Controller with PID logic for spectrogram-based control of blood-brain barrier opening

Abstract

A system for controllably changing a tissue property in a target volume. It comprises an ultrasound transducer that sonicates a target volume to change tissue properties by transmitting a sequence of acoustic pulses to the target volume, while a sensor array captures acoustic feedback generated by the sequence of acoustic pulses in the target volume. A controller (420) is provided, which (i) generates a spectrogram from the acoustic feedback, (ii) computationally calculates a degree of change of the tissue property in the target volume based on the spectrogram, and (iii) controls the ultrasound transducer based on the calculated degree of change of the tissue property to change the tissue property to a target degree set for a specific treatment.

Description

Controller With PID Logic For Spectrogram-Based Control Of Blood-Brain Barrier OpeningRELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Application No. 63 / 616,750, filed 12 / 31 / 2023, which is hereby incorporated by reference in its entirety.TECHNICAL FIELD

[0002] The present disclosure relates to systems and methods for controlling focused ultrasound treatments.BACKGROUND

[0003] The blood-brain barrier (BBB) is a highly selective mechanism that separates the circulating blood from the brain extracellular fluid. It plays a crucial role in maintaining the homeostasis of the brain microenvironment by regulating the exchange of nutrients, ions, and waste products between the blood and the brain. However, the BBB also poses a significant challenge for the delivery of therapeutic agents to the brain, as it restricts the entry of most drugs and large molecules.

[0004] Various methods have been developed to overcome the BBB and deliver drugs to the brain, including invasive techniques such as direct injection into the brain or implantation of drug delivery devices. However, these methods are associated with significant risks and limitations, such as infection, inflammation, and damage to the brain tissue.

[0005] Non-invasive methods for BBB opening have also been developed, such as focused ultrasound (FUS) BBB opening. FUS uses ultrasound waves to induce mechanical disruption of the BBB, using contrast agents (e.g., microbubbles, nanobubbles, or phase shift droplets) that are injected into the bloodstream and oscillate in response to ultrasound waves, causing transient disruption of the BBB. However, this method has limitations, such as the lack of precise real-time control over the BBB opening level, which could lead to over- or undertreatment, potentially damaging the brain tissue due to excessive or prolonged BBB opening.SUMMARY

[0006] In order to use FUS for BBB opening and enable effective drug delivery without causing damage, careful control of FUS parameters is essential. To ensure efficacious treatment and minimize collateral damage, such control should provide real-time feedback responsive both to the degree of BBB opening and the risks of tissue injury.

[0007] The present disclosure describes systems and methods for opening the BBB using a controller that is based on acoustic data obtained from ultrasonic receivers that capture the oscillations of microbubbles in and outside the treatment area. The system includes a controller that determines the optimal parameters for opening the BBB based on one or more spectrograms generated from the acoustic data, such as the pulse duration, acoustic intensity, acoustic frequency, or pulse repetition frequency (PRF) of ultrasound waves. The one or more spectrograms includes one spectrogram that is accumulated from the beginning of the treatment up to the real-time point of the treatment, and / or one spectrogram from the last N pulses, to calculate the total and / or local predictions, respectively. The controller uses a feedback-based control loop to adjust the parameters in real-time based on the total and local predictions to maintain the BBB opening level at a specific value. The controller can be powered by a machine learning artificial intelligence (Al) algorithm. The system can be used for drug delivery to the brain, as well as for diagnostics (e.g., liquid biopsy) and therapeutic purposes.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In the drawings, like reference characters generally refer to the same parts throughout the different views. Also, the drawings are not necessarily to scale, with an emphasis instead generally being placed upon illustrating the principles of this disclosure. In the following description, various embodiments of the present disclosure are described with reference to the following drawings, in which:

[0009] Figures 1A-1B depict an ultrasound system and an MRI system in accordance with some implementations.

[0010] Figure 2 depicts an implementation of an acoustic reflector substantially close to a target region in accordance with some implementations.

[0011] Figure 3 depicts a process for obtaining a spectrogram during a focused ultrasound procedure in accordance with some implementations.

[0012] Figure 4 depicts a process for predicting and controlling BBB opening levels based on spectrograms obtained during a focused ultrasound procedure in accordance with some implementations.

[0013] Figure 5 depicts training data preparation for use in BBB opening level predictions in accordance with some implementations.

[0014] Figure 6 depicts a schematic diagram of a process for predicting and controlling BBB opening levels during a focused ultrasound procedure in accordance with some implementations.

[0015] Figure 7 depicts results of Al-based BBB opening prediction in accordance with some implementations.

[0016] Figure 8 depicts results of Al-based BBB opening prediction in accordance with some implementations.DETAILED DESCRIPTION

[0017] Figure 1 A illustrates an exemplary ultrasound system 100 for generating and delivering a focused acoustic energy beam to a target region 101 within a patient’s body. The illustrated system 100 includes a phased array 102 of transducer elements 104, a beamformer 106 driving the phased array 102, a controller 108 in communication with the beamformer 106, and a frequency generator 110 providing an input electronic signal to the beamformer 106.

[0018] The array 102 may have a curved (e.g., spherical or parabolic) or other contoured shape suitable for placement on the surface of the patient’s body, or may include one or more planar or otherwise shaped sections. Its dimensions may vary between millimeters and tens of centimeters. The transducer elements 104 of the array 102 may be piezoelectric ceramic elements, and may be mounted in silicone rubber or any other material suitable for damping the mechanical coupling between the elements 104. Piezo-composite materials, or generally any materials capable of converting electrical energy to acoustic energy, may also be used. To assure maximum power transfer to the transducer elements 104, the elements 104 may be configured for electrical resonance at 50 Q, matching input connector impedance.

[0019] The transducer array 102 is coupled to the beamformer 106, which drives the individual transducer elements 104 so that they collectively produce a focused ultrasonic beam or field. For n transducer elements, the beamformer 106 may contain n driver circuits, each including or consisting of an amplifier 118 and a phase delay circuit 120; each drive circuit drives one of the transducer elements 104. The beamformer 106 receives a radio frequency (RF) input signal, typically in the range from 0.1 MHz to 10 MHz, from the frequency generator 110, which may, for example, be a Model DS345 generator available from Stanford Research Systems. The input signal may be split into n channels for the n amplifiers 118 and delay circuits 120 of the beamformer 106. In some embodiments, thefrequency generator 110 is integrated with the beamformer 106. The radio frequency generator 110 and the beamformer 106 are configured to drive the individual transducer elements 104 of the transducer array 102 at the same frequency, but at different phases and / or different amplitudes.

