Methods, devices, and systems for building a composite biomarker atlas of the brain using focused ultrasound

Abstract

Embodiments are directed to methods, devices, and systems for building a composite biomarker atlas of a brain of subject using focused ultrasound. An example method comprises processing a response of a subject to focused ultrasound applied to a specific region within a brain of the subject, wherein the response includes or corresponds to a biomarker, associating information from the focused ultrasound to coordinates of the specific region within the brain, and based thereon, building a composite biomarker atlas of the brain of the subject.

Description

1690.105.111METHODS, DEVICES, AND SYSTEMS FOR BUILDING A COMPOSITE BIOMARKER ATLAS OF THE BRAIN USING FOCUSED ULTRASOUNDCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This PCT Application claims benefit to U.S. Provisional Application No. 63 / 776,103, filed March 22, 2025, the entirety of which is incorporated herein by reference.BACKGROUND

[0002] Understanding the biological, chemical, electrical, mechanical, and other aspects of organs is key to the development of treatments for diseases and conditions in humans. Of the various organs in a human, the brain is one of the hardest to diagnose and treat partly due to its location within the skull, which makes it hard to access. The brain is also a complex organ with dense interconnections. It is relatively easy to obtain an image of the brain through techniques that involve magnetic resonance or computed tomography (using X-rays); however, there is much that still needs to be discovered at a more microscopic, cellular, and molecular level. In addition, the mechanical, electrical, and chemical composition of the properties of the brain and the changes these compositions and properties undergo as a function of the condition of the brain are not well understood. Collectively, the measurable mechanical, electrical, and chemical compositions and properties may be referred to as “biomarkers.” Thus, there is a need to understand the biomarker status of the brain as a function of the condition of the brain.BRIEF DESCRIPTION OF THE DRAWINGS

[0003] Fig. 1 A illustrates example biomarker maps and a composite biomarker atlas.

[0004] Fig. IB illustrates example biomarker maps and a composite biomarker atlas.

[0005] Fig. 1C illustrates an example device for building a composite biomarker atlas.

[0006] Fig. 2A illustrates an example system that includes an ultrasound device configured to apply sonication to a specific location within the brain of a subject.

[0007] Fig. 2B illustrates another example system that includes an ultrasound device configured to apply sonication to a specific location within the brain of a subject.

[0008] Fig. 3 illustrates at least portions of an example atlas builder system.1690.105.111

[0009] Fig. 4A illustrates an example transformation process between coordinates in an input image of a brain of a subject and coordinates of a corresponding location in a standardized image of a brain.

[0010] Fig. 4B illustrates an example transformation process between coordinates in a standardized image of a brain and coordinates of a corresponding location within an input image of a brain of a subject.

[0011] Fig. 5A is a flow chart illustrating an example method for building a composite chemical biomarker atlas.

[0012] Fig. 5B is a flow chart illustrating an example method for building a composite electrical biomarker atlas.

[0013] Fig. 6A illustrates an example flow chart for building or using the composite chemical biomarker atlas, including preplanning and sonication steps and use of the composite chemical biomarker atlas.

[0014] Fig. 6B illustrates an example flow chart for building or using the composite electrical biomarker atlas, including preplanning and sonication steps and use of the composite electrical biomarker atlas.

[0015] Fig. 7 illustrates an example use of a composite biomarker atlas.

[0016] Fig. 8 illustrates example blocks of an atlas builder system.DETAILED DESCRIPTION

[0017] This disclosure describes inventive concepts with reference to specific examples. However, the intent is to cover all modifications, equivalents, and alternatives of the inventive concepts that are consistent with this disclosure. It will be apparent, however, to one of ordinary skill in the art that the present approach can be practiced without these specific details. Thus, the specific details set forth are merely exemplary, and is not intended to limit what is precisely disclosed. The features implemented in one example may be implemented in another example where logically possible. The specific details can be varied from and still be contemplated to be within he spirit and scope of what is being disclosed.

[0018] The condition of the brain can be characterized in numerous ways, including health conditions (which include diseases and healthy conditions), age, and gender, among others, and which can be used for a variety of purposes. An example of how such understanding may be helpful is if the protein composition of the brain for a specific disease is known and if tools exist to obtain the protein composition, then by comparing the current protein composition to the known protein composition, along with other relevant medical1690.105.111information, the existence of a particular disease may be confirmed. For such a purpose, a location-specific composite chemical biomarker atlas of the brain in accordance with the present disclosure can be useful. The location in the phrase “location-specific composite chemical biomarker” refers to the location within the brain from where the biomarker originated. Embodiments of the present disclosure are directed to a composite biomarker atlas that may contain chemical biomarker (e.g., makeup and concentration) information for the (whole) brain; it may also contain information about when the brain is healthy or diseased. Further, the chemical biomarker information may be different as a function of disease, disease stage, age, sex, etc. Several types of chemical biomarkers can be detected and analyzed, including proteins, nucleic acids, and extracellular vesicles. For various purposes and / or in addition, other biomarkers may be useful, including mechanical biomarkers and / or electrical biomarkers, and which can be used to build a location-specific composite mechanical biomarker atlas, a location-specific composite electrical biomarker atlas, and / or a location-specific composite multi-modal biomarker atlas, as further described herein.

