imaging frames fired in a continuous mode.

[0048] A fourth aspect of the invention is directed to triggering destruction frames at a fixed point of a physiological signal and displaying, between the destruction frames, imaging frames that are triggered at a different fixed point of the physiological signal.

[0049] A preferred embodiment provides a user interface that gives the operator the option of choosing either the third or fourth aspect of the invention. Therefore, destruction frames are triggered at a fixed point of a physiological signal (e.g., every n beats of a cardiac cycle or respiratory cycle, where n is an integer), and between the destruction frames, either continuous imaging frames are displayed, or imaging frames are displayed that are triggered at a different time (e.g., every m beats of a cardiac cycle or respiratory cycle, where m is an integer) than the destruction frames. For example, the imaging frames can be triggered at the same point of the cardiac cycle, but in a different cardiac cycle than the destruction frames. Alternatively, the imaging frames can be triggered at a different point in the cardiac cycle, in the same cardiac cycle(s) as or different cardiac cycle(s) than the destruction frames.

[0050] Ideally, different aspects of the invention are combined. For example, HPRF destruction frames are combined with firing at different transmit foci and using multiple transmit beams to further improve the efficiency of the contrast agent destruction.

[0051] FIG. 5 provides a schematic diagram of an ultrasound system 500 that incorporates a preferred embodiment of the invention. Ultrasound system 500 preferably includes programmable transmit beamformer 504 and receive beamformer 505 coupled to ultrasonic transducer 506 via multiplexer 507. Disclosures of such beamformers are available: for example, receive beamformers are described in U.S. Pat. Nos. 5,685,308, 5,882,307, and 5,827,188 to Wright, et al.; and transmit beamformers are described in U.S. Pat. Nos. 5,856,955 and 5,675,554 to Cole, et al., but other beamformers in the art can be used. Transmit and receive parameters are fed to beamformers 504 and 505 by beamformer controller 502, such as, but not limited to, that described in U.S. Pat. No. 5,581,517 to Gee, et al. Beamformer controller 502 controls beamforming parameters, such as transmit center frequency and bandwidth, receive center frequency and bandwidth, transmit power, receive gain, pulse repetition timing and transmit line spacing interval. Preferably, transmit beamformer 504 is also capable of transmitting multiple beams simultaneously in different directions. Transducer 506 comprises a phased array of transducer elements in one of various formats, such as a linear, curvilinear, one-dimensional, 1.5-dimensional, and two-dimensional array. Transmit beamformer 504 provides transmit waveforms to transducer 506, causing transducer 506 to emit ultrasonic pulses into the body, which contains both tissue and contrast agent scatterers. Ultrasound pulse scattering from the scatterers in the body returns ultrasound pulses to transducer 506, which sends signal waveforms to receive beamformer 505 via the multiplexer 507. The multiplexer 507 contains a transmit/receive switch which can switch between transmit and receive states for imaging pulses, but which will otherwise remain in the transmit state for destruction pulses. Receive beamformer 505 may optionally take the received signal waveforms and form outputs, which can be passed on to display controller 514 and display 516 (typically via image processor 508 and optionally through comparator 510). Image processor 508 provides signal processing of the signals generated from receive beamformer 505, and optionally provides input data to comparator 510 and memory 512. Alternatively, image processor 508 can be integrated with comparator 510 and/or display controller 514. Before going to display controller 514, the signals from image processor 508 preferably pass through comparator 510, which can compare the signals to previous stored signals in memory 512. Display controller 514 then prepares the signals for display, e.g., by scan conversion and other processing known in the art. Display controller 514 optionally also superimposes the output of comparator 510 on a conventional B-mode image, harmonic B-mode image, color Doppler image, tissue Doppler image, pulse inversion B-mode image, or pulse inversion Doppler image. The output of comparator 510 can alternatively be displayed in graphical or tabular form. System controller 520 controls beamformer controller 502, image processor 508, and display controller 514. It also controls comparator 510, allowing optional different comparisons to be calculated and passed on display controller 514. Trigger signal source 518 is responsive to physiological signals (e.g., respiratory signals, or ECG signals supplied by an ECG device, not shown) or to internal or external timers (not shown). Trigger signal source 518 preferably includes conventional hardware or software which recognizes some portion of a physiological signal (e.g., the R-wave of an ECG signal) and generates a trigger signal based thereon. Alternatively, trigger signal source 518 can be activated by an internal timer or an external timer.

