Method and system for determining a therapeutic ultrasound amplitude

Abstract

A method (100) for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject. The method (101) comprises delivering ultrasound energy at a selected amplitude to in a subject's brain to cause cavitation of microbubbles in the subject's brain (102). An acoustic emission from the brain generated responsive to the delivering of the ultrasound energy is detected (104). The acoustic emission is analysed to obtain response intensity data comprising a determined intensity of the acoustic emission (106). A parameter of the response intensity data at the selected amplitude is determined (108) and compared to a target parameter range (110). Based on the comparison, when the determined parameter is outside of the target parameter range: the selected amplitude is modified (114); and the preceding method steps are repeated with the modified selected amplitude. When the determined parameter is within the target parameter range (118), the current selected amplitude is identified as a therapeutic amplitude (118).

Description

[0001] "Method and system for determining a therapeutic ultrasound amplitude"

[0002] Field

[0003] [1] The present disclosure relates to a method and system for delivering ultrasound energy to a subject, and particularly to a method and system for identifying a therapeutic amplitude for delivery of ultrasound energy to the subject.

[0004] Background

[0005] [2] Low-intensity ultrasound used in combination with intravenously injected microbubbles is an emerging modality for transiently and safely opening the blood-brain barrier (BBB) for therapeutic applications. Disruption of the BBB may be used to deliver molecules from blood vessels into the brain tissue, when the molecules are of a size that would otherwise be unable to pass through the BBB.

[0006] [3] Acoustic cavitation refers to the dynamic interaction between ultrasound and microbubbles. In an ultrasound field, a microbubble can expand, oscillate, and collapse, producing mechanical effects to blood vessels in the vicinity of the microbubble. These are recognised as factors contributing to BBB opening, although the exact underlying mechanisms are yet to be understood.

[0007] [4] Ultrasound pressure, more specifically the peak negative pressure (PNP), is the dominant parameter that determines the dynamics of microbubbles and the success of BBB opening. Safe BBB opening can be achieved over a therapeutic pressure range (or window). Below this window, the BBB remains closed, whereas above the window, the BBB will open, but damage such as bleeds and oedema can occur. The selected ultrasound amplitude must therefore not be too low (such that no bio-effect is achieved) nor too high (which may result in tissue damage).

[0008] [5] The skull of a human (or other large animal) may present variable and difficult to predict attenuation of ultrasound energy, as the skull is an inhomogeneous medium. For example, ultrasound attenuation can be affected by the skull’s properties such as density, thickness, and the composition of cortical and trabecular layers. Moreover, skull curvature and incident angle of the ultrasound beam can affect the extent of attenuation. Hence, the ultrasound input amplitude needs to be carefully determined for any given position of the ultrasound transducer relative to the skull, such that the resultant PNP at the targeted spot within the brain achieves safe and effective BBB opening. A major challenge for therapeutic applications in humans is therefore in selecting an appropriate ultrasound amplitude.

[0009] [6] Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present disclosure as it existed before the priority date of each of the appended claims.

[0010] Summary

[0011] [7] According to one aspect of the present disclosure, there is provided a method for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject, the method comprising: a) delivering ultrasound energy at a selected amplitude to a subject’s brain to cause cavitation of microbubbles in the subject’s brain; b) detecting an acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; c) analysing the acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; d) determining a parameter of the response intensity data at the selected amplitude; e) comparing the determined parameter to a target parameter range; f) based on the comparison: i) when the determined parameter is outside of the target parameter range: modifying the selected amplitude; and repeating steps a) to f) with the modified selected amplitude; and ii) when the determined parameter is within the target parameter range, identifying the current selected amplitude as a therapeutic amplitude.

[0012] [8] The ultrasound energy may be delivered to a target region in the subject’s brain.

[0013] [9] The ultrasound energy in step (a) may be delivered using an ultrasonic generation system. The ultrasonic generation system may include a transducer configured to deliver the ultrasound energy to the subject’s brain to cause cavitation of the microbubbles in the subject’s brain.

[0014]

[0010] In some examples, the ultrasound energy in step a) may be delivered in a sequence of short pulses. The pulses may be separated by a predetermined interval. The predetermined interval may be about 1 ms, for example. The predetermined interval may be about 250 ps, about 500 ps, about 750 ms, about 1 ms, about 2 ms, about 3 ms about 4 ms, about 5 ms or longer. In some examples, the further ultrasound energy (therapeutic ultrasound delivery) may be delivered using a longer pulse length than the delivery of ultrasound energy in step a).

[0015]

[0011] In some examples, the ultrasound energy in step a) may be delivered in a single continuous pulse.

[0012] Detecting an acoustic emission in step (b) may be performed using one or more sensors configured to detect the acoustic emission. For example, the sensor may be a passive cavitation detector (PCD). The method may comprise delivering further ultrasound energy to the subject’s brain at the identified therapeutic amplitude for a selected duration and / or to deliver a preselected dose of ultrasound energy. The further ultrasound energy may be delivered for a therapeutic purpose and may be considered therapeutic ultrasound delivery.

[0016]

[0013] Analysing the acoustic emission signal may comprise determining an intensity of a broadband emission component of the acoustic emission. The response intensity data may comprise the determined intensity of the broadband emission component of the acoustic emission.

[0017]

[0014] Analysing the acoustic emission signal may comprise filtering the acoustic emission with one or more band-pass filters. Analysing the acoustic emission signal may comprise determining an intensity of a filtered component of the acoustic emission. The response intensity data may comprise the determined intensity of the filtered component of the acoustic emission. The band-pass filter may be configured to filter the acoustic emission in the time domain and / or in the frequency domain. In some examples, the band-pass filter may be configured relative to a fundamental frequency ( / o) of the ultrasound energy. For example, the band-pass filter may be configured to filter the acoustic emission to a frequency range such as between about fo and about 2 / o, between about Ifo and about 3 fo, between about 3 / o and about 4 / o, between about 4 / o and about 5fo, between about fo and about 4 / o, between about fo to about IQ / o, between about Ifo to about IQ / o, between about 3 / o to about IQ / o, between about 4 / o to about 1 Q / o, between about Ifo to about 8 / o, or between about 4 / o to about 8 / o.

[0018]

[0015] Analysing the acoustic emission may comprise comparing the detected acoustic emission to a baseline acoustic emission. For example, analysing the acoustic emission may comprise comparing an intensity of the detected acoustic emission to an intensity of the baseline acoustic emission. The baseline acoustic emission may be an acoustic emission generated responsive to delivering ultrasound energy to the subject’s brain in the absence of microbubbles. The method may comprise delivering ultrasound energy to the subject’s brain in the absence of microbubbles and detecting a baseline acoustic emission generated responsive to the delivering of ultrasound energy in the absence of microbubbles. The method may comprise determining an intensity of the baseline acoustic emission. The method may comprise determining an intensity of a filtered component of the baseline acoustic emission. The method may comprise determining an intensity of a broadband component of the baseline acoustic emission.

[0016] Analysing the acoustic emission may comprise correlating the acoustic emission with depth based on arrival time of the acoustic emission. The method may comprise extracting a portion of the acoustic emission corresponding to a target depth and / or a target depth range. In some examples, the method may comprise determining a distance of the subject’s skull from a source of the ultrasound energy. The method may comprise determining a distance of a target region of the subject’s brain from a source of the ultrasound energy based on the determined distance of the subject’s skull from the source of the ultrasound energy and a predetermined distance from the subject’s skull to the target region.

[0019]

[0017] Determining a parameter of the response intensity data in step (d) may comprise determining an absolute or average intensity value of the data at the selected amplitude. In such examples, the target parameter range may comprise a target intensity range.

[0020]

[0018] In some examples, determining a parameter of the response intensity data may comprise: fitting a curve to the response intensity data; and determining a slope of the curve at the selected amplitude. In such examples, the target parameter range may comprise a target slope range.

[0021]

[0019] Comparing the determined parameter to the target parameter range in step (e) may comprise determining whether the determined parameter is within the target parameter range or outside of the target parameter range. In some examples, the target parameter range may be determined as above or below a predefined target parameter value. In some examples, the target parameter range may be defined as between a predefined minimum value and a predefined maximum value.

[0022]

[0020] In some examples, step f) may comprise: when the determined parameter is within the target range: incrementing a count; when the count has a value less than a count threshold, repeating steps a) to f) at the current value of the selected amplitude; and when the count reaches a value equal to the count threshold, identifying the current value of the selected amplitude as the therapeutic amplitude value.

[0023]

[0021] In some examples, the method may comprise delivering the microbubbles to the subject.

[0024]

[0022] According to another aspect of the present disclosure, there is provided a system for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject, according to the disclosed method.

[0023] In one aspect there, is provided a system for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject, comprising: an ultrasonic generation system including a transducer configured to deliver ultrasound energy to the subject’s brain to cause cavitation of microbubbles in the subject’s brain; a sensor configured to detect an acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; a controller including a processor in communication with the transducer and the sensor, the controller configured to: cause the transducer to deliver ultrasound energy at a selected amplitude; receive, from the sensor a detected acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; analyse the detected acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; determine a parameter of the response intensity data at the selected amplitude; compare the determined parameter to a target parameter range and, based on the comparison: i) when the determined parameter is outside of the target parameter range: modify the selected amplitude; and cause the transducer to apply ultrasonic energy at the modified selected amplitude; and ii) when the determined parameter within the target parameter range, identify the current selected amplitude as a therapeutic amplitude.

[0025]

[0024] The controller may be further configured to cause the ultrasonic transducer to deliver further ultrasound energy to the subject’s brain at the identified therapeutic amplitude for a predetermined duration and / or to deliver a preselected dose of ultrasound energy.