[0019] Embodiments of with the present disclosure are directed to methods, devices, and systems for building a composite biomarker atlas. Such embodiments may be associated with and / or include processing a response of a subject to focused ultrasound applied to a specific region within a brain of the subject, wherein the response includes or corresponds to a biomarker, associating information from the focused ultrasound to coordinates of the specific region within the brain of the subject, and based thereon, building the composite biomarker atlas of the brain of the subject. In some embodiments, the information includes parameters of sonication and / or the response of the subject. In some examples, embodiments include processing and analyzing the response to the focused ultrasound and, in response, associating the information including results of the analysis to the coordinates of the specific region within the brain. In some embodiments, associating the information includes first associating the information to coordinates of the specific region within the brain of the subject (e.g., patient specific) and then associating the coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain, such as by mapping the patient specific coordinates to coordinates in the standardized image. The composite biomarker atlas can be built for the brain over multiple subjects, and / or over a single sonication session or multiple sonication sessions for each of the multiple subjects.

[0020] As further described herein, the biomarker can be a chemical biomarker, an electrical biomarker, and / or a mechanical biomarker. In some embodiments, the response is multimodal and includes at least two biomarkers.1690.105.111

[0021] As used herein, the term “biomarkers” refer to and / or include any biomarkers from the brain, including mechanical, electrical, and / or chemical compositions or properties.Collectively, the measurable mechanical, electrical, and chemical compositions and properties are referred to as “biomarkers”. For purposes of clarity, the term “biomarkers” is used to refer to all types (e.g., mechanical, electrical, and chemical) of biomarkers; however, those that are associated with molecules are referred to as “chemical biomarkers", those associated with electrical activity are referred to as “electrical biomarkers”, and those associated with mechanical activity are referred to as “mechanical biomarkers”. In some embodiments, multiple composite biomarker atlases may be built, each with one type of biomarker, or alternatively, one composite biomarker atlas may contain information about multiple biomarkers and, therefore, may be multi-modal in nature.

[0022] Although the descriptions below are provided in relation to a human patient, the same concepts may be applied to animals as well. Thus, the term “subject” may be used instead of and / or interchangeably with “patient”.

[0023] Turning now to the figures, Fig. 1 A illustrates example biomarker maps and a composite biomarker atlas. The biomarkers are associated with a response to focused ultrasound at a location in the brain 112 of a subject 110, where optionally the blood-brain barrier (BBB) is opened with ultrasound sonication. As such, the response to the focused ultrasound can include or correspond to the biomarker. The biomarkers in the example Fig.1 A are chemical biomarkers. In particular, Fig. 1 A illustrates how information about the chemical biomarkers may be organized in a composite biomarker atlas 104.

[0024] As noted above, chemical biomarkers refer to and / or include compositions or properties from the brain which are associated with molecules. Example chemical biomarkers include a brain biomarker and a tumor biomarker, among other molecules, such as circulating tumor DNA (ctDNA), cell-free DNA (cfDNA), among others. A brain biomarker refers to and / or includes a brain tissue biomarker that may be associated with the opening of the BBB, and which is not associated with a disease state, such as a tumor, and may be referred to as a “non-disease brain biomarker”. Example brain biomarkers include glial fibrillary acidic protein (GFAP) and myelin basic protein (MPB), among others. A brain biomarker refers to and / or includes a brain tissue biomarker that may be associated with the opening of the BBB and is associated with a disease state, e.g., a tumor.

[0025] A subject 110 is shown in Fig. 1A. Although a human head and brain 112 is indicated, embodiments can be extended to other organs and non-human subjects as well. Two locations representing regions in the brain 112 are indicted as Pi and P2. Fig. 1A shows example1690.105.111biomarker maps 102-1, 102-2 that are associated with each location Pi and P2. A biomarker map 102-1, 102-2 may refer to and / or include a multi-dimensional matrix, where each cell in the matrix contains specific information about a response of the brain region it is associated with. Multi-dimensional includes a case where the matrix is a one-dimensional matrix. Each dimension of the multi-dimensional matrix may be specified by an axis. In some embodiments, multiple biomarker maps 102-1, 102-2 may be associated with a single region. A collection of biomarker maps is referred to herein as a composite biomarker atlas 104. In this example, two example biomarker maps 102-1 , 102-2 (map 1 and map 2) are constructed in a similar manner (e.g., the axes, parameters, and size of the matrix are all identical, while the values in each cell may be different), but in other examples, biomarker maps 102-1, 102-2 may be different from each other. As an example, the biomarker maps 102-1, 102-2 may have different numbers of axes with respect to one another. Referring to map 1 102-1 as an example, in column 1, the chemical biomarkers present at Pi are specified.

[0026] Columns 1 through m in the biomarker maps 102-1, 102-2 of Fig. 1 A specify the concentration of a specific chemical biomarker 108, for example, Bl at time points ti through tm. An example of a chemical biomarker may be a cell death chemical biomarker, caspase-3. An increase in caspase-3 may occur, for example, with the administration of chemotherapy to location Pi. Thus, in response to chemotherapy, the concentration of caspase-3 may increase, indicating that the therapy may be having an intended effect. Another example of a chemical biomarker is the isocitrate dehydrogenase (IDH) gene. Mutations in this gene can be associated with cancers, such as glioblastoma. Thus, in Fig. 1 A, a chemical biomarker 106 in column A can be the IDH gene, and the associated value in column 1 (or other columns depending on when the measurement is taken) can be the structure of the mutated version of the gene.

[0027] Embodiments are not limited to chemical biomarkers, and may include mechanical biomarkers and / or electrical biomarkers. As noted above, electrical biomarkers refer to and / or include compositions or properties from the brain which are associated with electrical activity. Example electrical biomarkers include brain wave frequencies (e.g., electroencephalograms (EEG)), electrical events and / or patterns (e.g., spikes in electrical pulses, event-related potentials, epileptic spikes), and / or magnetic fields produced by neuronal currents (e.g., magnetoencephalogram (MEG)). As noted above, mechanical biomarkers refer to compositions or properties from the brain which are associated with mechanical activity. Example mechanical biomarkers includes pulsations, stiffness, and structural changes which may be measured using imaging techniques. For example, the1690.105.111imaging techniques can include Magnetic Resonance Elastography (MRE) to measure stiffness and functional imaging (e.g., functional MR imaging, Event-Related Optical Signal (EROS), ultrasonic imaging, brain pulse analysis (MR imaging and / or Doppler ultrasound), among others) to measure tissue swelling, blood flow changes, and other information.