[0052] Preferably, the images created and displayed by the ultrasound system are any type of image which shows contrast agent. In carrying out the invention, any type of ultrasound data can be used, alone or in combination. Specifically, any of the following ultrasound imaging modalities can be used in accordance with the embodiments taught herein, with either fundamental, harmonic, or subharmonic imaging or a combination thereof: B-mode, Color Doppler Velocity, Color Doppler Energy, Color Doppler Variance, Doppler Tissue Velocity, Doppler Tissue Energy, Doppler Tissue Acceleration, pulse inversion B-mode, pulse inversion Doppler, or any combination thereof. Alternatively, B-mode images are acquired by using alternating line technology, such as alternating line phase (ALP), as described in a co-pending U.S. patent application Ser. No. 09/282,396, entitled “Diagnostic Ultrasound Imaging Method and System with Improved Frame Rate,” filed Mar. 31, 1999, assigned to the assignee of the present invention and hereby incorporated by reference. All of these imaging modes can be carried out in a variety of ways. For example, in color Doppler modes and B-mode, data can be acquired and interleaved by line, block (i.e., a group of lines less than a frame), or frame. Of these methods, B-mode harmonic imaging, subharmonic imaging, ALP, and pulse inversion imaging are preferred because of their ability for non-linear detection of contrast agent, as well as for their ability to work at low power, where contrast agent is not destroyed.

[0053] FIG. 6 shows a timing diagram illustrating how ultrasound system 500 of FIG. 5 implements a simple sequence of two destruction pulses followed by an imaging sequence. First, transmit beamformer 504 is programmed by beamform controller 502 with parameters to transmit destruction pulses, and receive beamformer 505 is turned off. Transmit beamformer 504 then fires the destruction pulses, and resets itself. In this case, a second destruction pulse is then fired, and transmit beamformer 504 is reset. Next, beamform controller 502 programs imaging parameters for transmit beamformer 504, and receive beamformer 505 is turned on to receive and process the returning data. Transmit beamformer 504 is then fired, and receive beamformer 505 receives and processes the received data. This entire sequence is optionally controlled by system controller 520.

[0054] In this case, different parameters are used to set up the imaging and destruction pulses. For example, if ultrasonic transducer 506 has an overall bandwidth from 1.5 to 4 MHz, then beamformer controller 502 sets up ultrasound system 500 for imaging using low power harmonic B-mode imaging. In this case, ultrasound system 500 is set up for a transmit frequency of 1.75 MHz, a transmit pulse with a Gaussian envelope having a 6 dB fractional bandwidth of 33%-50%, and a transmit power corresponding to a mechanical index (MI) of 0.4. Other transmit and receive parameters, such as transmit foci, receive filters, etc., are set properly for harmonic imaging. An example for destruction pulses is ultrasound system 500 set up for a transmit frequency of 1.75 MHz, a transmit pulse with a rectangular envelope having a 6 dB fractional bandwidth of 10%, and a transmit power corresponding to a MI of 1.9. These parameters are one example of a system set up for destruction of contrast agent. The use of the MI parameter incorporates both the pressure and the frequency dependence of pulses used for contrast agent destruction.

[0055] Destruction pulses are transmitted at a pressure high enough to destroy contrast agent. The best transmit frequency to use is transducer dependent, and may also depend on the particular bubble population being destroyed. Also, in general, the longer the pulse length, the more destructive the pulse. Generally, destruction pulses have a MI greater than 0.6 to destroy contrast agent. Preferably, in order to destroy contrast agent, the MI is above 1.0. Most preferably, the MI is above 1.5.