[0026]

[0025] According to one aspect, there is provided a method for identifying a therapeutic amplitude for use with an ultrasound transducer, the method comprising: receiving an acoustic emission detected from a brain, the acoustic emission generated responsive to delivering of ultrasound energy at a selected amplitude to cause cavitation of microbubbles in the brain; analysing the acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; determining a parameter of the response intensity data at the selected amplitude; comparing the determined parameter to a target parameter range and, based on the comparison: i) when the determined parameter is outside of the target parameter range: modifying the selected amplitude; and outputting a signal configured to cause a transducer to deliver ultrasound energy at the modified selected amplitude; and ii) when the determined parameter is within the target parameter range, identifying the current selected amplitude as a therapeutic amplitude.

[0027]

[0026] According to one aspect of the present disclosure, there is provided a non-transitory computer-readable medium configured to perform the method of the present disclosure.

[0028]

[0027] Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

[0029] Brief Description of Drawings

[0030]

[0028] By way of example only, embodiments are now described with reference to the accompanying drawings, in which:

[0031]

[0029] Figure 1 shows a flow diagram of a method for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject according to an embodiment of the present disclosure;

[0032]

[0030] Figure 2 shows, schematically, a system for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject according to an embodiment of the present disclosure;

[0033]

[0031] Figure 3 shows a 3D ultrasound beam profiled in free-field in panel (a), a 3D ultrasound beam profiled with a sheep skull in panel (b), and measurement of peak negative pressure (PNP) of ultrasound as a function of the driving amplitude of the ultrasound waveform in panel (c);

[0034]

[0032] Figure 4 shows representative frequency spectra for short pulse sonication in panel (a) for long pulse sonication in panel (b), and baseline for long pulse sonication in panel (c);

[0035]

[0033] Figure 5, shows, schematically, an in vitro experimental configuration to measure acoustic emission through an excised sheep skull as a function of the driving amplitude;

[0036]

[0034] Figure 6 shows an example intensity response curve with an indication of response onset;

[0037]

[0035] Figure 7 shows intensity of a broadband component of an acoustic emission measured through excised sheep skulls as a function of driving amplitude in panel (a) and as a function of peak negative pressure (PNP) in panel (b);

[0036] Figure 8 shows a flow diagram of a method for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject according to an embodiment of the present disclosure, using long pulse sonication and determination of broadband emission intensity as an input to a feedback controller;

[0038]

[0037] Figure 9 shows graphs of ultrasound driving amplitude and corresponding intensity of a broadband component of detected acoustic emissions from sonication of a sheep brain at three location ins panels (a), (b) and (c), respectively;

[0039]

[0038] Figure 10 shows an MR gadolinium signal enhancement image for a coronal section of a sheep brain in panel (a), graphs of driving ultrasound amplitudes (squares) and corresponding intensities of detected broadband (BB) component of acoustic emissions (circles) in panel (b), and BB emission analysis at low (circle) and high frequencies (square) in panel (c);

[0040]

[0039] Figure I la shows a time waveform of a burst composed of 20 pulses of 5 cycle pulses separated by 1 ms

[0041]

[0040] Figure 1 lb shows a PCD detected acoustic emission in response to the 1stpulse of Figure 1 la along depth obtained with varying amplitude;

[0042]

[0041] Figure 11c shows a PCD detected acoustic emission in response to 20 pulses at the selected depth, compared to baseline;

[0043]

[0042] Figure 1 Id shows the averaged acoustic emission intensities of Figure 11c at the selected depth plotted as a function of amplitude;

[0044]

[0043] Figure 12 shows a flow diagram of a method for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject according to another embodiment of the present disclosure, using short pulse sonication and determination of broadband emission intensity as an input to a feedback controller;

[0045]

[0044] Figure 13 shows a baseline detected acoustic emission response in panel (a), a detected acoustic emission response in the presence of microbubbles in panel (b), a maximal projection of depth axis revealing two major cavitation sources at 40 and 60 mm in panel (c), and a maximal detected acoustic emission response in the selected depth in comparison to TCL in panel (d);

[0046]

[0045] Figure 14 shows maximum detected acoustic emission responses at a selected depth range for four sheep skulls;

[0047]

[0046] Figure 15 shows sonication planning in a sheep brain with cranioplasty, including a test configuration including three different pressure levels applied across six discrete spots in panel (a), a coronal view of the planned six spots relative to skull on navigation system in panel (b), and a sagittal view showing three spots in the left hemisphere, assigned perpendicular to the implant in panel (c);

[0048]

[0047] Figure 16 shows a representative detected acoustic emission response using short pulse sonication from spot 6 of Figure 15, including baseline acoustic emission response in panel (a), detected acoustic emission response with microbubbles in panel (b), and maximum projection on depth axis in panel (c);

[0049]

[0048] Figure 17a shows a fluorescence image of leakage of Evans Blue indicating blood brain barrier;

[0050]

[0049] Figure 17b and Figure 17c show acoustic emission responses detected using short pulses;

[0051]

[0050] Figure 17d and Figure 17e broadband (BB) and higher harmonic (HH) emissions using long pulses in comparison to baseline emissions;

[0052]

[0051] Figure 18a shows a detected acoustic emission in response to short pulse sonication with a bandpass filtered waveform overlaid;

[0053]

[0052] Figure 18b shows an enlarged portion of the acoustic emission of Figure 18a correlating to the depth at which the response was detected;

[0054]

[0053] Figure 18c shows frequency spectra illustrating the filtering effect on the acoustic emission of Figure 18a; and

[0055]

[0054] Figure 18d shows a magnified intensity of a portion of the frequency spectra of Figure 18c, showing the difference in measured intensity using filtering in the time domain and frequency domain.

[0056] Detailed Description

[0057]

[0055] The present disclosure provides methods of identifying safe and appropriate amplitudes for delivery of ultrasound energy to the brain of a subject, for safe and effective BBB opening in said subject.

[0058]

[0056] One way to identify a therapeutic amplitude for delivery of ultrasound energy into a subject’s brain through an attenuating skull, is to exploit the ways in which ultrasound affects microbubbles in the brain. When microbubbles are delivered to the brain, and these microbubbles are exposed to ultrasound energy, the microbubbles interact with ultrasound and can emit a secondary ultrasound waveform called acoustic emission (or microbubble emission). The acoustic emissions occur as a result of the microbubbles cavitating. The detection of these acoustic emissions may be referred to as passive cavitation detection, with the detection device (sensor) being called a passive cavitation detector (PCD).

[0057] As used herein, “therapeutic amplitude”, and the like, refers to a parameter of ultrasound energy delivered to the brain of subject that is safe (i.e., does not cause tissue damage) and effective (i.e., is sufficient to cause microbubble cavitation of an intensity suitable for increasing the permeability of the BBB in the brain of said subject, when combined with / accompanied by the delivery of microbubbles to said subject).

[0059]

[0058] It would be understood by the skilled addressee that the disclosed methods require the delivery of microbubbles to the brain either prior to, concurrently or sequentially with the delivery of the ultrasound energy. Methods for the safe delivery of microbubbles to the brain are well established in the art and are encompassed herein without limitation.

[0060]

[0059] Figure 1 shows an example method 100 of identifying a therapeutic amplitude for delivery of ultrasound energy delivery according to the present disclosure. The method 100 comprises delivering ultrasound energy 102 at a selected amplitude to a subject’s brain to cause cavitation of microbubbles in the subject’s brain (i.e., delivering ultrasound energy at an amplitude high enough to generate a PNP sufficient to cause cavitation of the microbubbles) and detecting an acoustic emission 104 from the brain generated responsive to the delivering of the ultrasound energy. At 106, the acoustic emission is analysed to obtain response intensity data comprising a determined intensity of the acoustic emission. At 108, a parameter of the response intensity data is then determined at the selected amplitude and the parameter then compared to a target parameter range at 110. Based on the comparison, as indicated at 112, when the determined parameter is outside of the target parameter range, the selected amplitude is modified at 114 and the preceding steps are repeated (i.e. in a feedback loop) with the modified selected amplitude. Alternatively, as indicated at 116, when the determined parameter is within the target parameter range, the current selected amplitude is identified as a therapeutic amplitude, as indicated at 118.

[0061]

[0060] In the present disclosure, unless otherwise specified, selecting an amplitude of the ultrasound energy (i.e., the “selected amplitude” of 102) refers to selecting an input or driving amplitude (i.e. the output of a waveform generator). The selected input amplitude determines the amplitude of the output ultrasound waveform. However, the peak negative pressure (PNP) of the delivered ultrasound achieved at a target location in the brain may be affected by attenuation of the ultrasound energy (e.g. from the skull or other tissue). It will be readily understood that the driving amplitude of the ultrasound waveform may be controlled based on an input voltage amplitude to the waveform generator.

[0062]

[0061] Where step 102 is being performed for the first time when performing the disclosed method, the selected ultrasound parameters, including the initial selected amplitude, may be established during pre-treatment planning. In some examples, the method may also include defining a minimum and / or maximum value for the selected amplitude. It would be understood that the selected amplitude of the ultrasound energy delivered in step 102 in at least one cycle of the method of 100 would need to be sufficient to cause microbubble cavitation that could result in detectable acoustic emission.

[0063]

[0062] The ultrasound energy of the selected amplitude may be delivered by any suitable means, such as a transducer. The minimum and / or maximum value for the selected amplitude may be affected by a number of factors, such as, the type of transducer, the efficiency of the transducer, the performance of an amplifier in association with the transducer or others. Notwithstanding the acceptable levels of variations that may stem from such factors, the selected amplitude (i.e. the selected output of a waveform generator in association with the transducer) in any given iteration of step 102 may be in the range of about 1 mVpp to about 2 Vpp.