[0020] The amplification or attenuation factors oq-anand the phase shifts ax-animposed by the beamformer 106 serve to transmit and focus ultrasonic energy through the intervening tissue located between the transducer elements 104 and the target region onto the target region 101, and account for wave distortions induced in the intervening tissue. The amplification factors and phase shifts are computed using the controller 108, which may provide the computational functions through software, hardware, firmware, hardwiring, or any combination thereof. In various embodiments, the controller 108 utilizes a general- purpose or special-purpose digital data processor programmed with software in a conventional manner, and without undue experimentation, to determine the frequency, phase shifts and / or amplification factors necessary to obtain a desired focus or any other desired spatial field patterns at the target region 101. In certain embodiments, the computation is based on detailed information about the characteristics (e.g., the type, size, location, property, structure, thickness, density, structure, etc.) of the intervening tissue located between the transducer element 104 and the target and their effects on propagation of acoustic energy. Such information may be obtained from an imager 112. The imager 112 may be, for example, a magnetic resonance imaging (MRI) device, a computer tomography (CT) device, a positron emission tomography (PET) device, a single-photon emission computed tomography (SPECT) device, or an ultrasonography device. Image acquisition may be three-dimensional (3D) or, alternatively, the imager 112 may provide a set of two-dimensional (2D) images suitable for reconstructing a three-dimensional image of the target region 101 and / or other regions (e.g., the region surrounding the target 101 or another target region). Imagemanipulation functionality may be implemented in the imager 112, in the controller 108, or in a separate device. In addition, the ultrasound system 100 and / or imager 112 may be utilized to detect signals from an acoustic reflector (e.g., microbubbles 202, see Figure 2) located substantially close to the target region 101 as further described below. Additionally or alternatively, the system 100 may include an acoustic-signal detection device (such as a hydrophone or suitable alternative) 124 that detects transmitted or reflected ultrasound from the acoustic reflector, and which may provide the signals it receives to the controller 108 for further processing. In addition, the ultrasound system 100 may include an administrationsystem 126 for parenterally introducing the acoustic reflector into the patient’s body. The imager 112, the acoustic-signal detection device 124, and / or the administration system 126 may be operated using the same controller 108 that facilitates the transducer operation; alternatively, they may be separately controlled by one or more separate controllers intercommunicating with one another.

[0021] Figure IB illustrates an exemplary imager - namely, an MRI apparatus 112. The apparatus 112 may include a cylindrical electromagnet 134, which generates the requisite static magnetic field within a bore 136 of the electromagnet 134. During medical procedures, a patient is placed inside the bore 136 on a movable support table 138. A region of interest 140 within the patient (e.g., the patient’s head) may be positioned within an imaging region 142 wherein the electromagnet 134 generates a substantially homogeneous field. A set of cylindrical magnetic field gradient coils 144 may also be provided within the bore 136 and surrounding the patient. The gradient coils 144 generate magnetic field gradients of predetermined magnitudes, at predetermined times, and in three mutually orthogonal directions. With the field gradients, different spatial locations can be associated with different precession frequencies, thereby giving a magnetic-resonance (MR) image its spatial resolution. An RF transmitter coil 146 surrounding the imaging region 142 emits RF pulses into the imaging region 142 to cause the patient’s tissues to emit MR response signals. Raw MR response signals are sensed by the RF coil 146 and passed to an MR controller 148 that then computes an MR image, which may be displayed to the user. Alternatively, separate MR transmitter and receiver coils may be used. Images acquired using the MRI apparatus 112 may provide radiologists and physicians with a visual contrast between different tissues and detailed internal views of a patient’s anatomy that cannot be visualized with conventional x- ray technology.

[0022] The MRI controller 148 may control the pulse sequence, i.e., the relative timing and strengths of the magnetic field gradients and the RF excitation pulses and response detection periods. The MR response signals are amplified, conditioned, and digitized into raw data using a conventional image-processing system, and further transformed into arrays of image data by methods known to those of ordinary skill in the art. Based on the image data, the target region (e.g., a tumor or a target BBB) can be identified.

[0023] To perform targeted drug delivery or tumor ablation, it is necessary to determine the location of the target region 101 with high precision. Accordingly, in various embodiments, the imager 112 is first activated to acquire images of the target region 101and / or non-target region (e.g., the healthy tissue surrounding the target region, the intervening tissue located between the transducer array 102 and the target region 101 and / or any regions located near the target) and, based thereon, determine anatomical characteristics (e.g., the tissue type, location, size, thickness, density, structure, shape, vascularization) associated therewith. For example, a tissue volume may be represented as a 3D set of voxels based on a 3D image or a series of 2D image slices and may include the target region 101 and / or nontarget region.

[0024] To create a high-quality focus at the target region 101, it may be necessary to calibrate the transducer elements 104 and take into account transducer geometric imperfections resulting from, for example, movement, shifts and / or deformation of the transducer elements 104 from their expected locations. In addition, because the ultrasound waves may be scattered, absorbed, reflected and / or refracted when traveling through inhomogeneous intervening tissues located between the transducer elements 104 and the target region 101, accounting for these wave distortions may also be necessary in order to improve the focusing properties at the target region 101.

[0025] Referring to Figure 2, ultrasound waves transmitted from all (or at least some) transducer elements 104 are reflected by the acoustic reflectors 202. The acoustic reflectors 202 may consist essentially of microbubbles generated by the ultrasound waves and / or introduced parenterally by an administration system. In some embodiments, the administration system 126 introduces a seed microbubble into the target region 101; the transducer 102 is then activated to transmit ultrasound waves to the seed microbubble for generating a cloud of microbubbles. Approaches to generating the microbubbles and / or introducing the microbubbles to the target region 101 are provided, for example, in PCT Publication No. WO 2018 / 020315, PCT Application Nos. PCT / US2018 / 064058 (filed on December 5, 2018), PCT / IB2018 / 001103 (filed on August 14, 2018), PCT / US2018 / 064892 (filed on December 11, 2018), PCT / IB2018 / 000841 (filed on June 29, 2018), and PCT / US2018 / 064066 (filed on December 5, 2018), U.S. Patent Publication No.2019 / 0083065, and U.S. Patent Application No. 15 / 837,392 (filed on December 11, 2017), the contents of which are incorporated herein by reference.

[0026] FUS is currently being applied in the clinic to noninvasively treat CNS-related pathologies. FUS-mediated BBB opening is one of the most interesting FUS modalities that is extensively explored in clinical trials to increase and improve drug delivery into the brain parenchyma. However, disrupting the BBB excessively may also cause damage to the braintissue. In order to ensure the safety of the procedure while achieving the desired degree BBB opening for drug delivery, FUS sonication pattern control is required.

[0027] Figures 3-8 describe the use of a real-time controller based on acoustic spectral information acquired during treatment containing an algorithm trained on data from previous treatments correlating Gd measurements in the brain parenchyma and the acoustic spectrogram obtained from the same treated region.

[0028] Current controllers that use acoustic data in the field of BBB opening are based on specific harmonics or sub / ultra harmonics obtained from the tissue. Some use a combination of information from several spectral bands for controlling treatment rate. However, there is a wide variability in the data obtained between different patients, as well as different regions of the brain, which results in incorrect predictions of BBB opening in some instances.