[0056] FIG. 7 shows a timing diagram illustrating how ultrasound system 500 of FIG. 5 changes parameters between destruction pulses. First, transmit beamformer 504 is programmed by beamformer controller 502 with parameters to fire destruction pulses, and receiver 505 is turned off. Transmit beamformer 504 then fires the destruction pulse, and resets itself. Next, the transmit parameters are changed. For instance, the transmit focus is set for a new depth. Then transmit beamformer 504 fires the destruction pulse with the new transmit parameters, and resets itself. And the sequence is repeated a third time. For example, three destruction pulses are fired with three different transmit focal depths. Again, this entire sequence is optionally controlled by system controller 520.

[0057] FIG. 8 shows a flow chart 800 for one mode of operation of system controller 520 of FIG. 5. This mode of operation provides continuous imaging, with destruction pulses either triggered from a timer or a physiological signal (e.g., an ECG signal or a respiratory signal). If triggered from an ECG signal, a destruction pulse sequence is fired once every n beats. In this mode, in step 802, system controller 520 tests for a trigger from trigger signal source 518. If no trigger is detected, ultrasound system 500 images continuously: system controller 520 controls beamformer controller 502 to set up the correct imaging parameters for imaging in step 804; sets up transmit beamformer 504, multiplexer 507, and receive beamformer 506 for transmit and receive in step 806; and sets up display controller 514 to display the image in step 808. When a trigger from trigger signal source 518 is detected from step 802, system controller 520 controls beamformer controller 502 to set up the pulse parameters for destruction in step 810, and sets up transmit beamformer 504 and multiplexer 507 for transmit only in step 812. Ultrasound system 500 then returns to the imaging mode until the next trigger in step 802.

[0058] Preferably, ultrasound system 500 images by using a non-destructive imaging mode, and the destruction of contrast agent is achieved by HPRF destruction frames. Most preferably, the imaging mode uses low power harmonic imaging, subharmonic imaging, pulse inversion, or ALP, with destruction of contrast agent achieved by using HPRF destruction frames with multiple transmit foci and multiple simultaneous transmit beams. Since the destruction pulses are fired in triggered mode, it is possible to fire higher power pulses than could otherwise be fired without triggering. This is because the power transmitted over a period of time can be averaged to satisfy governmental regulatory restrictions on ultrasound transmitter power.

[0059] FIG. 9 shows a flow chart 900 for another mode of operation of the system controller 520 of FIG. 5. This mode of operation provides triggered imaging, with destruction pulses triggered using a different trigger interval, either from a timer or a physiological signal (e.g., an ECG signal or a respiratory signal). In this mode, in step 902, system controller 520 tests for a trigger from trigger signal source 518. Once a trigger is detected, system controller 520 determines in step 904 if it is a destruction trigger or imaging trigger. If it is a destruction trigger, ultrasound system 500 executes a destruction sequence in step 906 as described above. If it is not a destruction trigger, and hence an imaging trigger, it executes the imaging sequence in step 908 described above. For example, high power color harmonic imaging could be used, which potentially could destroy some contrast agent. Destruction pulses could be used to destroy the remaining contrast agent.

[0060] There are several ways the mode of operation shown in FIG. 9 could be used. For example, an image could be acquired once every heartbeat, while a destructive pulse could be fired once every 5 heartbeats. Alternatively, the triggers for the imaging and destruction pulses could have the same timing interval, while being staggered by some number of heartbeats. For example, both the destruction frame and imaging frame could be fired every 5 heartbeats, but the destruction frame could be fired 3 heartbeats after each imaging frame. In yet another alternative, if the trigger intervals are varied, then re-flow is measurable for a variety of different time intervals. The preferred methods for destruction and imaging are the same as discussed above for FIG. 8.

[0061] A fifth aspect of the invention is directed to using triggered destruction frames and imaging frames and comparing imaging frames. Here, the act of comparing imaging frames can be performed either by the operator or by the ultrasound imaging system.

[0062] When the operator performs the act of comparing imaging frames, the ultrasound imaging system assists the operator in comparing the imaging frames. In one embodiment of the invention, the ultrasound imaging system displays two or more imaging frames side-by-side for the operator to visually compare the images to assess differences. Alternatively, the ultrasound imaging system tags two or more imaging frames and stores them in some image memory system, such as a frame buffer or video buffer, for example. When the tagged imaging frames are retrieved from the image memory system, the imaging frames can be visually placed next to each other, or a switch can allow the operator to quickly alternate the display of two or more imaging frames. In another embodiment of the invention, parameters derived from two or more imaging frames are displayed, for example in a graphical format.