[0064]

[0063] In some examples, the minimum value for the selected amplitude for the ultrasound energy delivered in step 102 may be close to zero. For example, the minimum value for the selected amplitude for the ultrasound energy delivered in step 102 may be 1 mVpp. In other examples, the minimum value may be higher.

[0065]

[0064] In some examples, the maximum value for the selected amplitude for the ultrasound energy delivered in step 102 may be up to about 2 mVpp. For example, the maximum value for the selected amplitude for the ultrasound energy delivered in step 102 may be about 0.5 mVpp, about 1 mVpp, about 1 .5 mVpp, or about 2 mVpp.

[0066]

[0065] In some examples, the delivery of the ultrasound energy at the selected amplitude in step 102 may be by at least one short pulse, such as in a sequence of short pulses. This may be referred to as “short pulse sonication”. For example, each pulse may contain a predetermined number of cycles of the ultrasound waveform. Each pulse may contain about 2, 3, 4, 5, 6, 7, 8, 9 or 10 cycles, for example. In general, it has been found that reducing pulse length improves spatial resolution in depth. The pulses may be delivered in a burst, comprising a selected number of pulses. The number of pulses delivered in a burst may be about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 pulses, or more. The interval between pulses may be between about 100 ps and about 10 ms, such as between about 200 ps and about 5 ms, between about 300 ps and about 3ms, between about 500 ps and about 2 ms, or about 1 ms. In one example, the ultrasound energy is delivered in 20 pulses of 5 cycles, where each pulse is separated by 1 ms interval. In some examples, the pulses may each have the same amplitude. In other examples, a variable amplitude can be applied, (i.e., the 20 delivered pulses may have different amplitude values).

[0067]

[0066] In other examples, additionally or alternatively, the ultrasound energy in step 102 may be delivered as a single, substantially continuous pulse. This may be referred to as “long pulse sonication”. The pulse length may be about 10 ms, about 15 ms, about 20 ms, about 25 ms, about 30 ms or longer. The pulse length may be selected in combination with the initial selected amplitude. In general, a shorter pulse length requires higher pressure (PNP) to generate BBB opening and vice versa.

[0068]

[0067] Referring to step 104 of method 100 in Figure 1, an acoustic emission from the brain is detected. The acoustic emission is generated responsive to the delivering of the ultrasound energy to the brain. The acoustic emission may be detected using one or more sensors. For example, the signal may be detected using a passive cavitation detector (PCD).

[0069]

[0068] Referring to step 106 of method 100 in Figure 1, the detected acoustic emission is then analysed to obtain response intensity data, which comprises a determined intensity of the acoustic emission. The present disclosure refers to the “intensity” of the acoustic emission. However, it will be appreciated that the terms “intensity”, “power”, “strength” or “magnitude” may be used interchangeably as the context allows. For example, where the detected signal in step 104 is an electric signal indicative of the acoustic emission and not converted to an acoustic signal, determining the intensity of the acoustic emission may comprise determining a power of the electrical signal.

[0070]

[0069] Where step 106 is performed for the first time, the response intensity data may comprise a single acoustic emission intensity datapoint. However, where step 106 has been performed multiple times (for example, by following the feedback loop from step 114), the response intensity data may comprise a plurality of determined intensity datapoints. For example, the number of datapoints may correspond to a number of modified selected amplitudes for each repetition of the feedback loop. The response intensity data may be considered a bivariate dataset, in which the value of the acoustic response intensity is paired with a respective value of the selected amplitude. In some examples, steps 102, 104 and 106 may be performed multiple times to obtain a plurality of datapoints of the response intensity data (at the same selected amplitude or at different selected amplitudes), prior to proceeding to step 108.

[0071]

[0070] Analysing the acoustic emission signal 106 may comprise filtering the acoustic emission with one or more band-pass filters. The acoustic emission may be filtered in the time domain and / or in the frequency domain. It will be understood that, in signal processing, either time or frequency domains may be used. For time domain fdtering, a bandpass fdter or a series of bandpass fdters can be designed to extract either broadband (BB) or shockwave (SW) emission (as further explained below). Filter coefficients (or impulse responses) can be obtained using filter designing algorithms and filtering operation is performed using a convolution process. In frequency domain filtering, a Fourier transform may first applied to an input signal, then the spectral components of interest can be selectively extracted by simple masking on the frequency axis. In the end, however, it will be understood that the same outcome may be achieved by either time or frequency domain filtering.

[0072]

[0071] In some examples, analysing the acoustic emission signal 106 may comprise determining an intensity of a broadband (BB) emission component of the acoustic emission. A BB emission component may be defined a spectral component between nfO harmonics (where n is an integer), optionally excluding ultra-harmonics (n / mfO where n & m are integers but not equal).

[0073]

[0072] In some examples, the band-pass filter is configured relative to a fundamental frequency ( / o) of the ultrasound energy, where fo is the centre frequency of the ultrasound wave. This filtered portion of the acoustic emission is defined herein as “shockwave” (SW) emission.

[0074]

[0073] In general, because of skull attenuation increasing in proportional to frequency, components of the acoustic emission beyond 5 MHz may not be well detected. BB emission generally shows stronger power at higher frequency than BB emission near the driving frequency (fo). Therefore, in the method according to the present disclosure, superior sensitivity is expected in detecting BB emission given that a detection frequency higher than fo is selected. Detecting lower frequencies (e.g. / o-2 fo, 2 / o-3 fo, 3fo-4fo) may be preferrable when detecting a BB component of the acoustic emission. However, detection of SW emission at lower frequencies may be interfered with by harmonic component generated by non-linear distortion during propagation. As such, it may be beneficial to include higher frequencies when detecting SW emissions. In some examples, the band-pass filter may be configured to filter the acoustic emission to a frequency range of between about fo and about 1 Q / o, such as between about 4fo and about 8 / o.

[0075]

[0074] In some examples, analysing the acoustic emission 106 comprises comparing the detected acoustic emission to a baseline acoustic emission. The baseline acoustic emission may be an acoustic emission from the brain generated in response to delivering ultrasound energy to the subject’s brain in the absence of microbubbles. In such examples, the baseline acoustic emission may be determined prior to commencing the method 100. In other examples, the method 100 may include an additional step of determining the baseline acoustic emission. Determining the baseline emission may be performed after coupling an ultrasound transducer to the subject’s skull, prior to delivery of the microbubbles to the subject’s brain.

[0076]

[0075] In some examples, analysing the acoustic emission 106 may comprise correlating the acoustic emission with depth based on an arrival time of the acoustic emission. Analysing the acoustic emission 106 may further comprise extracting (e.g. by filtering) a portion of the acoustic emission corresponding to a target depth and / or a target depth range. In some examples, analysing the acoustic emission may comprise determining a distance of the subject’s skull from a source of the ultrasound energy (such as a transducer). The method 100 may comprise determining a distance of a target region of the subject’s brain from a source of the ultrasound energy based on the determined distance of the subject’s skull from the source of the ultrasound energy and a predetermined distance from the subject’s skull to the target region. This may enable isolating a portion of the detected acoustic emission coming from brain tissue, for example, and may enable exclusion of artefacts generated from the skull or other such structures. Additionally, or alternatively, this may enable targeting of a specific brain region at a predetermined depth or depth range, such as shallow or deep (relative to the skull).

[0077]

[0076] In some examples, the thickness of the skull at the target region and / or the depth of the target region relative to the surface of the skull may be known or may be estimated. In some examples, the method may include an additional step of determining the thickness of the skull and / or the depth of the target region relative to the surface of the skull. For example, the method may include performing pre-operative imaging, such as computed tomography (CT) and / or magnetic resonance imaging (MRI) to determine the thickness of the skull and / or the depth or depth range of the target region in the brain.

[0078]

[0077] In some examples, a known depth of a brain structure from the skull may be used to confirm depth targeting. For example, the posterior cingulate cortex (PCC) is known to be about 60 mm deep relative to the skull. Where the PCC is the target region, the known distance from the skull to the PCC may be used to determine whether the detected acoustic emission is originating from the PCC. Alternatively, or additionally, the known distance from the skull to the PCC may enable extracting (e.g. by filtering) a portion of the detected acoustic emission coming from the PCC. Although this example specifies the PCC as the target region, the same principle may be applied to any brain region with a known or determined depth from the skull, to confirm that that a detected acoustic emission is originating from the target region, rather than other areas / structures.

[0079]

[0078] Determining a parameter of the response intensity data 108 may comprise determining an intensity value of the data at the selected amplitude. In various examples, the parameter may be a value associated with an absolute value of the response intensity at the selected amplitude and / or a rate of change of the response intensity at the selected amplitude.

[0080]

[0079] For example, the intensity value may be an absolute value of a single datapoint for a given selected amplitude value. Alternatively, the intensity value may be an average value of multiple datapoints associated with the selected amplitude value. For example, the target parameter value may be determined as an absolute intensity of a BB emission component, or of a SW emission component, of the detected acoustic emission.

[0081]

[0080] In some examples, the parameter may comprise a slope of the response intensity data for the selected amplitude value. For example, determining a parameter of the response intensity data 108 may comprise fitting a curve to the response intensity data and determining a slope of the curve at the selected amplitude. In some examples, the target parameter value may be determined as the slope of a curve fitted to intensity data comprising the BB emission component or SW emission component of the detected acoustic emission.

[0082]

[0081] At step 110, the determined parameter is compared to the target parameter range. The target parameter range may be determined as above or below a predefined target parameter value, or as within a predefined minimum value and a predefined maximum value. The target parameter range and / or value may be determined experimentally prior to commencement of the method. For example, the target parameter range and / or value may be determined as indicative of a desired intensity of microbubble cavitation required for safely yet effectively opening the BBB. Further, the target parameter range and / or value may be determined with reference to a determined baseline acoustic emission intensity.