[0028] One of the dimensions in such biomarker maps 102-1, 102-2 may be a time axis. Fig.1A illustrates an example first map, map 1 102-1. Map 1 102-1 is a two-dimensional map where one of the axes includes the chemical biomarkers 106, as shown in column A. The second dimension is a temporal axis containing time instants ti through tm- Each cell in this example contains a value about the concentration of a specific chemical biomarker at a specific time instant.

[0029] An example aspect also shown in Fig. 1 A is the association of the chemical biomarkers to the location of where the chemical biomarkers originated from within the brain 112. In some embodiments, a coordinate system 114 can be established. The coordinates of the locations Pi and P2 can be specified in reference to the established coordinate system 114.

[0030] In some embodiments, the coordinate system 114 may be subject specific, e.g., are coordinates specific to the region of the brain of the subject 110. In some embodiments, the coordinate system 114 may be based on a standardized image of the brain, as further described herein. For example, the subject specific coordinates of the specific region of the brain 112 may be mapped to coordinates of a specific region of the brain in the standardized image of the brain. The specific regions of the brain or the same regions (or locations) of the brain (e.g., specific to the subject 110 and the standardized image and / or to another subject, as further described below) can include the same or substantially the same brain region with respect to one another. For example, the specific regions may at least partially overlap, but may not be identical in various embodiments. As further described below, the specific regions may overlap such that the centroids of the regions are displaced by distances measured in millimeters such as 1 millimeter (mm), 1.5 mm, 2 mm, 3 mm, or under 10 mm. The overlap may also be measured in fractions or multiple of wavelengths of the transmit frequency. At 250 kilohertz (kHz), the wavelength is about 6 mm in the brain.

[0031] As further described below, Fig. IB illustrates another example biomarker map and composite biomarker atlas. In Fig. IB, the biomarkers are associated with a location in the brain of a subject 110 where the brain is subjected to ultrasound neuromodulation. The biomarkers in the example of Fig. IB are electrical biomarkers.

[0032] Fig. 1C illustrates an example device for building a composite biomarker atlas. The device 103 may be used to form any of the biomarker maps and composite biomarker atlas1690.105.111described herein, such as those illustrated by Figs. 1A-1B, can form part of any system, such as those illustrated by Figs. 2A-2B, can be an implementation of or include at least part of the atlas builder system, such as those illustrated by any of Figs. 2A-3, and / or used to implement the example methods, such as those illustrated by Figs. 4A-6B. In some embodiments, the device 103 be separate from the ultrasound device and / or may be distributed. For example, the processor 105 and memory 107, as further described below, may include processing and memory resources which are distributed, such as via a Cloud-computing system. In some embodiments, the processor 105 and memory 107 may form part of computing device(s) which are local to or remotely located to the subject.

[0033] As shown by Fig. 1C, the device 103 comprises a processor 105 and memory 107. The memory 107 may include a computer-readable storage medium storing a set of instructions 109, 111, 113. The memory 107 may include Read-Only Memory (ROM), Random-Access Memory (RAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, a solid state drive, Electrically Programmable Read Only Memory aka write once memory (EPROM), physical fuses and e-fuses, and / or discrete data register sets. In some embodiments, memory 107 may be a non-transitory storage medium, where the term “non-transitory” does not encompass transitory propagating signals. The processor 105 may be coupled to the memory 107 to execute the instructions 109, 111, 113 to perform the actions, as further described below.

[0034] At 109, the processor 105 may obtain and process a response of a subject to focused ultrasound applied to a specific region within a brain of the subject. The response may include or correspond to a biomarker (e.g., a measurable mechanical, electrical, and chemical composition and / or property). For example, the response of the subject may include a chemical response and / or a physiological response. In some embodiments, the response of the subject may include or correspond to a biomarker, such as a chemical biomarker, a mechanical biomarker, and / or electrical biomarker. In some embodiments, the response is multi-modal, such as being associated with at least two biomarkers and / or multiple types of biomarkers (e.g., different combinations of chemical, mechanical, and / or electrical biomarkers), as further described herein.

[0035] In some embodiments, the processor 105 may be configured to execute the instructions 109 to obtain the response of the subject to the focused ultrasound via communication with an ultrasound device and at least one of: a chemical biomarker analysis system configured to capture a chemical response, a physiological waveform processing system configured to capture a physiologic response, an imaging device configured to capture1690.105.111an image of the brain of the subject, and a computing device in communication with the imaging device and / or ultrasound device. In some embodiments, the device 103 is in communication with at least three or all of the listed devices and systems.

[0036] At 111, the processor 105 may associate information from the focused ultrasound to coordinates of the specific region within the brain of the subject. In some embodiments, the information includes parameters of sonication and / or the response of the subject to the focused ultrasound. As used herein, parameters of sonication refer to and / or include parameters for or associated with applying the focused ultrasound. Example parameters of sonication include identification of transducers which are excited and defined transmit parameters, including amplitude, phase, timing, and sequence. Other example parameters of sonication include where the focused ultrasound is applied (e.g., brain location), pressure level, frequency of sonication, sonication duration, pulse length, and pulse repetition frequency, among others. In some embodiments, the processor 105 may be configured to execute the instructions 111 to process and analyze the response to the focused ultrasound and, in response, associate the information including results of the analysis to the coordinates of the specific region of the brain.