[0063] In a preferred embodiment of the invention, the ultrasound imaging system compares two or more imaging frames and produces a new imaging frame which is a linear or non-linear combination of the imaging frames. The linear combination of the imaging frames can be as simple as addition or subtraction of two frames, or can be a weighted sum or difference. The system can also combine the frames in a non-linear fashion, e.g., by multiplying the two frames or by applying a non-linear function to a sum or difference of the two frames. Alternatively, any of the combinations can be presented as a graphical display of one or more parameters derived from comparing the imaging frames, such as the mean difference between two frames over a region of interest.

[0064] In another embodiment, the system can subtract the complex demodulated values associated with each pixel. In this case, the subtracted image is color coded and persisted over the continuous B-mode image. Alternatively, the amplitude or energy value associated with each pixel is subtracted, color coded, and persisted over the B-mode image.

[0065] Finally, more complicated means of comparison can be used. For example, the cross-correlation between the two images at each pixel or between complex demodulated values can be calculated and displayed. In this case, a fixed threshold can be used to distinguish LOC effects. Other on-line or off-line comparisons can also be used.

[0066] Prior to any comparison, the pixels on the imaging frames may be spatially or temporally filtered. For example, prior to making a comparison, pixels may be spatially averaged to reduce the effects of noise or speckle variations. Alternatively, pixels from different imaging frames may be temporally averaged to help reduce the effects of electronic noise. Such filtering may also improve the accuracy of parameter estimates derived from imaging frame comparison techniques.

[0067] FIG. 10 shows one type of comparison 1000 that can be made using the present invention. Here, the images are video inverted, so that contrast agent is dark and the absence of contrast agent is white. The upper left contains an image I₀1002 that been stored in memory just prior to a destruction sequence. Since image I₀1002 was acquired just before destruction, it is full of contrast agent and accordingly, appears to be relatively dark in the perfused regions. Let “x₁” be the complex demodulated value associated with each pixel in image I₀1002. The areas where contrast agent exists will have correspondingly high values of “x₁.” Image I_n1004 on the upper right is the nth image obtained after destruction, and is accordingly much brighter in the unperfused region than in image I₀1002. Let “x₂” be the complex demodulated value associated with each pixel in image I_n1004. By subtracting the two complex demodulated values (“x₁”−“x₂”), a difference image 1006 can be created, shown in the lower part of FIG. 10. The value of each pixel of this difference image 1006 corresponds to the amount of contrast agent that has not returned by the time the nth image I_n1004 is obtained.

[0068] FIG. 11 shows another type of comparison 1100 that is carried out in a preferred embodiment of the present invention. Here, the images are video inverted, so that contrast agent is dark and the absence of contrast agent is white. The upper left contains an image I₁1102 that has been stored in memory immediately after a destruction sequence and accordingly, appears to be relatively bright. Since this image I₁1102 was acquired immediately after destruction, it is void or substantially void of contrast agent (i.e., at least about 70% and preferably at least 90% of the contrast agent was destroyed). Let “x₁” be the complex demodulated value associated with each pixel in image I₁1102. The image I_n1104 on the upper right is the nth image obtained after destruction, and is accordingly much darker in the perfused regions than in image I₀1102. Let “x₂” be the complex demodulated value associated with each pixel in image I_n1104. The areas where contrast agent exists will have correspondingly high values of “x₂.” By subtracting the two complex demodulated values (“x₂”−“x₁”), a difference image 1106 can be created, shown in the lower part of FIG. 11. The value of each pixel of this difference image 1106 corresponds to the amount of contrast agent that has returned by the time the nth image I_n1104 is obtained.