[0083]

[0082] Where the determined parameter is a value of the intensity data, such as an absolute value or an average value, the target parameter range may comprise a target range for the intensity value. For example, the target range may be defined as greater than or equal to a target intensity value, or within a predetermined tolerance of the target intensity value. Where the determined parameter is a slope of a curve fitted to the intensity data, the target parameter range may comprise a target range for the slope. For example, the target range may be defined as greater than or equal to a target slope, or within a predetermined tolerance of the target intensity slope.

[0084]

[0083] Where analysing the acoustic emission signal comprises determining an intensity of a BB emission component of the acoustic emission, in some examples, the target parameter may comprise a BB response intensity of between about 0.5 dB to about 20 dB above the determined baseline BB emission intensity, such as about 0.50 dB, about 1.0 dB, about 2.0 dB, about

[0085] 3.0 dB, about 5.0 dB, about 10.0 dB, about 15.0 dB or about 20.0 dB above the determined baseline BB emission intensity. In some examples, the target parameter may comprise a slope of a curve fitted to the BB response intensity data of between about 0.5 dB / mVpp to about

[0086] 6 dB / mVpp, such as about 0.5 dB / mVpp, about 1 dB / mVpp, about 2 dB / mVpp, about

[0087] 3 dB / mVpp about 4 dB / mVpp, about 5 dB / mVpp, or about 6dB / mVpp.

[0088]

[0084] Where analysing the acoustic emission signal comprises determining an intensity of a SW emission component of the acoustic emission, the target parameter may comprise a SW response intensity between about 0.5 dB to about 20 dB above the determined baseline SW emission intensity, such as about 0.50 dB, about 1.0 dB, about 2.0 dB, about 3.0 dB, about 5.0 dB, about 10.0 dB, about 15.0 dB or about 20.0 dB above the determined baseline SW emission intensity. In some examples, the target parameter may comprise a slope of a curve fitted to the SW response intensity data of about 0.05 dB / mVpp to about 2 dB / mVpp, such as about 0.05 dB / mVpp, about 0.10 dB / mVpp, about 0.20 dB / mVpp, about 0.30 dB / mVpp, about 0.50 dB / mVpp, about 1.0 dB / mVpp, about 1.5 dB / mVpp or about 2.0 dB / mVpp.

[0089]

[0085] It will be understood that the example values presented above for the target parameter range may be dependent on the efficiency of the ultrasonic generation system (including the transducer 210) as well as the sensitivity of the detection system (including the sensor 220). The values and ranges presented above were based on the ultrasonic generation system and detection system as described in the experimental examples of the present disclosure. Particularly, voltage (mVpp) is at the output of a waveform generator, which in these examples is an input to 50 dB radio-frequency amplifier. The values may vary accordingly with use of a system having different characteristics.

[0090]

[0086] At step 112, if the determined parameter is outside of the target range, the selected amplitude is then modified at step 114, and this modified selected amplitude is the input into step 102 for the next loop.

[0091]

[0087] In some examples, modifying the amplitude 114 may comprise increasing the amplitude or decreasing the amplitude by a selected amplitude increment. The selected amplitude increment may be the same each time the feedback loop is executed, or may vary.

[0092]

[0088] In some examples, the initial selected amplitude at 102 may be relatively low. In such case, the initial detected acoustic emission may provide a determined parameter (e.g. an intensity value) which is below the target value / range. In this example, the selected amplitude may be modified at 114 by increasing the amplitude by a selected increment each time feedback loop is executed, until a parameter within the target range is achieved. In some examples, the selected amplitude may be increased linearly (i.e. by the same increment each time). In other examples, the selected amplitude may be increased non-linearly. In one example, the selected amplitude is increased linearly using a predefined step size of 10 kPa. However, other step sizes may be used.

[0093]

[0089] In other examples, the initial selected amplitude at 102 may be relatively high, and the selected amplitude may be modified at 114 by decreasing the amplitude by a selected increment each time the feedback loop is executed, until a parameter within the target range is achieved at 116. In some examples, the selected amplitude may be decreased linearly (i.e. by the same increment each time). In other examples, the selected amplitude may be decreased non-linearly.

[0094]

[0090] In other examples, the selected amplitude may be increased or decreased relative to the initial selected amplitude. The amplitude increment may be selected based on the comparison of the determined parameter to the target range. For example, if the determined parameter is above the target range, the selected amplitude may be decreased (i.e. a negative increment) while if the determined parameter is below the target range, the selected amplitude may be increased (i.e. a positive increment).

[0095]

[0091] In some examples, a predefined pattern for modifying the selected amplitude at 114 may be followed. For example, a search pattern such as binary search (otherwise known as halfinterval search or logarithmic search) may be employed for modifying the selected amplitude.

[0096]

[0092] In some examples, the selected amplitude may be modified at 114 using a predictive algorithm. For example, once response intensity data including a plurality of determined intensities correlated to the respective input amplitude has been obtained, a modelled curve may be fit to the data and used to predict an amplitude at which a parameter within the target parameter range will be achieved.

[0097]

[0093] In general, the selected amplitude of 102 is varied until a determined parameter is achieved within the target range as indicated at step 116. However, as the determined parameter of the intensity data may vary or fluctuate slightly on repetition, the method 100 may comprise repeating the sequence of steps without further modifying the amplitude until a value within the target range has been achieved a predetermined number of times, to confirm accuracy of the result before proceeding. For example, when the determined parameter is within the target range, the method may comprise incrementing a count. When the count has a value less than a count threshold, the steps 102 to 110 of the method may be repeated without increasing the selected amplitude, that is, at the current value of the selected amplitude. When the count reaches a value equal to the count threshold, the method proceeds to step 118. In some examples, the count threshold may be equal to 2, 3, 4, 5 or more. In some cases, once a parameter within the target range is achieved, the method proceeds directly to step 118. It will be understood that this may be considered not incrementing a count, or reaching a count threshold equal to 1.

[0098]

[0094] Once the determined parameter is achieved within the target range as indicated at step 116 and, optionally, once the count threshold is satisfied, the method 100 proceeds to step 118, in which the current selected amplitude is identified as a therapeutic amplitude.

[0099]

[0095] In some examples, as indicated by the dashed line box 120 in the flowchart of Figure 1, the method may comprise delivering further ultrasound energy to the location within the subject’s brain at the identified therapeutic amplitude for a selected duration and / or to deliver a preselected dose of ultrasound energy (for example as defined by a predetermined number of pulses). This may be considered a “therapeutic ultrasound delivery” phase (as encompassed by, for example, step 120 of Figure 1), as opposed to the preceding “therapeutic level determination” phase (as encompassed by, for example, step 102 of Figure 1). In some examples, the total duration and / or dose of ultrasound delivered at the therapeutic amplitude may be adjusted to compensate for a duration and / or dose of ultrasound energy delivered in the preceding method step 102 for identifying the therapeutic amplitude. In some examples, the further ultrasound energy (i.e., step 120 of Figure 1) is delivered for a predetermined duration of 60 seconds. However, in other examples, the further ultrasound energy may be delivered for a predetermined duration of between about 1 second and about 1 hour.

[0100]

[0096] In some examples, where the further ultrasound energy at step 120 is delivered as a series of pulses, for each pulse, the steps of detecting an acoustic emission 104, analysing the acoustic emission to obtain response intensity data 108, determining a parameter of the response intensity data 108 and comparing the parameter to the target parameter range 110 may be repeated (as indicated by the dashed line arrow in Figure 1). If the determined parameter is outside of the target parameter range as per 112, the selected amplitude may be further modified at 114 (e.g. further increased) for subsequent delivery of pulses at step 120. However, as long as the determined parameter is within of the target parameter range as per 116, delivery of further ultrasound at step 120 may continue using the identified therapeutic amplitude. In such examples, the delivered dose may be determined based on the number and / or duration of pulses delivered pulses which generated a detected acoustic emission having an intensity data parameter within the target parameter range.

[0101]

[0097] The appropriate dose of ultrasound delivered at the therapeutic amplitude may be determined experimentally in order to achieve BBB opening.

[0102]

[0098] In some examples, the delivered dose may be determined using the formulae set out below. As set out below, dose may be defined in linear scale value, which can be converted to dB scale by taking 1 Oxlog 10(CD). SBase(t,n) and Scav(t,n) denote the PCD time waveform obtained in the absence and presence, respectively, of circulating microbubbles at a target region. For simplicity, SBase(t,n) and Scav(t,n) can be regarded as fdtered signals. iBase(n) and Icav(n) denote intensity of the PCD response at nthpulse. To calculate the dose, Icav(n) is divided by baseline intensity, iBase(n), then summed up.

[0103]

[0099] A desirable range for the preselected dose dose, as measured using the method described above, may be in the range of about 20-60 dB. In some examples the preselected dose may be about 35 dB. However, it should be noted that this range may vary depending on the PCD's sensitivity. Ideally, determining the delivered dose should be based on a deconvolved (i.e. conversion from voltage to pressure domain) response.

[0104]

[0100] In some examples, the further ultrasound energy (therapeutic ultrasound, i.e., step 120 of Figure 1) is delivered using a longer pulse length than the delivery of ultrasound energy in step 102. For example, where “short pulse” sonication is used for the therapeutic level determination in step 102, the method may comprise switching to “long pulse” sonication for the therapeutic ultrasound delivery in step 120. Alternatively, the therapeutic ultrasound delivery in step 120 may also be delivered using “short pulse” sonication, or both the delivery of ultrasound energy in step 102 and the further “therapeutic” ultrasound energy delivery in step 120 may be performed using “long pulse” sonication.

[0105]

[0101] Figure 2 shows an example system 200 suitable for performing the method of the present disclosure. The system 200 includes an ultrasonic generation system including a transducer 210.