[0037] In some embodiments, associating the information can be subject specific. For example, the information can be associated to coordinates within the specific region of the brain of the subject. In some embodiments, associating the information includes associating the information from coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain. For example, the processor 105 can first associate the information to coordinates of the specific region within the brain of the subject (e.g., patient specific) and then associate the coordinates of the specific region within the brain of the subject to coordinates of the specific region in the standardized image of the brain, such as by mapping the subject specific coordinates to coordinates in the standardized image.

[0038] As noted above, the processor 105 may be in communication with various devices and / or systems. In some embodiments, the processor 105 may obtain the parameters of sonication from the ultrasound device and / or the response of the subject to the focused ultrasound from a chemical biomarker analysis system, a physiological waveform processing system, an imaging device, and / or a computing device in communication with the imaging device and / or ultrasound device.1690.105.111

[0039] In some embodiments, the processor 105 may execute the instructions to process and analyze the response by comparing the response to sonication applied to the specific region of the brain of the subject and a response to sonication applied to a specific region of a brain of another subject. The specific regions of the brains of the subject and the other subject can be the same region, which as described above, can be at least overlapping. For example and in some such embodiments, the processor 105 can build the composite biomarker atlas for the brain over multiple subjects, including the subject, and / or over a single sonication session or multiple sonication sessions for each of the multiple subjects. In some embodiments, the composite biomarker atlas is built for the whole brain, a portion of the brain, or portions of the brain over the multiple subjects. For example, in the embodiments where the parameters are optionally modified on a subject-to-subject basis, the calculations may consider the parameters stored in the composite biomarker atlas for one or all of the subjects whose data contributed to the information at the location in question. In addition, the calculation may also consider if the previous sonication(s) resulted in providing acceptable biomarker data. As an example, the calculations may ignore sonication(s) that did not result in successful detection of biomarkers or successful level of biomarkers. Thus, the parameters for sonication for a new subject may be based on the prior sonication parameters used on prior subjects or the results of the biomarker analysis on prior subjects or both.

[0040] Accordingly, and in some embodiments, the processor 105 may be configured to use the composite biomarker atlas and / or information from the ultrasound device to provide additional information. For example, the processor 105 can calculate a location of the sonication in the subject based on a corresponding location in a standardized image of a brain. As another example, the processor 105 can obtain parameters of sonication for a particular location of the specific region in the brain of the subject from the information associated with the location in the composite biomarker atlas. Further, in some embodiments, the processor 105 can obtain the parameters of sonication from the composite biomarker atlas and which optionally are modified on a subject-to-subject basis, such as described above.

[0041] Fig. 2A illustrates an example system that includes an ultrasound device configured to apply sonication to a specific location within the brain of a subject. Fig. 2A additionally shows the data and information flow between the ultrasound device 246 and the atlas builder system 250. The ultrasound device 246 may be used to open the BBB of a subject 210. The system 230 can be used to create a composite biomarker atlas which includes the location of where the chemical biomarker originated from and is associated with the chemical biomarker itself, as further described herein.1690.105.111

[0042] The ultrasound device 246 can include a processor and memory (such as described above) which is configured to control delivery of focused ultrasound to the subject 210. The ultrasound device 246 can form part of an ultrasound system which includes the ultrasound device 246, an ultrasound cap 232 or other ultrasound transducer, and a registration device 248. The ultrasound system may include at least some of substantially the same features and attributes and / or otherwise be implemented as described in US Patent No. 11,534,630, issued on December 27, 2022, and entitled “Ultrasound Guided Opening of Blood-Brain Barrier”, which is incorporated herein by reference in its entirety for its teaching. For example, the ultrasound system can include the ultrasound cap 232 for placement on the head of the subject 210, optionally, components for providing a contrast agent to the subject 210 (e.g., stand, bag containing the contrast agent, electronically controlled valve, a cable coupling the valve to the ultrasound device 246), the ultrasound device 246 having the processor for controlling the focused ultrasound, and components for connecting the ultrasound cap 232 to the ultrasound device 246 (e.g., cables, cable interface).

[0043] The ultrasound cap 232 may include at some of substantially the same features and attributes and / or otherwise be implemented as described in US Publication No.2023 / 0082109, published on March 16, 2023, and entitled “Ultrasound Transducer Assembly”, which is incorporated herein by reference in its entirety for its teaching. In brief, the ultrasound cap 232 may include several types of transducer components: (1) a low-frequency transducer configured to provide the energy to open the blood-brain barrier, (2) a high-frequency transducer set or imaging array configured to image through the thin areas of the skull to image structures of the brain; the ultrasound images obtained from the high-frequency set may be aligned with the images obtained from another modality such as magnetic resonance (MR) where the energy for imaging can penetrate the skull, and / or (3) monitoring transducers which may be used in a feedback loop to open the BBB. In various embodiments, the monitoring transducers may be used to provide safe opening of the BBB, by capturing or receiving signals coming from structures within the brain and presenting these signals for further analysis by other components of the ultrasound system.

[0044] In some embodiments, the low-frequency range may be between 0.200 megahertz (MHz) to 10 MHz and, in some embodiments, may be between 0.25 MHz to 5 MHz. The high-frequency range may be between 2 MHz to 5 MHz, for example. The monitoring transducers may operate with a bandwidth 100 kHz to 10 MHz, for example.