[0069] Furthermore, parameters related to the rate of perfusion (i.e., the volume of fluid perfusion in a unit of time) can be estimated from these comparisons. For example, if one imaging frame is fired immediately after a destruction frame, and a second imaging frame is fired some time after the destruction frame, then the difference in signal levels between the two imaging frames indicates the amount of contrast agent that has perfused into a region over that time period. If the two imaging frames are subtracted, the signal level in the resultant difference frame would be equal to the amount of contrast agent that has perfused into the region. The rate of perfusion can then be estimated by the equation given below:

[0070] R=E/T  (1)

[0071] where “R” is the linear approximation of the rate of perfusion, “E” is the amount of contrast agent that has perfused into a region over a time period, and “T” is the time period. In this case, the rate of perfusion estimate is normalized by the total amount of contrast agent that can exist in a unit volume of tissue.

[0072] If multiple measurements are made at different times after a destruction pulse, then more complex models can be used. For example, if multiple imaging frames, I₁, I₂, . . . I_n, are acquired after a destruction frame at specific times, T₁, T₂, . . . T_n, then the amount of contrast agent that has perfused into the region at each of these times can be estimated by subtracting the imaging frame immediately following the destruction frame from each of these imaging frames. The resulting difference frames, F₁, F₂, . . . F_n, can then, on a pixel-by-pixel or region-by-region basis, provide data on the amount of contrast agent that has perfused. For example, the data could be fit into an exponential model of perfusion given below:

[0073] P=A{1−exp (−Bt)}  (2)

[0074] where “P” is the estimated perfusion at time “t,” and the values of “A” and “B” are curve-fitted to the data on the amount of contrast agent that has perfused into a region in the difference frames, F₁, F₂, . . . F_n. In this way, perfusion can be measured as a function of time.

[0075] One approach for curve-fitting an exponential equation to measured values of the amount of contrast agent that has perfused into a region is described in the article “Quantification of Myocardial Blood Flow With Ultrasound-Induced Destruction of Microbubbles Administered as a Constant Venous Infusion,” by K. Wei, et al., Circulation, Vol. 97, pp. 473-483 (1998), hereby incorporated by reference. Another approach is described in co-pending U.S. patent application Ser. No. 09/184,749, entitled “Ultrasonic System and Method for Imaging Subtraction,” filed Nov. 2, 1998, assigned to the assignee of the present invention and hereby incorporated by reference.

[0076] Alternatively, one could subtract the multiple imaging frames, I₁, I₂, . . . I_n, from an imaging frame immediately prior to the destruction frame. In this case, the resulting difference frames, F₁, F₂, . . . F_n, represent the amount of contrast agent that has not yet reperfused into a region. In this case, the resulting difference frames, F₁, F₂, . . . F_n, can provide data for curve-fitting the exponential equation given below:

[0077] Q=Aexp (−Bt)  (3)

[0078] where “Q” is the amount of contrast agent that has yet to be reperfused, and again “A” and “B” are constants derived from curve-fitting the data. Other models of perfusion, and other methods of comparison to estimate the amount of contrast agent in a region, can also be used.

[0079] A sixth aspect of this invention relates to combining destruction pulses with subharmonic imaging. Subharmonic imaging may provide unique abilities to distinguish between contrast agents and surrounding tissue, because while contrast agents may generate subharmonics at conventional diagnostic ultrasound power levels, tissue will not. Imaging using subharmonics is described in a co-pending U.S. patent application Ser. No. 09/282,603, entitled “Medical Diagnostic Ultrasound Imaging System and Method with Fractional Harmonic Seed Signal,” filed on Mar. 31, 1999, which is assigned to the assignee of the present invention and hereby incorporated by reference.

[0080] Although subharmonic imaging is frequently performed with a seed signal, it can be performed with or without a seed signal. Since subharmonic signals are typically not generated by tissue, it may be advantageous to combine subharmonic signals with fundamental signals to provide a background image, as described in a co-pending U.S. patent application Ser. No. 08/838,920, entitled “Ultrasound Imaging Enhancement Methods and Systems,” filed on Apr. 11, 1997, which is assigned to the assignee of the present invention and hereby incorporated by reference. It may also be advantageous to combine subharmonic signals with harmonic signals to provide a background image, as described in a co-pending U.S. patent application, entitled “Contrast Agent Imaging with Subharmonic and Harmonic Signals in Diagnostic Medical Ultrasound,” filed on the same day as this patent application, which is assigned to the assignee of the present invention and hereby incorporated by reference.