[0106]

[0102] The system 200 further comprises a sensor in the form of a passive cavitation detector (PCD) 220, which is configured to detect an acoustic emission from the brain generated responsive to the delivering of the ultrasound energy.

[0103] The system 200 also comprises a controller 230 including a processor in communication with the transducer 210 and the cavitation detector 220. In some examples, the controller may be a personal computer. In other examples, the controller may be a field programmable gated array (FPGA) or other suitable computing device. The controller may be equipped with a waveform generator 232 and a digitizer 234. For example, the waveform generator 232 and the digitizer 234 may be two PCIe-based cards. The controller 230 may be configured to facilitate the transmission of focused ultrasound (via the waveform generator 232)and detection of acoustic emission (via the PCD 220) to be performed at the same time. For example, the controller 230 may be configured to operate the digitiser 234 as a master in synchronisation with the waveform generator 232 via an external triggering circuit.

[0107]

[0104] The waveform generator 232 is communicatively coupled to the ultrasound transducer 210. For example, the output of the waveform generator 232 may be connected to the ultrasound transducer 210 via a radio-frequency (RF) amplifier, or other suitable communication means.

[0108]

[0105] The PCD 220 may be located within, such as at the centre of, the transducer 210, for example as shown in Figure 2. The PCD 220 is communicatively coupled to the digitiser 234 such that the digitiser receives detected signals indicative of detected acoustic emission from the PCD 220. For example, the PCD 220 may be connected to the digitizer 234 input via preamplifier 218, as shown in the Figure 2.

[0109]

[0106] The transducer 210 is configured to deliver ultrasound energy to cause cavitation of microbubbles in the subject’s brain. For example, the probe unit containing the transducer 210 and the PCD 220 may be coupled to the target (e.g., the subject’s skull or scalp 10) via degassed water in a coupling cone 212 and coupling gel 20. The transducer 210 may be configured to deliver the ultrasound energy to a target region in the subject’s brain. The target region within the brain may be determined, in part, by the location over the skull 10 at which the transducer 210 is placed. The target region may be further determined based on a determined focus of the ultrasound transducer (for example, as described in more detail in Example 1, below).

[0110]

[0107] In use, the controller 230 commences the sonication process at step 102 of method 100 with a given set of ultrasound parameters such as amplitude, pulse length, pulse repetition frequency, and sonication duration. These parameters are all programmable and may be updated or adjusted during an interval between pulses. The amplitude and / or pulse length may be updated in real time based on the presently disclosed method for identifying a therapeutic amplitude for delivery of the ultrasound energy.

[0111]

[0108] The examples describes herein may allow determination of an input amplitude required in order to achieve a desired level of cavitation intensity for BBB opening. The method may be utilised as a feedback controller for an ultrasound system to account for the variable attenuation of ultrasound signals through the skull of large subjects such as sheep, humans or the like.

[0112] Experimental examples

[0113]

[0109] A system according to the present disclosure was used to identify a therapeutic amplitude for delivery of ultrasound energy according to the method of the present disclosure, as further described in the examples below.

[0114] Device configuration and characterisation

[0115] [HO] The device used in the examples below was configured in accordance with the example shown in Figure 2. The device included a host system, including the real time processor running a feedback controller 230, a focused ultrasound generation system 210 and a passive cavitation detection system 220. In this example, the host system is a personal computer equipped with two PCIe-based cards, one the waveform generator 232 and the other the digitizer 234. The feedback controller 230 runs the software on the computer (although this could also be implemented on hardware such as a field programmable gated arrays, FPGAs) and operates the digitizer 234 as a master in synchronisation with the waveform generator 232 via an external triggering circuit, allowing delivery of ultrasound energy and acoustic emission detection to be performed at the same time.

[0116] [Hl] The output of the waveform generator 232 is connected to the ultrasound transducer

[0117] 210 via a radio-frequency (RF) amplifier. The input to the digitizer 234 is connected to the PCD 220 placed at the centre of the transducer 210 via a preamplifier 218. The probe unit containing the transducer 210 and the PCD 220 is coupled to the target (e.g., the human skull 10) via degassed water in the coupling cone 212 and coupling gel 20.

[0118]

[0112] The feedback controller 230 commences the sonication process (i.e. delivering ultrasound energy as in step 102) with a given set of ultrasound parameters, including an initial selected amplitude (established during the pre-treatment planning), pulse length, pulse repetition frequency, and sonication duration. These parameters are all programmable and can readily be updated during an interval between pulses. In this example, based on analysis of detected acoustic emission, the selected amplitude (i.e., PNP) and pulse length were updated in real time based on the proposed algorithm for the feedback controller 230, as discussed in more detail below with reference to Figures 8 and 12.

[0119]

[0113] To determine the focus of the ultrasound transducer 210, 3D scanning was performed in an acoustic measurement tank (Onda Corp.). The 3D beam profile was first measured in free- field and then in the presence of a rehydrated sheep skull. Measuring the beam profile confirmed the focus at the expected location, as shown in Figure 3. In Figure 3, the Z axis is the direction of ultrasound propagation, with Z = 0 being the back side of the transducer 210. Apart from a distortion compared to the beam profile that had been measured in free-field, neither the focal size nor its location were significantly altered. Pressure as a function of the selected amplitude (output of the waveform generator) was measured with a skull and in free-field, revealing a skull attenuation for this specific spot of 76%. The dots in Figure 3, panel c, represent measured data points, while the dotted line is the best fitted line obtained by ID linear regression analysis. PCD signal processing

[0120]

[0114] PCD signals (i.e. detected acoustic emission in this example) are typically analysed in the frequency domain, based on the spectrum obtained by applying a Fourier transform to raw PCD signal waveforms. This reveals several components of the acoustic emission, which are higher harmonic (HH), sub-harmonic (SH), ultra-harmonic (UH), and broadband (BB) emissions. Traditionally, HH as well as SH and UH emissions have been considered indicators of ‘stable cavitation’ of the microbubbles, which refers to the repetitive oscillation of microbubbles driven by periodic pressure alterations between compression and rarefaction phases of the primary ultrasound waveform. When stronger ultrasound pressures are applied, the microbubbles inflate to a larger size in the rarefaction phase such that the ‘inertia’ of the surrounding medium starts exerting a force towards the centre of the microbubble. This drives the microbubble to deflate and eventually to collapse. Of note, it is the inertia of the medium, and not the compression phase, that has an effect on the microbubble dynamics, and this phenomenon is hence termed ‘inertial cavitation’. Since a broad frequency response of a shock wave is emitted when the microbubble collapses, BB emission has traditionally been considered as an indicator of inertial cavitation and associated with tissue damage.

[0121]

[0115] This classification of cavitation has been accepted in the field for a long time. As a result, ‘stable cavitation’ has been considered desirable and ‘inertial cavitation’ has been considered undesirable in achieving safe opening of the BBB. However, in more recent studies, SH has now been associated with de-stabilisation of microbubbles. Moreover, a microscopic observation of the dynamics of single microbubbles provided new insight into the mechanism of the different components of the PCD signature, revealing that HH emission can in fact be produced by periodically collapsing microbubbles. Given that microbubble collapse is a unique feature of inertial cavitation, this finding implies that the detection of HH emission does not necessarily indicate stable cavitation. When a stronger pressure was applied, microbubble fragmentation was found to occur during its collapse and the jitter in the timing of the recorded shock waves correlated with the detection of BB emission. When the pressure was further increased, the period of the microbubble collapse increased by a factor of two, producing clear peaks at the SH and UH frequencies.

[0122]

[0116] In short, all four emission components (HH, SH, UH, and BB) have been observed under a regime of inertial cavitation, suggesting that the traditional association of stable cavitation with BBB opening, based on the specific detection of HH, UH and SH emissions, is likely inadequate. In contrast, micro-jetting and shock emission, which are signature phenomenon of inertial cavitation, may be the underlying mechanism of BBB opening.

[0123]

[0117] As discussed above, the skulls of large animals or humans may present variable and difficult to predict attenuation of ultrasound energy. By contrast, for small animals, such as mice and rats, variability of ultrasound energy attenuation is reasonably negligible, and a nominal attenuation can therefore be applied. When BBB opening is achieved in small animals, the detection of BB emission often correlates with extravasation of red blood cells. As such, prior ultrasound feedback controllers to date have typically been developed based only on detection of HH or SH / UH emission.

[0124]

[0118] However, the inventors have found, surprisingly, that BB emission may be detected even at pressures lower than the threshold pressure of BBB opening. Moreover, it has been found that the intensity of BB emission reveals a strong correlation with PNP up to 0.7 MPa, when intense bleeds were consistently found. While HH emission also shows a correlation with PNP, the inventors have found that the performance of BB emission as a predictor of PNP may be superior to HH.

[0125]

[0119] The only ultrasound feedback controller that has been clinically deployed so far uses a SH-based ramp up. In these studies, the PNP was monotonically increased (by increasing amplitude) until SH emission was detected. Subsequently, the amplitude was reduced to half of the amplitude at which SH emission was detected. Then, the amplitude was maintained throughout the duration of the sonication (during which SH was not detected). In another SH- based example, a modified controller also commences sonication at a low pressure and then monotonically increases pressure until SH emission is detected. Interestingly, upon SH detection, the pressure was not reduced but maintained, to compensate for reduced microbubble concentration due to infusion. Sonication was terminated when the dose (defined in this example as the sum of SH power by individual pulses) reached a pre-defined value.

[0126]

[0120] By contrast, the method and system of the present disclosure utilises a feedback controller based on BB and / or SW emission. In large animals and humans, the threshold pressure of BBB opening is expected to be sufficiently high to generate BB emission.