[0045] The registration device 248, in some embodiments, may form part of the ultrasound device 246 or is separate therefrom. The registration device 248 may include a processor and1690.105.111memory (such as previously described) which is configured to perform registration and adjustments calculations. The registration device 248 may include at least some of substantially the same features and attributes or may be used to perform at least some of substantially the same registration and adjustment calculations as described in: US Patent No.11,534,630; US Publication No. 2024 / 0042242, published on February 8, 2024, and entitled “Rapid Calculation of Parameters for Delivering Ultrasound Energy to Selected Locations in the Brain”; and / or PCT Publication No. WO2024 / 238674, published on May 15, 2024. entitled “Methods and Systems for Optimization of Ultrasound-Facilitated Liquid Biopsy”, which are each incorporated herein by reference in their entirety for their teaching.

[0046] Fig. 2A shows a subject 210 with the ultrasound cap 232 placed on the head of the subject 210. As described above, the ultrasound cap 232 comprises single or multiple ultrasound transmitters. In addition, the ultrasound cap 232 may also have one or multiple ultrasound receivers. The ultrasound transmitters in the ultrasound cap 232 can be electrically connected to the ultrasound device 246. Each transmitter can be individually excited with a precalculated phase and amplitude. Due to the phase and amplitude control, the ultrasound beam from the ultrasound device 245 can be focused inside the brain through the skull at a desired location, which is programmable.

[0047] A registration process can be used to ensure that the focused ultrasound is accurately guided to the desired location. The registration process can involve mapping or aligning the images of the head of the subject 210 from magnetic resonance (MR) or other imaging device 242 to the actual head of the subject 210, such as MR or computed tomography (CT) images. An imaging device refers to and / or includes a processor and other components configured to capture an image of the brain of the subject. Example imaging devices can include an MR imaging device, a CT imaging device, and a fused MR and CT imaging device, among other known imaging devices. Example aspects and features of a non-invasive registration process are described in US Patent No. 11,534,630. For registration of the images to the physical face, a technique can be used to associate the physical world with the computer world where the images are contained. There are several techniques to accomplish this, but one example technique is shown in Fig. 2A. Here, a registration device 248 is shown electrically connected to three sensors (a cap sensor 236, a subject sensor 234, and a pointer sensor 238). The registration device 248 may work using several techniques, such as optical techniques or electromagnetic techniques. In each example implementation, the registration device 248 can define a home base that can establish a coordinate frame of reference in the physical world and transmit optical or electromagnetic energy from the home base. Distance, position, and1690.105.111motion of the sensors 234, 236, 238 can be tracked in real-time as the sensors 234, 236, 238 receive the energy and send signals back to the home base. In the case of optical sensing, the sensors 234, 236, 238 may not be connected electrically to the registration device 248. In Fig.2A, an electromagnetic (EM)-based registration device is shown. The cap sensor 236, the subject sensor 234, and the pointer sensor 238 all have small (around 1-2 mm) EM sensors that receive the EM energy from the home base of the registration device 248 and send the received signals back to the registration device 248 from which the position, distance, and motion can be calculated. In some embodiments, the cap sensor 236 can be coupled to the ultrasound cap 232 and the subject sensor 234 can be coupled to the subject 210 with a removable adhesive. Multiple anatomical landmarks, such as the corner of the eyes (canthi), the region between the eyes (nasion), and the pointed eminence of the ear (tragus), are specified in the CT image of the subject 210. The same landmarks are be selected or pointed to by the subject sensor 234. In some embodiments, a user (e.g., medical professional) can hold the subject sensor 235 and point to these locations on the face of the subject 210.

[0048] Once the locations of the physical landmarks are entered into the registration device 248, a transformation matrix between the physical world and the computer world can be computed by computer-executable instructions (e.g., software) residing in memory of the registration device 248. The transformation matrix can involve functions like scaling, translation, and rotation. After the transfomiation matrix is calculated, the coordinates of any point in the computer world (for example, in the MR image) can be transformed into the coordinates in the physical world (within the brain of the subject 210) and vice-versa. Thus, the CT image can be first aligned to the physical head of the subject 210. The MR and CT images are then aligned with one of several computer-executable instruction packages, such as Brainlab Elements, TeraRecon Intuition etc. With these steps, the MR and CT images are aligned to the head of the subject 210.

[0049] In some embodiments, the ultrasound device 246 can receive MR and CT images of a subject 210 from the hospital network 247. An user, such as a physician, can open the files of the subject and examine the MR and CT images using a display screen that may be coupled to the ultrasound device 246. The user can specify a region of interest where the BBB is to be opened, particularly if the subject 210 is being encountered for the first time with a particular disease condition. For the subsequent encounters with the subject 210, an alternate configuration may be used, as is further described below. In some embodiments, the ultrasound device 246 computes a physical and acoustic model of the subject 210 based on the MR and the CT images of that subject 210 and as obtained using the MR imaging or other1690.105.111imaging devices 242. The ultrasound device 246 can further compute which transducers within the ultrasound cap 232 are to be excited with particular transmit parameters, such as amplitude, phase, timing, and sequence, so that an ultrasound focus is formed at and near the vicinity of the specified region of interest of brain of the subject 210.

[0050] In some embodiments, the ultrasound cap 232 may include receivers. These receivers can receive the ultrasound energy reflected back from the tissue and, in various embodiments, from microbubbles 241 that are injected into the subject 210 for the purposes of the BBB opening, such as via an intravenous catheter 240. The signals from the microbubbles 241 can be strong, such as in the frequency band outside the fundamental frequency (fo ) or the frequency of the transmitted energy. The frequency content of the received signals provides information about how the microbubbles 241 are vibrating inside the body as a result of the magnitude of ultrasound energy. If the microbubbles 241 are vibrating in a stable cavitation mode or in a state that results in the safe opening of the BBB, the signals at the harmonic and sub (0.5 / o ) and ultra harmonic ((2n+l) / 2 fo , n=l,2..) and harmonic (2n, n=l,2,..) are strong. If the microbubbles 241 are vibrating violently as a result of high transmitted pressure (for example, higher than about 0.5 MPa at 250 kHz), which results in the unsafe opening of the BBB, the signals can have a high energy content across a broad range of frequency bands, for example, at a non-rational multiple of fo (II / o). Thus, the electronics (e.g., processor, memory, and other components) associated with the ultrasound device 246 and coupled to the receivers in the ultrasound cap 232 and the signal processing capabilities in the ultrasound device 246 can monitor the signals coming back from the tissue continuously while the ultrasound device 246 causes the transmission of ultrasound energy into the tissue. The ultrasound device 246 can determine if the BBB has been opened safely using this method.