[0081] For example, referring to FIG. 5, if ultrasonic transducer 506 has an overall bandwidth from 1.5 to 4 megahertz (MHz), and the contrast agent has a resonance frequency of about 2 MHz, then beamform controller 502 sets up ultrasound system 500 for imaging with subharmonics using the following parameters. In this case, ultrasound system 500 can transmit an ultrasound pulse centered at 4 MHz, a pulse length with a Gaussian shaped envelope which has a 6 dB fractional bandwidth of 10%, and a transmit power corresponding to a mechanical index (MI) of 1.0. In this case, the transmit frequency is set to about twice the resonance frequency, because that is where the threshold for subharmonic generation is lowest. Programmable receive beamformer 505 is set up to receive signals at the first subharmonic frequency, 2 MHz, and filter signals at the fundamental frequency, 4 MHz. This can be done by demodulating the received signal by 2 MHz, so that the subharmonic frequency resides at baseband, and applying a filter with a 6 dB bandwidth of about 500 kilohertz and about 50 dB of rejection at the fundamental frequency, so that the fundamental signal is effectively removed. The exact demodulation frequency must be modified to account for frequency downshift due to tissue attenuation. For example, in this case, an initial demodulation frequency of 2 MHz could be set, and the demodulation frequency could be decreased by 100 Hz for every 1 mm of depth. Of course, these parameters should be optimized properly for subharmonic imaging. Finally, the gain must be set high enough in programmable receive beamformer 505 to display the subharmonic signal.

[0082] Because subharmonic signals are not generated by tissue, they could potentially provide excellent contrast between contrast agent and tissue. This is very useful in detecting and quantifying the presence of contrast agent. By combining subharmonics with destruction pulses, physiological functions, such as perfusion, can be measured.

[0083] For example, the firing of destruction pulses can be triggered off an ECG signal or timer, and subharmonics could be substantially continuously imaged to measure the reflow of contrast agent into a region of interest. Alternatively, both the destruction pulses and imaging pulses can be triggered. All of the methods previously described for using destruction and imaging pulses to assess physiological function using contrast agents can be applied to subharmonics. For example, by measuring the amount of contrast agent using subharmonics after destruction, perfusion parameters can be estimated by modeling perfusion with an exponential function, as described earlier.

[0084] Various or all aspects of the preferred embodiments described above can be combined for improved contrast agent imaging. In a particularly preferred embodiment, the system images at a low power, either in continuous or triggered mode, so that contrast agent is not substantially destroyed (i.e., at least about 70% and preferably at least 90% of the contrast agent is not destroyed). At some time, one or more HPRF destruction frames are fired. This destruction frame firing is triggered from a timer or a physiological signal, such as an ECG signal or a respiratory signal. These HPRF destruction frames consist of high power, HPRF pulses designed to destroy contrast agent. The HPRF destruction pulses are swept across the imaged tissue, transmitted across several lines and with several transmit foci on each line. Additionally in a preferred embodiment, multi-beam capability exists to transmit pulses in different directions or multiple pulses in the same direction. After the HPRF destruction frames are transmitted, the system returns to imaging at low power. Images are acquired in a specific sequence and then compared to assess perfusion. Several specific imaging sequences which allow such comparisons are described below.

[0085] In a first sequence, an imaging frame is acquired, followed by an HPRF destruction frame. Then, a series of imaging frames is acquired afterwards. In the expression below, “I” stands for an imaging frame, and “D” stands for an HPRF destruction frame. An example is given below:

[0086] I₀D I₁, I₂, I₃, . . . I_n  (4)

[0087] Then each of the imaging frames I₁, I₂, I₃, . . . I_nis compared to the original frame, I₀. In this case, the reference image I₀is substantially or fully perfused with contrast agent. Comparisons of the reference image with subsequent images will indicate the amount of contrast agent that is perfusing into the region after the destruction frame. In essence, a curve can be drawn at each point in the image to indicate the amount of perfusion of contrast agent. For cardiac applications, the number of frames that are used must be kept low, however, so that there is not excessive motion between these frames and the original frame, I₀. However, for other applications, the number of imaging frames that are compared to the original frame can be quite high, lasting a number of seconds after the destruction frame.