[0029] Some techniques may analyze the acoustic feedback during treatment based on one or more bands of interest. A band is a narrow continuous part in the spectrum that is related to a specific area of interest. Usually, frequencies around (e.g., + / - 3%) harmonics of the transmission frequency and also the frequencies around half of the transmission frequency may be considered bands. In some cases, the frequencies between bands may be referred to as broad band frequencies, and some may be considered bands as well. For each band, some techniques may accumulate the signal measured during treatment and control the treatment based on the information gathered for each band independently. One of the novel aspects of this disclosure includes exploiting the information regarding concurrent relationships between the bands along the sonication in addition to the information used in the aforementioned techniques. Another of the novel aspects of this disclosure includes the use of an algorithm trained on one or more spectrograms obtained experimentally from the target regions, improving the confidence in the measured BBB opening calculated in real time. As used in this disclosure, the term “spectrogram” refers to a measurement of information of frequencies that contains frequencies from two or more bands.

[0030] In one embodiment, the proposed controller works in the following way: First, the target region is sonicated (optionally with microbubbles). Next, the acoustic spectrum from the target region is acquired by the receivers in the transducer (same as transmitters or different ones). Then, the controller computationally calculates what will be the predicted BBB opening level in the target region at the end of the sonication, based on the acoustic spectrum using an embedded algorithm, described below. Alternatively, the predicted BBBopening level can be calculated from the dynamics of the one or more acoustic spectrograms between two separate sonications at the same target region. Finally, based at least in part on at least one of the calculated predicted BBB opening levels, the controller instructs the transducer to: (i) increase power if the calculated predicted BBB opening and / or treatment rate is below the desired opening for that specific treatment, (ii) maintain power if the calculated BBB treatment rate equals the desired opening for that specific treatment, (iii) decrease power or stop treatment if calculated BBB opening and / or treatment rate is higher than the desired opening for that specific treatment, or (iv) finish the sonication if the calculated BBB opening level equals the desired BBB opening for that specific treatment. In some cases, over treatment might cause lowering the opening level and therefore sometimes treatment will be stopped even though the prediction suggests that target opening level was not achieved. The aforementioned algorithm can be a neural network developed and trained using both the acoustic spectrogram and the measured gadolinium (Gd) concentration acquired from patients’ brains during past treatments.

[0031] The disclosed system uses the whole acoustic spectrogram to predict BBB opening, rather than specific harmonics or sub / ultra-harmonics acquired from the target region. This approach improves the accuracy and reliability of the BBB opening prediction, enabling better control of the BBB opening treatment. This system can be used to control BBB opening and other acoustic effects for a variety of treatments such as drug delivery for tumors, neurodegenerative diseases, neurologic disorders, neuropsychiatric disorders, CNS infections, enzyme replacement, and gene therapies. This system can also be used to control BBB opening in liquid biopsies. It can also be used to control other acoustic effects. For example, in neuromodulation acoustic sonication may be used without microbubbles for scenarios in which the brain is subject to acoustic treatment without the need for opening the BBB.

[0032] Thus, the present disclosure provides systems and methods for opening the BBB using a controller that is based on acoustic data obtained from ultrasonic receivers that capture the oscillations of microbubbles in the treatment area. The system includes a controller to determine parameters for opening the BBB based on one or more spectrograms generated from the acoustic data, processed by an algorithm described in more detail below.

[0033] The one or more spectrograms can be built by one or more of the following: pulses' spectra of at least one receiver, pulses' spectra of a plurality of receivers, pulses' weighted sum of spectra of X receivers, pulses' spectra which are compensated by skulltransmission (either continuous or discrete) of different acoustic frequencies, and / or pulses' spectra obtained by beamforming of a plurality of receivers signals in time domain and application of Fourier transform on these signals, after beamforming.

[0034] Figure 3 depicts a process for obtaining a spectrogram during a focused ultrasound procedure in accordance with some implementations. A spectrogram is a representation of the signal strength of a signal over time at various frequencies present in a particular waveform. Here, the waveform is the acoustic response (also referred to as acoustic feedback) of the bubbles in the target region 101. Not only does a spectrogram show whether there is more or less energy at different frequencies, but it also shows how energy levels vary over time. The spectrogram can also show the combination of frequencies presented in each time point. The pattern of combinations of frequencies and the changes of those patterns along time governs the treatment results.

[0035] Referring to Figure 3, each of a plurality of sub-spots 303 in a treatment area 302 of the target region 101 is treated with a sequence of acoustic FUS pulses. Each pulse of the sequence of pulses can be characterized according to a pulse spectrum 304, and each pulse spectrum 304 is compiled into a spectrogram 306. Each column of the spectrogram 306 represents one pulse spectrum 304. As treatment continues, successive pulses result in the accumulation of more pulse spectrums, which causes the spectrogram 306 to grow. In some implementations, one or more pulse spectrums 304 may contain only one or more parts of the measured pulse spectrum. For example, in some implementations, the system may be prone to measurement artifacts in some frequencies and therefore some areas may be excluded from the spectrogram.

[0036] Figure 4 depicts a process for predicting and controlling BBB opening levels based on one or more spectrograms obtained during a focused ultrasound procedure in accordance with some implementations. Spectral data 402 accumulates in real time during a treatment procedure. Spectrogram data 404 accumulates from the beginning of the treatment up to the real-time point of the treatment, and spectrogram data 406 represents the most recently obtained portion of data 404, obtained from the last n pulses. Accumulative spectrogram data 404 and / or local spectrogram data 406 are used by a spectrogram analysis module 410 to calculate total and / or local BBB opening level predictions, respectively. The controller 420 uses a feedback-based control loop logic (e.g., proportional-integral-derivative (PID) logic, or any other feedback-based control loop logic) to adjust acoustic treatment parameters in real-time based on the total and local predictions to maintain the BBB openinglevel on a specific value. For example, a feedback-based controller may increase treatment power by a certain amount (e.g., by 1%) when a desired affect has not yet been achieved, decrease the treatment power by a certain amount (e.g., by 2%) when the predicted affect is too strong with respect to the desired effect, and so forth.

[0037] In some implementations, the spectrogram analysis module 410 uses one or more algorithms trained by one or more neural networks to analyze spectrogram data 404 and / or 406 to obtain predictions for BBB opening levels. More details regarding how such neural network(s) may be trained and used to provide such predictions are provided below with reference to Figures 5-6.

[0038] In some implementations, the spectrogram analysis module 410 uses one or more algorithms other than neural networks to analyze spectrogram data 404 and / or 406 to obtain predictions for BBB opening levels. For instance, in some implementations, pre-determined parameter matrices of the same size as the spectrograms may be used, which can be multiplied and utilized to calculate the prediction. One way to utilize such matrices is as follows: Aij *Bij=Cij . sum(Cij)=(BBB Opening prediction). Here, matrix A represents the spectrogram, matrix B is the pre-determined parameter matrix, and the sum of Cij results in the BBB opening prediction.