[0121] As discussed above, PCD signals are often analysed in several components, depending on the selected frequency band. Figure 4 illustrates the frequency bands used in the present examples. Figure 4 shows a representative frequency spectra obtained through a sonolucent implant. Figure 4, panel a, shows the frequency spectra for 5 cycles of short pulse sonication. Enhanced frequency response by microbubbles (solid line) compared to baseline (dotted line) is depicted. Figure 4, panels b and c, show the frequency spectra for long pulse (20 ms) sonication, depicting the enhanced frequency response from microbubbles (panel b) compared to baseline (panel c).

[0127]

[0122] The shaded band in Figure 4, panel (a) represents the selected frequency range for “shockwave” (SW) emission. When the ultrasound energy was delivered using a short pulse, because of poor spectral resolution as shown in Figure 4, panel (a), the detected acoustic emission was filtered to extract a continuous band from 2 to 4 MHz, which was defined as shockwave (SW) emission. In this example, fo was 500 Hz and the selected frequency range of 2 to 4 MHz therefore corresponded to a range of 4 / o to 8 / o. The intensity (or magnitude) of the SW emission represents the strength of microbubble collapse. When the ultrasound energy was delivered using a single continuous pulse (long pulse sontication - 500 kHz sinusoidal wave), both conventional HH and BB emission components were measured using four discrete bands as depicted by the arrows and the shaded bands, respectively, in Figure 4, panel (b).

[0128]

[0123] An alternative method for analysing an acoustic emission using the time domain is illustrated in Figure 18. This example shows filtering of the acoustic emission to extract SW emission from a raw PCD waveform in the time domain. In Figure 18a, the dotted line represents a raw PCD waveform obtained by using a short pulse sonication. A bandpass filtered waveform is represented by the solid line, which reveals that cavitation activity occurred at 48- 70 mm in depth. Figure 18b shows zoomed in waveforms in that depth region. A 'bandpass' function in Matlab (“firl” function) was used with cut-off frequencies at 2 and 4 MHz. The frequency spectra in Figure 18c clearly shows the filtering effect, while the dashed line represents a frequency mask used for frequency-domain filtering. In Figure 18d, a passband (2-4 MHz) in which intensity is measured is zoomed in to show a small discrepancy between the raw and filtered spectra, which resulted in a small gap in the measured intensities when filtering in the time vs frequency domain.

[0129] Example 1 - Long pulse sonication

[0130] A. In vitro experiment

[0124] An in vitro experiment was conducted as described below. The experimental setup for the in vitro experiment is illustrated in Figure 5.

[0131]

[0125] A probe unit as described above, including transducer 210 and PCD 220 was coupled to a degassed and rehydrated sheep skull 10 in an acoustic measurement tank fdled with degassed water. 3D scanning was then conducted to identify the focus for any given spot of the skull 10. The attenuated pressure was measured using a needle hydrophone (HNR-1000, Onda Corp.) 40, calibrated by the National Physics Laboratory (NPL, Teddington, UK). Under visual guidance, based on the hydrophone 40 position, a capillary tubing 30 connected to a 3D-axis manual stage was positioned at the focus, as illustrated in Figure 5.

[0132]

[0126] To determine baseline acoustic emission levels, ultrasound delivery was performed while flowing only saline into the tubing 30. Then, Defmity (Lantheus) microbubbles were activated and diluted with saline in a 50 m syringe to achieve a 1:50 concentration. The microbubble solution was delivered into the tubing 30 at 5 mL / h using an infusion pump (Harvard Apparatus). While the microbubbles were flowing, delivery of ultrasound at a selected amplitude using 2 ms continuous pulses (long pulse) was performed. The selected amplitude was modified by increased monotonically by a fixed increment of 10 mVpp on the waveform generator for every pulse. Simultaneously, acoustic emissions were detected by the PCD 220 and recorded. The saved data were processed offline using a custom-developed Matlab script. Data acquisition was repeated at least three times.

[0133]

[0127] The detected acoustic emission was analysed to obtain response intensity data. The response intensity data was then analysed to determine an intensity of a broadband (BB) emission component. An example curve of BB emission intensity as a function of selected ultrasound amplitude is shown in Figure 6. In general, it was found that BB emission intensities increased linearly beyond the onset amplitude. Considering a small variability of baseline intensity, the onset was defined as a 1 dB increase in intensity above baseline, as illustrated in Figure 6. This value was set arbitrarily and may take other values in other examples.

[0134]

[0128] The intensity of BB emission (cavitation intensity) as a function of the selected driving amplitude obtained from the four sheep skulls in this example is plotted in Figure 7. Attenuation of the sheep skulls ranged from 66 to 84%. The attenuation was found to correlate with several factors including onset amplitude, rate of intensity increase (i.e., slope), and absolute intensity at a given selected driving amplitude. For skull 3 (68% attenuation) BB emission developed at the lowest selected amplitude, increased in intensity most rapidly, and reached the highest intensity level at any driving amplitude. On the contrary, for skull 2 (84% attenuation) BB emission developed at the highest selected amplitude, increased in intensity most slowly, and reached the lowest intensity level at any driving amplitude. Skull 1 (76% attenuation) and Skull 4 (66% attenuation) sat between Skull 2 and Skull 3. Skull 4 exhibited a slightly weaker response, although the attenuation was similar. Considering the measured PNP data in Figure 7, panel (b), this is considered due to higher attenuation on detection frequency with skull 4. Because the BB emission intensities are plotted in PNP values, the intensity drop from free-field to skull response reflects detection attenuation. Skull 4 showed stronger attenuation than Skull 3. This indicates that primary factor affecting BB emission intensity may be the attenuation on transmission.

[0135] B. Sheep experiments

[0136]

[0129] The method 100 according to the present disclosure was tested in live sheep, using the system 200 depicted in Figure 2.

[0137]

[0130] Figure 8 shows an example of an algorithm utilising a determined intensity of a broadband (BB) component of a detected acoustic emission as an input to a feedback controller.

[0138]

[0131] Referring to Figure 8, first, an amplitude range and initial selected amplitude input to the driving system is selected and a pulse count (Npuise) is set to 0. A pulse count threshold (Nnioid) is also set at this stage. The pulse count threshold may be selected as an integer number between about 10 and about 300. Next, prior to delivering microbubbles, ultrasound energy is delivered to determine a baseline acoustic emission intensity, which is analysed to determine the intensity level of BB emission (BBnase). A target parameter value is set, which in this case is the targeted broadband emission intensity level (TBL). The target parameter range is therefore equal to or above TBL. Of note, TBL and Nnioid are predetermined experimentally. The controller 230 stays in an idle state until it receives trigger input. The microbubbles are delivered to the subject, then, the first trigger is given by the operator, commencing the delivering of ultrasound energy as per step 102. For the remainder of the sequence, triggers initiating delivering of ultrasound energy at repetitions of step 102 are provided automatically by the detection system 200. As the system 200 delivers ultrasound energy at step 102, a cavitation signal (acoustic emission) is generated from the brain and detected at the passive cavitation detector 220 as per step 104. The acoustic emission is analysed as per step 106 by the processor of the controller 230 and an intensity of a BB component of the acoustic emission is determined.

[0139]

[0132] The determined BB component intensity is compared to the target parameter range. As indicated in Figure 8, as long as the determined BB component intensity is smaller than TBL (i.e. outside of the target range as per 112), the feedback loop continues, with the system modifying the selected amplitude as per step 114 at each repeat of the feedback loop. In the present example, the selected amplitude is linearly increased by a predefined step (10 kPa in this example). In other examples, the amplitude may be modified in other ways as previously described. When the determined BB component intensity is greater than or equal to TBL (i.e. within the target parameter range as per 116), the current selected amplitude is identified as a therapeutic amplitude as per 118. The selected amplitude is maintained and further “therapeutic” ultrasound delivery is commenced as per step 120, with the pulse count increased by one for each pulse generating an acoustic emission with a BB emission intensity having a value within the target parameter range (above TBL). When the pulse count reaches the threshold (Nnioid), sonication was terminated.

[0140]

[0133] Using the algorithm depicted in Figure 8, sonication was delivered to three spots (SOI, S02, and S03) in a sheep brain. Here, TBL was set to be 10 dB higher than the baseline and Nrhoid was 60. BBB opening was confirmed by gadolinium leakage via T1 MRI scans.

[0141]

[0134] Figure 9 shows the selected amplitude (squares) and the corresponding determined intensities of the BB component of the detected acoustic emission (circles). As is evident from the amplitude profiles, the sonication commenced at a sufficiently low selected amplitude such that BB emission was not detected in the acoustic emission. Upon increasing the selected amplitude, broadband emission became detectable. For example, as indicated by the grey arrows, after 42 s, 29 s, and 29 s, respectively, BB emissions were detected with a small rise in intensity (from 18 dB to 20 dB) in SOI, S02, and S03, respectively. Subsequently, intensities in the three spots gradually increased towards TBL. As indicated by the black arrows, the intensities reached TBL at 65 s, 43 s, 58 s in the three spots, respectively. Given some signal fluctuations, although the TBL had been reached, there were a few further selected amplitude increases for some of the subsequent sonication events, whenever the detected intensity of the BB component of the acoustic emission happened to be lower than TBL (i.e. outside of the target parameter range). The sonication was terminated when the number of delivered pulses satisfying pulse count reached 60, which was the case at 143 s, 115s, and 121 s for SOI, S02, and S03, respectively.

[0142]

[0135] For SOI, from 120 s onwards (vertical dashed line), further increases in detected BB component intensities were detected, which is considered to be due to additional cavitation activity in the coupling space 212 because of the high driving amplitude of 445 mVpp. Further inspection revealed that there was a high frequency component in the detected PCD signal, that was assumed not to be brain-derived given the high attenuation of the skull at this high frequency range.