[0051] As shown in Fig. 2A, the ultrasound device 246 can receive and send messages to other devices and / or systems. For example, the ultrasound device 246 can output messages, including data related to the location of BBB opening (or coordinate(s) of BBB opening), how the BBB was opened (e.g., safe or unsafe), the frequency content of the signals received from the receivers, and / or other data. Some or all of the messages or data may be used in subsequent steps, as further described below. In relation to the BBB opening location, an example method to confirm the location of the BBB opening is described in US Patent No.12,017,093, issued on June 25, 2024, and entitled “Methods and systems for confirming focus of ultrasound beams’’, which is incorporated herein by reference in its entirety for its teachings. Thus, in some embodiments, the BBB opening location message provided by the1690.105.111ultrasound device 246 can be described as the “confirmed BBB opening location’’ in contrast to the “planned opening location’’, which is can be specified by a user in an MR image of the brain of the subject 210.

[0052] While Fig. 2A shows an ultrasound device 246 and a separate MR imaging or other imaging device 242, in various embodiments, the ultrasound device 246 can also include an MR imaging or other imaging device 242. Thus, in various embodiments, the MR or other imaging device 242 and ultrasound device 246 can be a combined device. In some embodiments, the subject 210 can be transferred to a standalone MR imaging or other imaging device 242 after the ultrasound sonication procedure with the ultrasound device 246 is complete. The MR imaging or other imaging device 242 can generate and send messages that can be used in subsequent steps and that can include at least some of the same data as described above, such as the location of the BBB opening and whether the BBB opening is safe or unsafe.

[0053] Fig. 2A illustrates a blood draw device 259. In some embodiments, the blood draw device 259 may be a catheter inserted into the subject 210 to collect blood. Blood samples can be collected prior to the ultrasound sonication procedure for opening the BBB and after opening the BBB. After collection, the blood can be sent for analysis to a chemical biomarker analysis system 258, as further described below. In some embodiments, blood can be drawn at specific times after the BBB opening. Embodiments are not so limited, and other types of fluids may be drawn. For example, a fluid draw device can be configured to collect the fluid, and may include a catheter, a needle, collection tube or syringe, among other types of devices. Non-limiting examples of fluids include blood, urine, and / or cerebrospinal fluid.

[0054] Fig. 2A further illustrates an atlas builder system 250. Data, messages, and / or information from the ultrasound device 246, from the chemical biomarker analysis system 258, and from other systems, such as for obtaining MR and / or CT images via the hospital network 247 (from the MR imaging or other image device 242 and / or via a computing device 244), can be input to the atlas builder system 250. Information related to the BBB opening procedure may also be entered by a user via computing device 244 and / or display / input-output 252. This information may be used to add tags or labels to the information sent by the ultrasound device 246. Tagging or labeling can be useful when the associated information is used with a machine learning and / or other artificial intelligence models.

[0055] A chemical biomarker analysis system 258 can be configured to capture a chemical response. For example, the chemical biomarker analysis system 258 can include a processor and memory (and other components) configured to detect and analyze chemical biomarkers1690.105.111in body fluids, such as in blood, urine, and / or cerebrospinal fluid. The chemical biomarker analysis system 258 can include a fluid draw device configured to collect the fluid and provide the fluid to an analysis device including the processor and memory. The analysis device of chemical biomarker analysis system 258 may analyze the fluid specimen, such as by using a liquid biopsy assay(s) or other tests to detect the chemical biomarker. Various techniques can be used, including enzyme-linked immunosorbent assay (ELISA), polymerase chain reaction (PCR), next-generation sequencing (NGS), digital droplet PCR (ddPCR), and / or mass spectrometry (MS).

[0056] Example systems of the present disclosure are not limited to systems including all components as illustrated by Figs. 2A-2B. An example system, in accordance with various embodiments, can include the ultrasound device 246 and the processor which may form part of the atlas builder system 250. The ultrasound device 246 can be configured to apply focused ultrasound to a specific region within a brain of a subject 210, and optionally in response, open a BBB of the subject 210. In some embodiments, the ultrasound device 246 is configured to apply the focused ultrasound non-invasively. In some embodiments, the ultrasound device 246 is configured to apply the focused ultrasound to the specific region within the brain via a sonication procedure which is uninterrupted in response to subject motion such that the specific region of the brain is sonicated independent of the subject motion. In some embodiments, and as a practical matter, different regions may be sonicated due to subject motion may overlap such that the centroids of the regions are displaced by distances measured in millimeters such as 1 mm, 1.5 mm, 2 mm, 3 mm, or under 10 mm. The overlap may also be measured in fractions or multiple of wavelengths of the transmit frequency. At 250 kHz, the wavelength is about 6 mm in the brain.