[0039] In addition to or as an alternative to predicting BBB opening levels, the spectrogram analysis module 410 can analyze spectrogram data 404 and / or 406 to obtain predictions for treatment risk and / or BBB closing time. The neural network or matrix inputs can additionally or alternatively include the planned sonication duration, the number of the performed treatments to specific treatment area or patient, tissue characterization (e.g. by MRI), and / or real-time MRI, like T2* and fMRI.

[0040] Figure 5 depicts training data preparation for use in BBB opening level predictions in accordance with some implementations. In some embodiments, the controller 420 is implemented at least in part by an artificial intelligence (Al) network-trained spectrogram analysis module 410. The spectrogram analysis module 410 of the controller may be trained on data of total accumulative spectra of defined sonication time. Specifically, the inputs are spectrogram data, and the outputs (also referred to as input labels) are BBB opening levels, which are obtained by analysis of MRI images. The MRI analysis for obtaining the BBB opening level that corresponds to a given spectrogram may be based on MR sequences from which T1 relaxation times were extracted from pre- and post-treatmentimages with and without Gadolinium (Gd) injection, to calculate the 1 / T1=R1 values. After the extraction of these values, Delta R1 volumes are calculated, from which Gd concentration changes (which represent the BBB opening levels) in the treatment volumes are calculated. Additionally, the analysis for obtaining the BBB opening level can be based on MR T1 qualitative images obtained pre- and post-treatment, with and without Gadolinium injection. The without-Gadolinium T1 images undergo N4 bias analysis for Bl and receptive field correction, which is used to correct Bl and receptive field of with-Gadolinium injection volumes. This is achieved by division of the with-Gadolinium and the without-Gadolinium N4 bias uncorrected volumes and multiplication of this division by the corrected N4 bias without-Gadolinium volume. After the N4 bias analysis, the with- and without-Gadolinium volumes undergo white-stripe normalization analysis using a segmentation cube (e.g., a 4x4x4 cm3segmentation cube, or any other size segmentation cube larger or smaller than 4x4x4 cm3) in the center of the brain (or center of the brain ventricles), from which its histogram white matter parameters are extracted (white matter peak and standard deviation). The volumes are then subtracted by the white matter peak value and then divided by the standard deviation value. Once these analyses are performed, the analyzed volumes are subtracted to unravel the Gadolinium changes in the treatment region (which represent the BBB opening level).

[0041] Referring back to Figure 4, the spectrograms 404 and / or 406 may be used for prediction of treatment results (e.g., predicting BBB opening level). In some implementations, the original measured spectrogram may be processed (e.g., stretched) in order to compare partial measurements (e.g., measurements from part of the sonication) to results learned from “full” measurements (e.g., measurements gathered along a full sonication). In some embodiments, the processing turns the measurement into an image 402 that can be fed into neural network. The processing method can use interpolation and extrapolation to add missing measurements. For example, accumulative spectrogram 404 obtained from the beginning of treatment to a current point of treatment can be augmented to include extrapolated spectrogram data 408 representing data yet to be obtained from the current point of treatment to the end of treatment. Further, local spectrogram 406 obtained from the last n pulses of the treatment can be extrapolated backwards to represent data going back to the beginning of treatment (the beginning of 404), and forward to represent data going to the end of treatment (408). In some implementations, the extrapolation can be doneby reusing one or more measurements acquired in a constellation which is similar to the constellation expected in the rest of the sonication (e.g. the current power).

[0042] Figure 6 depicts a schematic diagram of a process for predicting and controlling BBB opening levels during a focused ultrasound procedure in accordance with some implementations. In this example, the feedback-based control loop is implemented as a PID controller. In general, however, any feedback-based control implementation can be used without departing from the inventive concepts described herein. For example, a feedbackbased controller can be implemented to increase a treatment parameter by a certain amount (e.g., increase power by 1%) when the desired affect has not yet been achieved, decrease the treatment parameter by a certain amount (e.g., decrease power by 2%) when the predicted affect is too strong with respect to the desired effect, and so forth.

[0043] A sensor array captures acoustic feedback (e.g., pulse spectra 304 in Figure 3) generated by the acoustic pulses that were transmitted during treatment (e.g., at sub-spots 303 in Figure 3). The controller (e.g., 420, Figure 4) receives the acoustic feedback and generates one or two spectrograms: an accumulated spectrogram (e.g., 404, Figure 4) and / or a local spectrogram (e.g., 406, Figure 4). The controller extrapolates each of the one or two spectrograms to extend the data to a dataset representing an entire treatment from beginning to end (e.g., 404+408, Figure 4), and each extrapolated spectrogram is processed by the neural network (e.g., by spectrogram analysis module 410, Figure 4). Stated another way, the neural network applies an algorithm to each extrapolated spectrogram to predict a corresponding BBB opening level as described above with reference to Figures 4-5.Specifically, the neural network applies the algorithm to the accumulative spectrogram of all N pulses of the treatment from beginning to current point and extended with extrapolated data to obtain a total predicted BBB opening level, and / or the neural network applies the algorithm to the local spectrogram of the last n pulses of the treatment from a time after the beginning to the current point and extended with extrapolated data to obtain a local predicted BBB opening level.

[0044] The total and / or local predicted BBB opening levels are then compared to the target BBB opening level. The comparison of the total predicted BBB opening level to the target BBB opening level yields first comparison data, sometimes referred to as position data, as it describes the current difference between target and predicted levels. The comparison of the local predicted BBB opening level to the target BBB opening level yields second comparison data, sometimes referred to as velocity data, as it describes more recent changesin the difference between target and predicted levels. The first and second comparison data may be weighted differently. Since the local spectrogram data represents more recent data overall than the accumulated spectrogram data, the second comparison data may be weighted higher when compared to the target BBB opening level. The first comparison data is assigned weight KP 1, and the second comparison data is assigned weight KP 2.

[0045] In some implementations, the target BBB opening level is compared once: either to the predicted BBB opening level associated with the local spectrogram of the last n pulses (for a recent subset of the treatment) after extrapolation, or to the predicted BBB opening level associated with the accumulated spectrogram of all N pulses (for the whole treatment up to the current point) after extrapolation.

[0046] In some implementations, the target BBB opening level is compared twice: first to the predicted BBB opening level associated with the local spectrogram of the last n pulses (for a recent subset of the treatment) after extrapolation, and second to the predicted BBB opening level associated with the accumulated spectrogram of all N pulses (for the whole treatment up to the current point) after extrapolation.

[0047] The total and / or local (first and / or second) comparison data are then compared. As the predicted BBB opening levels get closer to the target BBB opening level, the first and second comparison data should be very close. Thus, the smaller the result of this comparison step, the closer the predicted BBB opening levels are to the target BBB opening level. Based on a result of this comparison, a feedback-based control loop (e.g., PID) feature of the controller assigns a proportional weight KP 3, an integral weight KI 3, and a derivative weight KD 3to the result, and adjusts a power level P(n) (or any other acoustic parameter value) based on those weights. The FUS system 100 uses the adjusted power level (or other parameter level) as to successively transmit more acoustic pulses while the treatment continues. The additional acoustic pulses generate additional acoustic feedback (accumulative and / or local spectrogram data) and the process repeats. As the process continues to repeat, the system stabilizes as in the example sonication graph in Figure 4.