[0143]

[0136] After completing sonication of SOI, S02 and S03, the sheep had T1 MRI scans which showed the anatomical structure of the brain. In the follow-up T1 scan performed after injecting gadolinium, leakage was detected as hyperintense pixels. By comparing both MR scans, signal enhancement in terms of percentage was calculated and mapped showing three clearly visible foci of BBB openings.

[0144]

[0137] The same algorithm was applied to treat a second sheep, but failed to achieve BBB opening, as shown in Figure 10. In contrast to sheep 1, the BB component of the acoustic emission showed discrete responses and was detected at a lower selected amplitude. The first detection was seen at 23 sec (indicated by an arrow), and the determined BB intensity suddenly increased to above the TBL threshold (Figure 10 (b)). For the remaining ultrasound pulses, most detected BB component intensities were highly variable with stronger intensity than TBL, even though selected amplitude was at least 100 mVpp lower than in sheep 1. Hence, the BB emission was further analysed at low and high frequencies. The low frequency BB emission was found to be close to that detected during sonication. High frequency BB emission was analysed at 5 MHz. As skull attenuation increases in proportion to frequency, it is assumed that BB emission from brain would not be detected through the skull at this frequency. However, as seen in Figure 10, panel (c), strong high frequency BB emission was detected, particularly in correlation with low frequency BB emission (Figure 10(c)). Therefore, the scalp and skull are suggested as potential cavitation sources other than the brain, given both require blood supply and thus may contain microbubbles. In addition, the pressure above and inside the skull can be a few times stronger than in the brain, and acoustic emission signals are less attenuated, which would allow for detection of strong BB emission even for a weakly vascularised scalp or skull. Therefore, it was assumed that the scalp / skull may be an additional cavitation source that can generate a significantly stronger PCD response acting as an interference overriding the PCD response from brain.

[0145] Example 2 - Short pulse sonication

[0146] A. In vitro experiment

[0147]

[0138] To be able to spatially resolve the various cavitation sources, and to measure cavitation activity solely from the brain, as a strategy, ultrasound energy was delivered using a short pulse sonication (e.g., 5 cycles) was employed in the ramp up sequence, as illustrated in Figure 11.

[0148]

[0139] Short pulses are commonly used in ultrasound imaging, as they improve the resolution in the propagation axis. To assess the feasibility of the method 100 using short pulse ultrasound energy, the same in vitro setting as in Figure 5 was configured, i.e. using capillary tubing. The ultrasound energy was delivered as a set of 20 pulses (i.e., a burst) at the same selected amplitude, separating 5 -cycle pulses by 1 ms (Figure 11(a)). The 1 ms pulse interval was chosen to allow for sufficient time for signals (e.g., reflection, scattering, microbubble acoustic emission) to travel back to the PCD detector 220, such that signals from different source pulses would not overlap. Whenever a subsequent burst was excited, the selected amplitude was increased by a small step (e.g., 10 mVpp) starting at 50 and going up to 450 mVpp.

[0149]

[0140] Figure 1 lb shows a heatmap of the detected acoustic emission response from the 1stpulse out of 20 pulses. An increased acoustic emission response was only observed at position 50-100 mm (in the Z axis) where the capillary tubing had been placed. Since there was no microbubble supply to the skull, the detected PCD response from the skull was relatively low as indicated by the white arrow. All detected responses at the selected depth to the 20 pulses (Figure 11c) were averaged and plotted as a function of the selected amplitude at the target depth (Figure 1 Id), showing that the acoustic emission intensity increased linearly compared to the baseline response (lower dataset, Figure l id). The experiment was repeated, as indicated by the two response datasets in Figure 1 Id. Although the experiment was performed with only one skull, factoring in the long pulse results shown in Figure 7, it is reasonable to assume that an acoustic response using short pulse sonication would vary similarly depending on skull attenuation.

[0150] B. BBB opening in sheep with an intact skull

[0151]

[0141] The flowchart of the feedback controller algorithm based on the short pulse sonication is presented in Figure 12. As with the algorithm using long pulse sonication discussed in relation to Figure 8, baseline acoustic emission levels (PCDnase) were measured prior to delivering microbubbles. The baseline level and the target parameter range (in this case equal to or greater than a target acoustic response intensity level (Target cavitation level = TCL) are given as an input to the controller 230 after which the microbubbles were delivered. While the microbubbles circulate in the brain, delivery of ultrasound energy commences at a low selected amplitude. The selected amplitude is modified (in this case, gradually increasing) for each subsequent burst delivery of ultrasound energy. In this example, during the therapeutic level determination phase, the selected amplitude was monotonically increased (thus named “ramp-up” sequence).

[0152]

[0142] In this example, TCL is measured at the target depth (Drarget), which can be determined based on neuro-navigation or the envelope signal (as indicated in Figure 13, panel (c)). As soon as the acoustic emission detected at the PCD exhibits a response at the target depth having a determined intensity value which is larger than TCL (i.e. within the target parameter range), a count (N) is increased by one. Delivery of ultrasound energy is repeated with the selected amplitude held constant until the count (N) reaches the count threshold (the count threshold was set at three for this example but may be any positive integer value), at which point the current selected amplitude is identified as a therapeutic amplitude. The sonication mode is converted to “follow-up” (further or therapeutic ultrasound delivery) mode. In follow-up mode, the ultrasound energy is performed using a fixed amplitude set at the identified therapeutic amplitude, as determined during the ramp-up sequence, for a pre -determined duration (e.g., 60 sec).

[0153]

[0143] This ramp up algorithm may provide relatively simple control, and the output provides a good overview of the acoustic emission response as a function of the selected amplitude / PNP. However, the ramp-up sequence may be modified (e.g. using alternative means of modifying the amplitude other than linearly increasing), which may enable using fewer sonication events in identifying the therapeutic amplitude. This may reduce the overall sonication duration per target area and ultimately per treatment session.

[0154]

[0144] A representative result from one target area (Spot 6) of sheep 3 through an intact skull is shown in Fig. 13. Baseline acoustic emission levels in Figure 13, panel (a), show a low intensity level relative to the acoustic emission response in the presence of microbubbles (Figure 13, panel (b)), which revealed two major acoustic emission responses at 40 mm and 60 mm. Here, the distance was calculated based on the arrival time of the signals using 1500 m / s as the constant speed of sound. At 40 mm there was a significantly strong response, detected at lower amplitude (approximately 70 mVpp) than at 60 mm. Maximum projection revealed that the detected response at 40 mm coincided with the envelope signal (Figure 13, panel (c)) which is the reflection of the skull by pulse-echo. Therefore, it was concluded that the response at 40 mm was from the skull and accordingly, the response at 60 mm from the brain, validating the assumption made above.

[0155]

[0145] The two dotted lines in Figure 13, panel (b), indicate the range of the selected target depth, determined based on the envelope signal obtained prior to injecting microbubbles (Figure 13, panel (c)) and knowing the skull thickness. From the selected range of depth, the maximal intensity of the acoustic emission detected at the PCD was derived for each selected amplitude and then compared to TCL during sonication in real time. Figure 13, panel (d), shows the maximal acoustic emission intensity in comparison to the TCL (horizontal dotted line). The controller stopped sonication at 380 mVpp as it was programmed to stop after a count reached three detected acoustic emissions having a determined intensity higher than TCL (N = 3). Responses for S01-S04 are presented in Fig. 14, confirming that the controller operated as programmed. Follow up ultrasound delivery with long pulse sonication (20 ms) using the identified therapeutic amplitude opened the BBB, as verified by subsequent gadolinium signal enhancement imaging. C. Sheep cranioplasty experiment

[0156]

[0146] To further validate the findings in sheep 3, the study was repeated in a fourth sheep (sheep 4) that had undergone cranioplasty. Here, a 40 mm x 50 mm portion of the skull was cut out and replaced by the same size of 3 mm thick homogeneous sonolucent implant with a known attenuation. As shown in Figure 15, three pressure levels distributed to six target regions (i.e. “spots” 1-6) were tested in an attempt to determine the threshold pressure for BBB opening and tissue damage.

[0157]

[0147] Due to the sonolucent implant having a weaker attenuation than natural sheep skull, the selected amplitude range in the ramp up sequence was lowered to, for example, 10 to 180 mVpp for spots 3 and 6. For the other spots, the selected amplitudes were adjusted such that the peak amplitudes were reached at the end of the ramp-up sequence.

[0158]

[0148] Figure 16 shows results obtained for the ramp up sequence in spot 6. In Figure 16(a), in the absence of microbubbles, the baseline acoustic emission was kept low for all selected amplitudes. In contrast, Figure 16(b) showed that delivery of microbubbles enhanced the detected acoustic emission response intensity by 20 dB. Maximum projection in Figure 16(c) revealed that a part of the PCD response coincided with the envelope signal, which is a reflection from the implant. This is because the length of the window where the PCD was measured is 5 times larger than the implant thickness (i.e., 15 vs 3 mm) and thus, the PCD response from the tissue sitting under the implant was included in the measurement. The strong acoustic emission response consistently detected under the implant was the brain.

[0159]

[0149] In each spot, once the ramp up sequence was completed, the sonication mode was immediately converted to “follow up” (or therapeutic ultrasound delivery), whereby further ultrasound energy was delivered using a long pulse at the identified therapeutic amplitude (i.e. within the target parameter range, when the detected acoustic emission has an intensity value equal to or greater than TCL) for 60 sec. This resulted in variable BBB opening, depending on the applied pressures.

[0160]

[0150] Figure 17a shows a fluorescence image of Evans blue leaked into brain tissue, confirming BBB opening. The largest opening was found for S06, weaker opening for S02 and S05, and no opening for SOI and S04. This correlates well with the applied pressure, except for S03, which may be due to the location of the spot in the ventricle.