[0057] The processor can: (i) obtain and process a response of the subject 210 to the focused ultrasound applied to the specific region within the brain of the subject 210, (ii) associate information from the focused ultrasound to coordinates of the specific region within the brain of the subject 210, and (iii) based thereon, build a composite biomarker atlas of the brain of the subject 210. As previously described, in some embodiments, the processor is configured to associate the information from the focused ultrasound to coordinates of the specific region within the brain by associating parameters of sonication and / or the response of the subject 210 to the coordinates. In some embodiments, the processor is configured to associate the information from the focused ultrasound to coordinates of the specific region within the brain by associating the information from coordinates of the specific region within the brain of the subject 210 to coordinates of a specific region in a standardized image of the brain.1690.105.111

[0058] In some embodiments, the processor can process and analyze the response to the focused ultrasound and, in response, associate the information including results of the analysis to the coordinates of the specific region in the standardized image. For example, the processor can calculate a location of the sonication in the subject 210 based on a corresponding location in a standardized image of a brain.

[0059] As previously described, the response can include or correspond to a biomarker, such as a chemical biomarker, a mechanical, and / or an electrical biomarker. In some embodiments, the response is multi-modal, such as corresponding to at least two of a chemical biomarker, a mechanical, and an electrical biomarker.

[0060] In some embodiments, the processor is configured to obtain the response of the subject 210 to the focused ultrasound via communication with the ultrasound device 246 and at least one of a chemical biomarker analysis system 258 configured to capture a chemical response, a physiological waveform processing system (262 of Fig. 2B) configured to capture a physiologic response, an imaging device 242 configured to capture an image of the brain of the subject 210, and a computing device 244 in communication with the imaging device 242. For example, the processor can be in communication with at least two of the ultrasound device 246, a biomarker analysis system, the imaging device 242, and the computing device 244. A biomarker analysis system includes a diagnostic tool which is configured to detect and analyze biomarkers, including but not limited to chemical, electrical, and mechanical biomarkers. The biomarker analysis system can include a processor and memory configured to detect and / or analyze the biomarker. Example biomarker analysis systems include the chemical biomarker analysis system 258 and a physiological waveform processing system, as further described below.

[0061] In some embodiments, the processor of the system is configured to process and analyze the response by comparing the response to sonication applied to the specific region of the brain of the subject 210 and a response to sonication applied to a specific region of the brain of another subject. For example, the processor can build the composite biomarker atlas for the brain over multiple subjects, including the subject 210, and over a single sonication session or multiple sonication sessions for each of the multiple subjects.

[0062] In various embodiments, the processor can process additional information from the composite biomarker atlas. For example, the processor can obtain parameters of sonication for a particular location of the specific region in the brain of the subject 210 from the information associated with the location in the composite biomarker atlas. In some1690.105.111embodiments, the processor can obtain the parameters of sonication from the composite biomarker atlas and which optionally are modified on a subject-to-subject basis.

[0063] As further described below, Fig. 2B illustrates another example system that includes an ultrasound device configured to apply sonication to a specific location within the brain of a subject. Fig. 2B shows the data and information flow between the ultrasound device 246 and the atlas builder system 250. As previously described, the ultrasound device 246 in may be used to apply ultrasound neuromodulation to a subject 210.Process steps within atlas builder

[0064] Fig. 3 illustrates at least portions of an example atlas builder system. The atlas builder system 370 can include a device comprising a processor and memory, such as previously described. In some embodiments, the atlas builder system 370 of Fig. 3 can include an implementation of the device 103 of Fig. 1C.

[0065] Fig. 3 shows example processing steps 372 within the atlas builder system 370. The atlas builder system 370 may accept data input 371 from several sources, such as the ultrasound device, the biomarker analysis system, and other devices and systems as illustrated by Figs. 2A-2B. A data check can be initially performed, as shown at 373. The data check can assess and, optionally, ensure that the data being input at 371 has the necessary information. As an example, if the data input is the result of the chemical biomarker analysis, the data check step can assess and optionally ensure that the result is associated with the correct location where BBB was opened. As another example, if the data input is a message from an ultrasound device, the data check step can assess if the message includes if the BBB was opened safely or not.

[0066] At 374, an MR image of the subject (e.g., an input MR image) can be mapped to a standardized MR image (e.g., a target image). A standardized image, as used herein, refers to and / or includes a template image of a brain which includes common structures of the brain. Typically, such standardized images are three-dimensional (3D) templates of the brain, including the common structures, and which is shaped and sized to normalize variances between subjects. Example standardized images include standardized MR images and / or CT images. There are several standardized MR images that are commercially available. One example is the ICBM 452 T1 Atlas. As may be appreciated, the description of the ICBM 452 T1 Atlas states that the ICBM 452 is an average MRI of normal young adult brains. It is an average of intensities and spatial positioning based on multiple human subjects.1690.105.111

[0067] The mapping from the MR image of the brain of the subject to a standardized MR image is further shown by Figs. 4A-4B. Fig. 4A illustrates an example transformation process between coordinates in an input image of a brain of a subject and coordinates of a corresponding location in a standardized image of a brain (which can be based on multiple subjects). Fig. 4B illustrates an example transformation process between coordinates in a standardized image and coordinates of a corresponding location within an input image of a brain of a subject.

[0068] There are multiple techniques to map an input image of a subject to a standardized image. One example method may use affine transformations, which is a linear geometric transformation that involves translation, rotation, scaling, and shearing. The transformation equation may be described as:~x' rmum12m13mu-iy' m21m22m23m24m3.... Eqn. 1z'1m32m33m34z.1 0 0 0 1 LlJWhere x’, ’, z are the 3D coordinates of a point in the standardized image 482, and x, y, z are the 3D coordinates of the corresponding point in the input image (e.g., the MR image 480 mnmi2mi3mof the subject), and21m22m23m24m31m32m33m34is the transformation matrix (M) 481 and where mi4, 0 0 0 1 Jni24, 1U34 are translation parameters in x-, y -and z- directions and the remaining 3 x 3 matrix ([m-i ]; i, j = 1,2,3) represents the rotation, scaling, and shearing parameters.