[0048] Thus, the process described with reference to Figure 6 may use either or both of the two branches (the local spectrogram analysis and prediction branch on the bottom, and / or the accumulative analysis and prediction branch on the top) to predict BBB opening levels (or any other acoustic treatment effect) in order to determine changes to an acoustic treatment power (or to any other acoustic treatment parameter) for optimizing the treatment effect.

[0049] Usually, spectrograms span all of the frequencies up to the half of the sampling rate. However, in some implementations, some parts of the spectrogram might be ignored (e.g., the area in the spectrogram that is influenced by the transmission frequency which might be not relevant to the measurement and treatment control). In other implementations, one or more areas of the spectrogram that were found to be dominant in prediction of the treatment results may be selected to represent the entire spectrogram.

[0050] In some implementations, different parts (ranges) of the spectrograms may be selected for prediction of safety, efficacy and / or deterioration in the BBB opening. The ranges might overlap with each other.

[0051] In some implementations, the one or more spectrograms may be measured and processed differently by pixel in the treated volume. The differentiation may be related to different transmission of acoustic data from a pixel / area to the measurement system. In some implementations, beamforming methods may be used to assign calculated spectrograms to pixel / area. In some implementations, the target area may include tissue that should be treated and tissue that should not be treated. In these implementations, one pixel may represent information from tissue that should be treated, and another pixel may represent information from tissue that should not be treated. In other implementations, the pixel data facilitates performing a specific treatment level for specific tissues within the target area. In other implementations, pixel data is used to assure uniform treatment. In other implementations, pixel data can be used to assure that acoustic feedback comes from target area.

[0052] In some implementations, the one or more spectrograms may be used to predict BBB opening level, risk to tissue or patient, deterioration of the BBB opening due to over treatment, and / or closing time of the BBB opening (e.g., getting to a permeability level that is not significantly different from the base line or below a certain level).

[0053] There may be a need for different parameters for the algorithm for different constellations. Parameters can be a function of planned sonication duration, tissue type, number of previous treatments, tissue characterization (e.g. by MRI), and / or disease (e.g. patients with AD are known to have a more delicate vascular system).

[0054] When neural networks are used to process the one or more spectrograms, the networks can be fed by a combination of spectrogram and real-time MRI. Predictor training may be done on full treatment data of total spectrograms vs. either MRI-based BBB opening prediction or measurement of markers in the treated tissue.

[0055] The system presented in this disclosure allows for the following acoustic-based dynamics of the controller: (a) analyzing spectrogram data (e.g., using an Al network trained on full treatment data of total spectrograms vs. MRI-based BBB opening prediction, or using one or more non-AI algorithms based on, for example, pre-determined parameter matrices of the same size as the spectrograms) to control dynamic the BBB opening process or to control any other acoustic treatment parameter and corresponding result; (b) obtaining stable acoustic fingerprints of components of different spectrograms to obtain predicted BBB opening- related processes over different treatment times; (c) using accumulated spectrum and local spectrum to stabilize the PID controller, via temporary Al networks predictions, when the accumulated spectrum temporary prediction is equivalent to “position” and local spectrum temporary prediction is equivalent to “velocity”; (d) using an Al network, which was trained on full treatment data of total spectrograms vs. MRI-based BBB prediction process (opening level, safety, BBB closing times), to control dynamic BBB opening process (opening level, safety, BBB closing times); (e) affect the controller response, according to the presence of unique spectral or acoustic power dynamic components; and / or (f) working with beamformed signals.

[0056] The inputs discussed above can also include information of: MRI-related tissue type information, MRI-related hemodynamic response signal (fMRI), neural activity according to EEG signal, CT related structural information, and / or PET related metabolic information.

[0057] The controller output is mainly the sonication power; however, alternatively or additionally the controller output may be changes in pulse duration, pulse repetition frequency, acoustic pulse frequency, control bubble concentration, bubble type, and / or drug administration.

[0058] In some implementations, personalized treatment planning inputs may include tissue type, age, medical history, genetic profile, number of performed treatments, heart rate, blood pressure, and / or blood flow rate measured by acoustic feedback.

[0059] In some implementations, as an alternative to Al, the system presented in this disclosure may use multivariate regression between past spectrograms and corresponding BBB openings to real-time BBB opening levels that correspond to currently accumulating spectrograms. For example, pre-determined parameter matrices of the same size as thespectrograms may be used as described above with reference to the spectrogram analysis module 410.

[0060] The system presented in this disclosure can be used for drug delivery to the brain to treat ailments including but not restricted to tumors, neurodegenerative diseases (such as Parkinson’s, Alzheimer’s and other Tauopathies), genetic diseases and brain infections, as well as for diagnostic (example, liquid biopsy indications) and therapeutic purposes, such as neuromodulation and microbubble-enhanced therapeutic tissue death.

[0061] In some embodiments, the system includes a device to adjust infusion rate responsive to the controller. In some embodiments, the device adjusts the infusion rate of microbubbles. In other embodiments, the device adjusts the infusion rate of therapeutic agents. In other embodiments, the device adjusts the infusion rate of both microbubbles and of therapeutic agents. In all embodiments, adjust can refer to increasing, decreasing, maintaining, or halting the infusion rate after each sonication.

[0062] In other embodiments, the real-time control over the BBB opening level provided by the control system presented in this disclosure increases the BBB opening treatment accuracy, while reducing the risk of over- or under-treatment and minimizing the potential for damage to the brain tissue.

[0063] In other embodiments, the control system presented in this disclosure is not limited to the field of FUS. It can be applied to any real-time dynamic process that involves sensors (process inputs) for monitoring. The system uses Al training or other types of algorithms to correlate process inputs to outputs, allowing the same principles as presented before to be used to control the process. This involves using accumulated and local (spatial and / or temporal) sensor data with time / space interpolations and / or extrapolation as input to the trained algorithm to predict the current state of the process. The PID controller is then use this prediction information to adjust the physical parameters and control the process in realtime. By doing so, the output of the controlled process will reflect the real process output at the end of the process period for which the training was carried out. Examples of such processes could include:

[0064] (1) HVAC systems: Heating, ventilation, and air conditioning (HVAC) systems are critical for maintaining comfortable and healthy indoor environments. By using sensors to monitor temperature, humidity, and air quality, the innovative control system could be used to adjust the physical parameters of the HVAC system in real-time. This could includeadjusting the temperature setpoint, fan speed, or air flow rate to maintain optimal indoor conditions.