[0088] The second sequence is similar to the first, except the process is repeated after n number of frames. An example is given below:

[0089] I₀D I₁I₂I₃. . . I_nD I_n+1I_n+2I_n+3. . .  (5)

[0090] This imaging sequence will generate a series of curves as described above to indicate the amount of perfusion of contrast agent.

[0091] A third sequence combines either of the above with gating from a physiological signal, such as an ECG signal. An example is given below:

[0092] I₀D I₁I₂I₃. . . I_nI_n+1I_n+2I_n+3. . . I_pI_p+1I_p+2I_p+3  (6)

[0093] Here, I₀, I_nand I_pare all triggered at the same point on the R-wave of an ECG signal. In this case, the images I_nand I_pwhich are compared to the original image I₀are acquired at the same point of the ECG signal. Images I₀, I_nand I_pare continuously acquired, but only images around the same point of the ECG signal are compared to image I₀. Also, only in the first set of images is a destruction frame present. Therefore, the perfusion of contrast agent is observed over a longer range of time than for methods using a destruction frame for every set of images, such as for every cardiac cycle. This process is repeated after some set number of heart beats. This sequence can also involve triggering from a timer rather than a physiological signal so that, for example, images are acquired every second and compared to the original frame.

[0094] Yet another sequence involves triggering both the imaging frames and destruction frames from an ECG signal, where the two triggers are based on a different integer number of heartbeats. For example, imaging frames are triggered and acquired once every heartbeat, while destruction frames are fired every 5 heartbeats. An example is given below:

[0095] I₀D I₁I₂I₃I₄I₅D  (7)

[0096] Here, each imaging frame is triggered from the same point of the R-wave.

[0097] Another set of sequences is similar to those discussed above, but instead of comparing a frame with the frame immediately preceding the destruction frame, a frame is compared to the frame immediately following the destruction frame. For example, analogous to the first sequence described, frames I₂, I₃, I₄, I₅are compared to I₁instead of I₀. In this case, the reference frame is a frame devoid or substantially devoid of contrast agent rather than a frame fully or substantially perfused with contrast agent. Again, the re-flow of contrast agent over time is observed. This same approach for comparing the first image following the destruction frame to other images can be applied to the other imaging sequences as well.

[0098] An alternative sequence involves imaging at high power immediately before and after HPRF destruction frames. In this case, the system images continuously at a low power so that contrast agent is not substantially destroyed. At some time triggered by a time trigger or some physiological signal, such as an ECG signal or a respiratory signal, a high power imaging frame is transmitted. Then, one or more HPRF destruction frames are fired, followed by one or more high power imaging frames. The system then returns to low power imaging until the next trigger. The advantage of this over the first sequence is that the higher the transmit power of the imaging pulses used to obtain an imaging frame, the more clearly the contrast agent is seen, for some contrast agents.

[0099] Another embodiment of the invention involves the use of destruction pulses on individual ultrasound lines instead of destruction frames. In this approach, an imaging pulse is fired along an ultrasound image line. Then a series of HPRF destruction pulses are sent along the same line. As discussed above, these destruction pulses are sent very rapidly, since the system does not have to wait for them to return and be processed. These HPRF destruction pulses are fired for a variety of different focal depths to ensure contrast agent destruction along the whole line. Then a second imaging pulse is sent along the same line, and the received signal is compared to the received signal from the first pulse.

[0100] Another embodiment involves the use of destruction blocks. A block is a set of lines. A destruction block is a set of destruction pulses fired into a block. For example, a set of imaging pulses, a set of HPRF destruction pulses and a second set of imaging pulses are fired into the block for comparing the signals received from the imaging pulses before and after the destruction of the contrast agent in the block.

[0101] Whether destruction pulses, blocks, or frames are used, there must be a pause between the HPRF destruction pulses and the imaging pulses to remove artifacts from the destruction pulses. The artifacts from the destruction pulses are either caused by reverberations or by not waiting for the pulses to die down. Furthermore, if Doppler processing is used, pulses must be fired in order to set the reverberation environment correctly. For destruction frames, such pauses and imaging pulse firings cause a negligible reduction of the system frame rate, while for destruction blocks, they cause a larger reduction in the system frame rate. However, these pauses and imaging pulse firings cause a greatly reduced system frame rate when destruction pulses are fired only along one line at a time.