[0161]

[0151] The intensity of the acoustic emission responses from the ramp up sequence increased linearly as a function of the selected amplitude, and 5 out of responses 6 (except for SOI) overlaid with each other, demonstrating the repeatability of the procedure (Figure 17b, c). From the determined BB component of the acoustic emission induced by the long pulse sonication, shown in Fig. 17d, a potential correlation of pressure with BBB opening was found, particularly at 0.35 - 0.50 MPa, believed to be within the BBB opening threshold. BB emission at higher pressure than the opening threshold PNP is distinguished by smaller error bar. This suggests that BBB opening in sheep may require pressures that consistently generate BB emission. HH emission in Figure 17e was strongly detected even at 0.35 MPa with small variability but the intensity did not significantly vary when higher pressures were applied.

[0162]

[0152] Through the series of experiments described above, the feasibility of using a determined BB component of detected acoustic emission (with long pulse sonication) and / or SW component of detected acoustic emission (with short pulse sonication) as an input to a feedback controller for ultrasound delivery to achieve opening the BBB was verified.

[0163]

[0153] In vitro results showed that both BB and SW emission components linearly increased as a function of ultrasound pressure (PNP). Particularly, the rate of increase in BB emission intensity well reflected skull attenuation. A lower rate of intensity rise was found to be associated with a higher attenuation. In case of SW emission, the experiment was performed with one skull, assuming that the rate of intensity rise will also vary depending on skull attenuation, (i.e., a higher rate of increase in intensity with a lower attenuation).

[0164]

[0154] The sheep experiments described above demonstrate that a BB emission-based feedback controller could achieve BBB opening successfully provided that sksull / scalp interference can be ruled out. To avoid this interference, short pulse sonication was used, with pulse length reduced from 20 ms to 10 ps (= 5 cycles), which allowed for the cavitation source to be spatially resolved. One alternative approach to reducing interference, in combination with the methods described herein, is passive acoustic / cavitation mapping or using a tightly focused PCD or transducer (i.e., low f-number = radius of curvature / diameter). For example, in the case of InSightec’s hemisphere array (f-number = 0.5), a focus is generated by using multiple transducer elements such that individual interference would be minimal.

[0165]

[0155] By using short pulse sonication, SW emission was measured rather than BB emission because of degraded spectral resolution due to the short time window in Fourier transform. The SW emission-based feedback controller guided pressures and successfully generated BBB openings, confirmed by leakage of MR contrast agent. Repeatability of the method was demonstrated by cranioplasty.

[0166]

[0156] By cranioplasty, analysis of acoustic emission generated in response to long pulse sonication used in the follow up ultrasound delivery further confirmed that BB emission may be superior to HH emission for monitoring, and therefore as an input to a feedback controller. While HH emission showed small changes in intensities within the tested pressure range, BB emission responded more sensitively at around the threshold pressure of BBB opening (0.27 - 0.38 MPa). Variance of intensity was also significantly reduced when pressure was higher than the threshold pressure.

[0167]

[0157] With regards to a human application, it can be reasonably assumed that the threshold pressure of BBB opening in human would be sufficiently high to generate BB emission. Indeed, in our cranioplasty experiment, consistent BB emission was associated with BBB opening, suggesting that BB emission is an indicator of BBB opening, not of damage. Therefore, it validated that a simple principle of the proposed control algorithm is the command: ‘modify selected amplitude (e.g. to increase pressure) until BB emission is consistently detected’.

[0168]

[0158] Where a short pulse ultrasound delivery is used to minimise interference, SW emission is instead detected that is correlated to BBB opening. However, in some examples, sophisticated signal processing techniques such as wavelet transform may be used to allow extraction of the BB emission component from a short pulse spectrum.

[0169]

[0159] Since both BB and SW emissions rise gradually as the selected amplitude is increased, the method according to the present disclosure may allow consecutive steps in the “therapeutic level determination” phase to be shortened to just a few steps. For example, a few pulses with variable selected amplitude could be excited, and the detected acoustic emission intensity responses fit to a modelled curve to predict the selected amplitude at which the target level and / or range will be achieved.

[0170]

[0160] It will be appreciated by persons skilled in the art that numerous variations and / or modifications may be made to the above-described embodiments, without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

CLAIMS:

1. A method for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject, the method comprising: a) delivering ultrasound energy at a selected amplitude to in a subject’s brain to cause cavitation of microbubbles in the subject’s brain; b) detecting an acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; c) analysing the acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; d) determining a parameter of the response intensity data at the selected amplitude; e) comparing the determined parameter to a target parameter range; f) based on the comparison: i) when the determined parameter is outside of the target parameter range: modifying the selected amplitude; and repeating steps a) to f) with the modified selected amplitude; and ii) when the determined parameter is within the target parameter range, identifying the current selected amplitude as a therapeutic amplitude.

2. The method of claim 1, comprising delivering further ultrasound energy to the subject’s brain at the identified therapeutic amplitude for a selected duration and / or to deliver a preselected dose of ultrasound energy.

3. The method of claim 1 or claim 2, wherein analysing the acoustic emission signal comprises determining an intensity of a broadband emission component of the acoustic emission.

4. The method of any one of the preceding claims, wherein analysing the acoustic emission signal comprises filtering the acoustic emission with one or more band-pass filters.

5. The method of claim 4, wherein the band-pass filter is configured relative to a fundamental frequency (fo) of the ultrasound energy, wherein the band-pass filter is configured to filter the acoustic emission to a frequency range of between about f0and about 10fo.

6. The method of claim 5, wherein the band-pass filter is configured to filter the acoustic emission to a frequency range of between about 4f0and about 80.

7. The method of any one of the preceding claims, wherein analysing the acoustic emission comprises comparing the detected acoustic emission to a baseline acoustic emission generated responsive to delivering ultrasound energy to the subject’s brain in the absence of microbubbles.

8. The method of any one of the preceding claims, wherein analysing the acoustic emission comprises correlating the acoustic emission with depth based on arrival time of the acoustic emission.

9. The method of claim 8, comprising extracting a portion of the acoustic emission corresponding to a target depth and / or a target depth range.

10. The method of claim 8 or claim 9, comprising determining a distance of the subject’s skull from a source of the ultrasound energy.

11. The method of claim 10, further comprising determining a distance of a target region of the subject’s brain from a source of the ultrasound energy based on the determined distance of the subject’s skull from the source of the ultrasound energy and a predetermined distance from the subject’s skull to the target region.

12. The method of any one of the preceding claims, wherein the ultrasound energy in step a) is delivered in a sequence of short pulses.

13. The method of claim 12, wherein the pulses are separated by an interval of about 1 ms.

14. The method of claim 12 or 13, when dependent on claim 2, wherein the further ultrasound energy is delivered using a longer pulse length than the delivery of ultrasound energy in step a).

15. The method of any one of claims 1 to 11, wherein the ultrasound energy in step a) is delivered in a single continuous pulse.

16. The method of any one of the preceding claims, wherein step f) comprises: when the determined parameter is within the target range: incrementing a count; when the count has a value less than a count threshold, repeating steps a) to f) at the current value of the selected amplitude; and when the count reaches a value equal to the count threshold, identifying the current value of the selected amplitude as the therapeutic amplitude value.

17. The method of any one of the preceding claims, further comprising delivering the microbubbles to the subject.

18. The method of any one of the preceding claims, wherein determining a parameter of the response intensity data comprises determining an absolute or average intensity value of the data at the selected amplitude and wherein the target parameter range comprises a target intensity range.

19. The method of any one of the preceding claims, wherein determining a parameter of the response intensity data comprises:fitting a curve to the response intensity data; and determining a slope of the curve at the selected amplitude, wherein the target parameter range comprises a target slope range.

20. A system for identifying a therapeutic amplitude for delivery of ultrasound energy to a subject, the system comprising: an ultrasonic generation system including a transducer configured to deliver ultrasound energy to the subject’s brain to cause cavitation of microbubbles in the subject’s brain; a sensor configured to detect an acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; a controller including a processor in communication with the transducer and the sensor, the controller configured to: cause the transducer to deliver ultrasound energy at a selected amplitude; receive, from the sensor a detected acoustic emission from the brain generated responsive to the delivering of the ultrasound energy; analyse the detected acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; determine a parameter of the response intensity data at the selected amplitude; compare the determined parameter to a target parameter range and, based on the comparison: i) when the determined parameter is outside of the target parameter range: modify the selected amplitude; and cause the transducer to apply ultrasonic energy at the modified selected amplitude; and ii) when the determined parameter within the target parameter range, identify the current selected amplitude as a therapeutic amplitude.

21. The system of claim 20, wherein the controller is further configured to cause the ultrasonic transducer to deliver further ultrasound energy to the subject’s brain at the identified therapeutic amplitude for a predetermined duration and / or to deliver a preselected dose of ultrasound energy.

22. A method for identifying a therapeutic amplitude for use with an ultrasound transducer, the method comprising: receiving an acoustic emission detected from a brain, the acoustic emission generated responsive to delivering of ultrasound energy at a selected amplitude to cause cavitation of microbubbles in the brain;analysing the acoustic emission to obtain response intensity data comprising a determined intensity of the acoustic emission; determining a parameter of the response intensity data at the selected amplitude; comparing the determined parameter to a target parameter range and, based on the comparison: i) when the determined parameter is outside of the target parameter range: modifying the selected amplitude; and outputting a signal configured to cause a transducer to deliver ultrasound energy at the modified selected amplitude; and ii) when the determined parameter is within the target parameter range, identifying the current selected amplitude as a therapeutic amplitude.

23. A non-transitory computer-readable medium configured to perform the method of claim 22.

Images