[0065] (2) Manufacturing processes: Many manufacturing processes involve complex machinery and equipment that require precise control to ensure high-quality output. By using sensors to monitor various process parameters, such as temperature, pressure, and flow rate, the innovative control system could be used to adjust the physical parameters of the equipment in real-time. This could include adjusting the speed of a conveyor belt, the temperature of a furnace, or the flow rate of a chemical reaction to maintain optimal process conditions and ensure consistent product quality.

[0066] (3) Acoustic underwater communication: Underwater communication is a critical aspect of many industries, including offshore oil and gas, marine research, and underwater exploration. Acoustic communication is often used in these applications due to its ability to transmit signals over long distances in water. By using sensors to monitor the acoustic signals and the surrounding environment, the innovative control system could be used to adjust the physical parameters of the acoustic communication system in real-time. This could include adjusting the frequency, amplitude, or pulse duration of the acoustic signals to maintain optimal communication conditions and ensure reliable data transmission.

[0067] Figure 7 depicts results of Al-based BBB opening prediction in accordance the techniques described above. These results were obtained from human data analysis. While current methods (graph on the left) that use clinical acoustic doses to determine BBB opening levels have a weak correlation between the input (mean acoustic dose) and the output (mean Gd quantity as observed in MRI data), the system presented in this disclosure (graph on the right) that use Al predictions based on spectrograms has a relatively strong correlation between the input (mean Gd quantity as predicted) and the output (mean Gd quantity as observed in MRI data).

[0068] Figure 8 depicts results of Al-based BBB opening prediction in accordance with the techniques described above. While these results were obtained from rat data analysis, the corresponding techniques equally apply to human data analysis. While current methods (graph in the lower right) that use clinical acoustic dose to determine BBB opening levels have a weak correlation between the input (acoustic dose) and the output (Gd quantity as observed in MRI data), the system presented in this disclosure (graph in the upper right) that use Al predictions based on spectrograms has a relatively strong correlation between theinput (Gd quantity as predicted) and the output (Gd quantity as observed in MRI data). In addition, referring to the charts in the lower left, by predicting the BBB opening levels in real-time during treatment, the power output level over time becomes steady as the treatment attains a desired BBB opening level at a desired rate.

[0069] The following section describes several embodiments of the systems and methods described above in the form of clauses.

[0070] Clause 1 : A system for controllably changing a tissue property in a target volume, the system comprising: an ultrasound transducer configured to transmit a sequence of acoustic pulses (e.g., 303) to the target volume to change the tissue property; a sensor array configured to capture acoustic feedback (e.g., 304) generated by the sequence of acoustic pulses (e.g., in the target volume or outside the target volume); and a controller configured to: (i) generate a spectrogram (e.g., 306) using two or more bands of frequencies included in the acoustic feedback; (ii) computationally calculate a degree of change of the tissue property in the target volume (e.g., predict a BBB opening level) based on the spectrogram, and (iii) control the ultrasound transducer (e.g., adjust an output power level) based on the calculated degree of change of the tissue property to change the tissue property to a target degree set for a specific treatment (e.g., to reach a desired BBB opening level). In some implementations, the controller generates one or more spectrograms (e.g., accumulative spectrogram 404 and / or local spectrogram 406) and proceeds with steps (ii) and (iii) based on each of the one or more spectrograms.

[0071] Clause 2: The system of clause 1, wherein the spectrogram includes acoustic feedback captured from a beginning of treatment to a current point of treatment (e.g., accumulative spectrogram 404).

[0072] Clause 3: The system of clause 1, wherein the spectrogram includes a subset less than all of most recently captured acoustic feedback (e.g., local spectrogram 406).

[0073] Clause 4: The system of any of clauses 1-3, wherein the spectrogram includes extrapolated acoustic data (e.g., 408) extending the acoustic feedback from a current point of treatment to an end of treatment.

[0074] Clause 5: The system of any of clauses 1-4, wherein the spectrogram includes at least one of: spectra of acoustic feedback captured by at least one receiver, a weighted sum of spectra of acoustic feedback captured by a plurality of receivers, spectra of acoustic feedback that are compensated by skull transmission of different acoustic frequencies, or spectraobtained by beamforming of a plurality of receiver signals in a time domain and application of a Fourier transform of the plurality of receiver signals after beamforming.

[0075] Clause 6: The system of any of clauses 1-5, wherein the spectrogram includes spectra of a plurality of frequency ranges used for safety, efficiency, and deterioration of changes in tissue property.

[0076] Clause 7: The system of any of clauses 1-6, wherein the controller is configured to calculate the degree of change based on outputs of a neural network trained on total accumulative spectra of defined sonication time.

[0077] Clause 8: The system of clause 7, wherein outputs of the neural network include degree of change of tissue property, treatment risk, or deterioration of the degree of change of tissue property.

[0078] Clause 9: The system of any of clauses 1-8, wherein the controller is configured to control the ultrasound transducer by adjusting one or more treatment parameters in real time to maintain the target degree of change, wherein the one or more treatment parameters includes sonication power, sonication duration, sonication frequency, microbubble concentration, microbubble infusion rate, or therapeutic agent infusion rate.

[0079] Clause 10: The system of any of clauses 1-9, wherein the change to the tissue property is: a disruption of a tissue barrier; a neuromodulation of tissue neurons; an activation of a sonodynamic therapy drug; an activation of contrast agent carriers for drug and / or gene delivery; a thrombolysis; an indication of tissue death and / or an induction of an ischemic effect.

[0080] Clause 11 : The system of any of clauses 1-10, wherein the change to the tissue property is a disruption of a tissue barrier to increase permeability of the barrier, wherein the tissue barrier is a blood-brain barrier, a blood-retina barrier, skin, a mucosal membrane, a cell membrane, or a nuclear membrane.

[0081] Clause 12: The system of clause 11, wherein the permeability is increased sufficiently to permit passage of a therapeutic agent therethrough, wherein the therapeutic agent is selected for treatment of a tumor, a neurogenerative disease, a neurologic disorder, a neuropsychiatric disorder, an enzymatic deficiency, or a CNS infection.

[0082] Clause 13: The system of clause 11, wherein the permeability is increased sufficiently to permit passage of a biomarker from within brain tissue to one or more bloodvessels, enabling or enhancing detection through liquid biopsy and / or another diagnostic technique.

[0083] Clause 14: A method of controllably changing a tissue property in a target volume, the method comprising: applying a sequence of acoustic pulses (e.g., 303) to the target volume to change the tissue property; detecting acoustic feedback (e.g., 304) generated by the sequence of acoustic pulse (e.g., in the target volume or outside the target volume); generating a spectrogram (e.g., 306) using two or more bands of frequencies included in the acoustic feedback; computationally calculating a degree of change of the tissue property in the target volume (e.g., predict a BBB opening level) based on the spectrogram, and controlling application of the acoustic pulses (e.g., adjust an output power level) based on the calculated degree of change of the tissue property to change the tissue property to a target degree set for a specific treatment (e.g., to reach a desired BBB opening level). In some implementations, one or more spectrograms are generated (e.g., accumulative spectrogram 404 and / or local spectrogram 406) in step (i), and the method proceeds with steps (ii) and (iii) based on each of the one or more spectrograms.