[0102] An alternative embodiment of the invention provides a user-interface, preferably a graphical user interface (GUI), where a doctor or sonographer presses a button on a screen or keyboard, clicks on a mouse, trackball or touchpad, or otherwise selects an option on the user-interface, which will instruct the ultrasound system to generate and transmit HPRF destruction pulses. There may be occasions when a doctor wants to override a programmed sequence of ultrasound pulse firings and individually control the firing of destruction pulses. That is, rather than using a programmed sequence of HPRF destruction pulses, the doctor independently activates the firing of HPRF destruction pulses from the GUI.

[0103] Many of the embodiments of the invention described herein do not require HPRF destruction pulses. In an alternative embodiment, the system images continuously without substantially destroying contrast agent, and at a time triggered by a physiological signal (e.g., an ECG signal or a respiratory signal) or a timer, a series of destruction pulses is fired to substantially destroy contrast agent. The system then returns to normal imaging. This is the approach described in FIG. 8, but without HPRF destruction pulses. Similarly, as shown in FIG. 9, a system can both image and transmit destruction pulses in triggered mode, where the two triggers are invoked at different times, keyed to fixed points of a physiological signal or a timer signal, without using HPRF destruction pulses.

[0104] Another embodiment not requiring HPRF pulses involves firing a sequence of imaging and destruction pulses such that a first image is acquired, a destruction frame is fired, and a second image is acquired for comparison to the first image, wherein the entire sequence is triggered from a source, such as an ECG signal. In this case, image brightness is used to gauge the amount of contrast agent that has perfused over that trigger interval. Such comparisons can be made using any of the imaging and comparison approaches described earlier.

[0105] Of course, all of these approaches are improved by using HPRF signals. Another approach to improve destruction of contrast agent involves using chirp signals, which can increase the time duration of a pulse without reducing the bandwidth of the pulse. Using HPRF chirp signals may further enhance destruction of contrast agent. In particular, this invention can be combined with the use of chirp signals or coded excitation as described in co-pending U.S. patent application Ser. No. 09/282,510, entitled “Medical Ultrasound Diagnostic Imaging Method and System with Nonlinear Phase Modulation Pulse Compression,” filed Mar. 31, 1999, and/or co-pending U.S. patent application Ser. No. 09/283,346, entitled “Medical Ultrasound Diagnostic Imaging System

[0106] Using Coded Transmission Pulses”, filed Mar. 31, 1999. Both patent applications are assigned to the assignee of the present invention and are incorporated by reference.

[0107] The exemplary embodiments described herein are for purposes of illustration and are not intended to be limiting. For example, while an ECG signal is the most frequently discussed physiologic signal source for triggering destruction or imaging pulses, other cardiac signals or respiratory signals can also be used for triggering destruction or imaging pulses. Moreover, the sequence of steps for firing imaging pulses and destruction pulses could be altered from what is disclosed and still achieve the intended benefits of the invention. Still further, one of skill in the art will appreciate that the system block of FIG. 5 and the flowcharts of FIGS. 8 and 9 are intended to illustrative and not limiting. Specific details of implementation may be varied without departing from the scope of the invention. For example, the system, beamformer and display controller can be optionally integrated into a single controller, or, in appropriate alternative architectures, dispensed with entirely. Likewise, transmit and receive beamformers can be optionally integrated; and while it is preferred that they be programmable, this is not necessary in order to carry out the invention. Likewise, the memory and comparator are optional in some embodiments of the invention. Finally, while destruction pulses are defined herein as not being used for imaging or display, it will be apparent to one of ordinary skill in the art that displaying such pulses on an auxiliary display or even in an irrelevant manner on the imaging display itself (e.g., in the first centimeter or so of the extended field of view) will be within the scope of this invention. Therefore, those skilled in the art will recognize that other embodiments could be practiced without departing from the scope and spirit of the claims set forth below.