[0084] Clause 15: The method of clause 14, wherein the spectrogram includes acoustic feedback captured from a beginning of treatment to a current point of treatment (e.g., accumulative spectrogram 404).

[0085] Clause 16: The method of clause 14, wherein the spectrogram includes a subset less than all of most recently captured acoustic feedback (e.g., local spectrogram 406).

[0086] Clause 17: The method of any of clauses 14-16, wherein the spectrogram includes extrapolated acoustic data (e.g., 408) extending the acoustic feedback from a current point of treatment to an end of treatment.

[0087] Clause 18: The method of any of clauses 14-17, wherein the spectrogram includes at least one of: spectra of acoustic feedback captured by at least one receiver, a weighted sum of spectra of acoustic feedback captured by a plurality of receivers, spectra of acoustic feedback that are compensated by skull transmission of different acoustic frequencies, or spectra obtained by beamforming of a plurality of receiver signals in a time domain and application of a Fourier transform of the plurality of receiver signals after beamforming.

[0088] Clause 19: The method of any of clauses 14-18, wherein the spectrogram includes spectra of a plurality of frequency ranges used for safety, efficiency, and deterioration of changes in tissue property.

[0089] Clause 20: The method of any of clauses 14-19, wherein the controller is configured to calculate the degree of change based on outputs of a neural network trained on total accumulative spectra of defined sonication time.

[0090] Clause 21 : The method of clause 20, wherein outputs of the neural network include degree of change of tissue property, treatment risk, or deterioration of the degree of change of tissue property.

[0091] Clause 22: The method of any of clauses 14-21, wherein the controller is configured to control the ultrasound transducer by adjusting one or more treatment parameters in real time to maintain the target degree of change, wherein the one or more treatment parameters includes sonication power, sonication duration, sonication frequency, microbubble concentration, microbubble infusion rate, or therapeutic agent infusion rate.

[0092] Clause 23: The method of any of clauses 14-22, wherein the change to the tissue property is: a disruption of a tissue barrier; a neuromodulation of tissue neurons; an activation of a sonodynamic therapy drug; an activation of contrast agent carriers for drug and / or gene delivery; a thrombolysis; an indication of tissue death and / or an induction of an ischemic effect.

[0093] Clause 24: The method of any of clauses 14-23, wherein the change to the tissue property is a disruption of a tissue barrier to increase permeability of the barrier, wherein the tissue barrier is a blood-brain barrier, a blood-retina barrier, skin, a mucosal membrane, a cell membrane, or a nuclear membrane.

[0094] Clause 25: The method of clause 24, wherein the permeability is increased sufficiently to permit passage of a therapeutic agent therethrough, wherein the therapeutic agent is selected for treatment of a tumor, a neurogenerative disease, a neurologic disorder, a neuropsychiatric disorder, an enzymatic deficiency, or a CNS infection.

[0095] Clause 26: The method of clause 24, wherein the permeability is increased sufficiently to permit passage of a biomarker from within brain tissue to one or more blood vessels, enabling or enhancing detection through liquid biopsy and / or another diagnostic technique.

[0096] Clause 27: A method for opening a blood-brain-barrier (BBB) during treatment, the method comprising: at a controller communicatively coupled to an ultrasound transducer configured to emit ultrasound waves and a sensor array configured to capture acoustic feedback: (a) obtaining acoustic feedback (e.g., 304) from the sensor array; (b) generating (i)a first spectrogram (e.g., 404) from acoustic feedback accumulated from a beginning of the treatment up to a current point of the treatment, and (ii) a second spectrogram (e.g., 406) from acoustic feedback selected from a subset the treatment leading up to the current point of the treatment; (c) determining a total predicted BBB opening level from the first spectrogram (e.g., Figure 6 top branch) and a local predicted BBB opening level from the second spectrogram (e.g., Figure 6 bottom branch); (d) determining an adjusted parameter (e.g., output power) for opening the BBB based on (i) a difference between a target BBB opening level and the total predicted BBB opening level (e.g., Figure 6, position), and (ii) a difference between the target BBB opening level and the local predicted BBB opening level (e.g., Figure 6, velocity); and (e) causing the ultrasound transducer to emit ultrasound waves using the adjusted parameter to open the BBB (e.g., the target BBB opening level at a desired BBB opening rate).

[0097] Clause 28: The method of clause 27, further comprising, subsequent to causing the ultrasound transducer to emit the ultrasound waves: (f) monitoring a current BBB opening level using the sensor array to obtain additional acoustic feedback; and (g) restarting the method at step (a) using the additional acoustic feedback.

[0098] Clause 29: The method of clause 27 or clause 28, wherein determining the adjusted parameter in step (d) includes using proportional-integral-derivative (PID) logic to determine the adjusted parameter in real-time based on the total and local predictions.

[0099] Clause 30: The method of any of clauses 27-29, wherein obtaining the acoustic feedback in step (a) includes capturing oscillations of microbubbles in a treatment area.

[0100] Clause 31 : The method of any of clauses 27-30, further comprising causing a drug delivery system to deliver a drug through the opened BBB.

[0101] Clause 32: The method of any of clauses 27-31, wherein determining the total and local predicted BBB opening levels in step (c) includes using a machine learning algorithm trained on spectrograms generated from acoustic feedback.

[0102] Clause 33: The method of clause 32, wherein using the machine learning algorithm includes extrapolating the acoustic feedback to stretch each of the first and second spectrograms to a size representing a completed treatment, and applying the machine learning algorithm to the extrapolated first and second spectrograms to respectively obtain the total and local predicted BBB opening levels.

[0103] Clause 34: The method of clause 32, wherein the machine learning algorithm is trained using past treatment spectrograms as inputs and BBB opening levels respectively corresponding to the past treatment spectrograms as input labels.

[0104] Clause 35: The method of clause 34, wherein the BBB opening levels corresponding to the past treatment spectrograms are obtained from MRI analysis based on MR sequences from which T1 relaxation times were extracted from pre- and post-treatment images with and without Gadolinium (Gd) injection, 1 / T1=R1 values were calculated, and delta R1 volumes were calculated, from which Gd concentration changes in the treatment volumes were obtained.

[0105] Clause 36: The method of clause 34, wherein the BBB opening levels corresponding to the past treatment spectrograms are obtained from MRI analysis based on MR T1 qualitative images obtained pre- and post-treatment, with- and without-Gd injection, wherein the images without Gadolinium T1 underwent N4 bias analysis for Bl and receptive field correction, which was used to correct Bl and receptive fields of with-Gd injection volumes.

[0106] Clause 37: The method of any of clauses 27-36, wherein the sensor array includes one or more sensors selected from the group consisting of: MRI, CT, PET, EEG, and fMRI.

[0107] Clause 38: The method of any of clauses 27-37, wherein each of the first and second spectrograms includes at least one of: spectra of acoustic feedback captured by at least one receiver, a weighted sum of spectra of acoustic feedback captured by a plurality of receivers, spectra of acoustic feedback that are compensated by skull transmission of different acoustic frequencies, or spectra obtained by beamforming of a plurality of receiver signals in a time domain and application of a Fourier transform of the plurality of receiver signals after beamforming

[0108] Clause 39: The method of any of clauses 27-38, wherein each of the first and second spectrograms includes spectra of a plurality of frequency ranges used for safety, efficiency, and BBB opening deterioration.

[0109] Clause 40: A non-transitory computer-readable storage medium containing instructions for a computer system to perform the method of any of clauses 14-26.

[0110] Clause 41 : A non-transitory computer-readable storage medium containing instructions for a computer system to perform the method of any of clauses 27-39.

[0111] Reference have been made in detail to various implementations, examples of which are illustrated in the accompanying drawings. In the above detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention and the described implementations. However, the invention may be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the implementations.

[0112] It will be understood that, although the terms “first,” “second,” etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first device could be termed a second device, and, similarly, a second device could be termed a first device, without changing the meaning of the description, so long as all occurrences of the first device are renamed consistently and all occurrences of the second device are renamed consistently. The first device and the second device are both devices, but they are not the same device.

[0113] The terminology used herein is for the purpose of describing particular implementations only and is not intended to be limiting of the claims. As used in the description of the implementations and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. For example, “A, B, and / or C” means: A only; B only; C only; A and B; A and C; B and C; or A, B, and C. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0114] As used herein, the term “if’ may be construed to mean “when” or “upon” or “in response to determining” or “in accordance with a determination” or “in response to detecting,” that a stated condition precedent is true, depending on the context. Similarly, the phrase “if it is determined [that a stated condition precedent is true]” or “if [a stated condition precedent is true]” or “when [a stated condition precedent is true]” may be construed to mean “upon determining” or “in response to determining” or “in accordance with a determination”or “upon detecting” or “in response to detecting” that the stated condition precedent is true, depending on the context.

[0115] The foregoing description, for purpose of explanation, has been described with reference to specific implementations. However, the illustrative discussions above are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The implementations were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various implementations with various modifications as are suited to the particular use contemplated.

Claims

CLAIMSWhat is claimed is:

1. A system for controllably changing a tissue property in a target volume, the system comprising: an ultrasound transducer configured to transmit a sequence of acoustic pulses to the target volume to change the tissue property; a sensor array configured to capture acoustic feedback generated by the sequence of acoustic pulses; and a controller configured to:(i) generate a spectrogram using two or more bands of frequencies included in the acoustic feedback;(ii) computationally calculate a degree of change of the tissue property in the target volume based on the spectrogram, and(iii) control the ultrasound transducer based on the calculated degree of change of the tissue property to change the tissue property to a target degree set for a specific treatment.

2. The system of claim 1, wherein the spectrogram includes acoustic feedback captured from a beginning of treatment to a current point of treatment.

3. The system of claim 1, wherein the spectrogram includes a subset less than all of most recently captured acoustic feedback.

4. The system of any of claims 1-3, wherein the spectrogram includes extrapolated acoustic data extending the acoustic feedback from a current point of treatment to an end of treatment.

5. The system of any of claims 1-4, wherein the spectrogram includes at least one of: spectra of acoustic feedback captured by at least one receiver, a weighted sum of spectra of acoustic feedback captured by a plurality of receivers, spectra of acoustic feedback that are compensated by skull transmission of different acoustic frequencies, or spectra obtained by beamforming of a plurality of receiver signals in a time domain and application of a Fourier transform of the plurality of receiver signals after beamforming.

6. The system of any of claims 1-5, wherein the spectrogram includes spectra of a plurality of frequency ranges used for safety, efficiency, and deterioration of changes in tissue property.

7. The system of any of claims 1-6, wherein the controller is configured to calculate the degree of change based on outputs of a neural network trained on total accumulative spectra of defined sonication time.

8. The system of claim 7, wherein outputs of the neural network include degree of change of tissue property, treatment risk, or deterioration of the degree of change of tissue property.

9. The system of any of claims 1-8, wherein the controller is configured to control the ultrasound transducer by adjusting one or more treatment parameters in real time to maintain the target degree of change, wherein the one or more treatment parameters includes sonication power, sonication duration, sonication frequency, microbubble concentration, microbubble infusion rate, or therapeutic agent infusion rate.

10. The system of any of claims 1-9, wherein the change to the tissue property is: a disruption of a tissue barrier; a neuromodulation of tissue neurons; an activation of a sonodynamic therapy drug; an activation of contrast agent carriers for drug and / or gene delivery; a thrombolysis; an indication of tissue death and / or an induction of an ischemic effect.

11. The system of any of claims 1-10, wherein the change to the tissue property is a disruption of a tissue barrier to increase permeability of the barrier, wherein the tissue barrier is a blood-brain barrier, a blood-retina barrier, skin, a mucosal membrane, a cell membrane, or a nuclear membrane.

12. The system of claim 11, wherein the permeability is increased sufficiently to permit passage of a therapeutic agent therethrough, wherein the therapeutic agent is selected for treatment of a tumor, a neurogenerative disease, a neurologic disorder, a neuropsychiatric disorder, an enzymatic deficiency, or a CNS infection.

13. The system of claim 11, wherein the permeability is increased sufficiently to permit passage of a biomarker from within brain tissue to one or more blood vessels, enabling or enhancing detection through liquid biopsy and / or another diagnostic technique.

14. A method of controllably changing a tissue property in a target volume, the method comprising: applying a sequence of acoustic pulses to the target volume to change the tissue property; detecting acoustic feedback generated by the sequence of acoustic pulse; generating a spectrogram using two or more bands of frequencies included in the acoustic feedback; computationally calculating a degree of change of the tissue property in the target volume based on the spectrogram, and controlling application of the acoustic pulses based on the calculated degree of change of the tissue property to change the tissue property to a target degree set for a specific treatment.

15. A method for opening a blood-brain-barrier (BBB) during treatment, the method comprising: at a controller communicatively coupled to an ultrasound transducer configured to emit ultrasound waves and a sensor array configured to capture acoustic feedback:(a) obtaining acoustic feedback from the sensor array;(b) generating (i) a first spectrogram from acoustic feedback accumulated from a beginning of the treatment up to a current point of the treatment, and (ii) a second spectrogram from acoustic feedback selected from a subset the treatment leading up to the current point of the treatment;(c) determining a total predicted BBB opening level from the first spectrogram and a local predicted BBB opening level from the second spectrogram;(d) determining an adjusted parameter for opening the BBB based on (i) a difference between a target BBB opening level and the total predicted BBB opening level, and (ii) a difference between the target BBB opening level and the local predicted BBB opening level; and(e) causing the ultrasound transducer to emit ultrasound waves using the adjusted parameter to open the BBB.

Images