[0069] Eqn. 1 can be simplified to Eqn. 2:S = MP ...Eqn. 2'x''Where S' = y' or the coordinates of the standardized image 482 and P y or thez' z .1.-1-coordinates of the input image 480 of the subject.

[0070] The transformation matrix M 481 can be found for each subject and for every image and can be stored in the memory 375 of the atlas builder system 370 of Fig. 3.

[0071] Example methods of the present disclosure are not limited to those illustrated by Figs.4A-4B and / or 5A-6B. An example method can comprise: (i) processing a response of a subject to focused ultrasound applied to a specific region within a brain of the subject, wherein the response includes or corresponds to a biomarker, (ii) associating information1690.105.111from the focused ultrasound to coordinates of the specific region within the brain of the subject, and (iii) based thereon, building a composite biomarker atlas of the brain of the subject. In some embodiments, the information includes parameters of sonication and / or the response of the subject to the coordinates. For example, the method can include associating parameters of sonication and / or the response of the subject to the coordinates. In some embodiments, the method includes associating the information from coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain, as previously described.

[0072] In some embodiments, the biomarker is a chemical biomarker. In some embodiments, the biomarker is an electrical biomarker. In some embodiments, the biomarker is a mechanical biomarker. In some embodiments, the response is multi-modal.

[0073] In some such embodiments, the method can further include applying the focused ultrasound to the specific region within the brain of a subject, and processing and analyzing the response to the focused ultrasound and, in response, associating the information including results of the analysis to the coordinates of the specific region. In some embodiments, the focused ultrasound is applied non-invasively. In some embodiments, the method can comprise applying focused ultrasound to the specific region within the brain of the subject via a sonication procedure which is uninterrupted in response to subject motion such that the specific region of the brain is sonicated independent of the subject motion.

[0074] In some embodiments, the method comprises processing and analyzing the response by comparing the response to sonication applied to the specific region of the brain of the subject and a response to sonication applied to a specific region of a brain of the another subject.

[0075] In some embodiments, the method comprises calculating a location of the sonication in the subject based on a corresponding location in a standardized image of a brain.

[0076] In some embodiments, the method comprises obtaining parameters of sonication for a particular location of the specific region in the brain of the subject from the information associated with the location in the composite biomarker atlas. For example, the parameters of sonication can be obtained from the composite biomarker atlas and optionally are modified on a subject-to-subject basis.

[0077] In some embodiments, the method comprises building the composite biomarker atlas for the brain over multiple subjects, including the subject, and over a single sonication session or multiple sonication sessions for each of the multiple subjects.1690.105.111

[0078] Fig. 5A is a flow chart illustrating an example method for building a composite chemical biomarker atlas. A flow chart 583 of some of the steps to build a composite chemical biomarker atlas is now described in Fig. 5A.Process of building the atlas for chemical biomarkers

[0079] At 584 in Fig. 5A, the MR images or other images of the subject are received by the atlas builder system. The MRI or other images may be sent to the atlas builder system over the hospital network and / or by the ultrasound device. At 585, a transformation matrix is generated that transforms the image of the subject to a stored standardized image (e.g., a target image). While Fig. 5A shows only one transformation, embodiments include generating multiple transformation matrices, each associated with a different standardized image. At 586, the atlas builder system receives information, such as the location of the BBB opening from the ultrasound device or from the information stored in the hospital network. If the location information comes from the ultrasound device, then the registration of the face of the subject to their brain image can be performed, as described above. In some such embodiments, the image of the subject can be received from the ultrasound device itself and the coordinates for the BBB opening location and the images may be in reference to the same coordinate system. If the image(s) of the subject are transferred from the hospital network and the BBB location is received from the ultrasound device, then a transformation matrix as described above may be used to convert the coordinates of the image(s) of the subject and the BBB opening location to the same frame of reference. As described above, the ultrasound device can send the confirmed BBB opening location; however, in some embodiments, the planned BBB opening location can also be transferred via the hospital network. At 587, the atlas builder system can receive information from the chemical biomarker analysis system. The process of inputting the results of the analysis into the atlas builder system, whether it is automatic, manual, or combined, can ensure that the results become associated with the correct subject and the correct location of BBB opening. At 588, the coordinates of the BBB opening location input at 587 are transformed using the transformation matrix calculated at 585. This step can assist with verifying that the coordinates of the BBB opening location are transformed into the coordinates of the standardized image. At 589, the chemical biomarker information can be associated with the location calculated at 588. At 590, the atlas builder system may allow for the input of labels or tags by a user for the particular result for the particular subject. These labels or tags can be useful when implementing a machine learning or other artificial intelligence model.1690.105.111

[0080] The process outlined in Fig. 5A may thus be used for building an example chemical biomarker atlas, in various embodiments.Motion-compensated location-specific liquid biopsy

[0081] In some embodiments, the motion of the head of the subject relative to the ultrasound cap can be compensated for during the process of opening the BBB with focused ultrasound. An example technique of motion-compensation for BBB opening with focused ultrasound and microbubbles is described in US Patent Publication No. US2024 / 0042242, published on February 8, 2024, and entitled “Rapid calculation of parameters for delivering ultrasound energy to selected locations in the brain”, which is incorporated herein by reference in its entirety for its teachings. When the BBB is opened for the purpose of obtaining a liquid biopsy of a subject, during the process of sonication, the motion compensation technique ensures that the region of BBB that is being sonicated remains consistent within acceptable margins despite subject motion. This ensures the reliability and consistency of the data in the composite biomarker atlas.Building the atlas over multiple sonication sessions for a. subject for a location for a disease

[0082] During the course of treatment associated with a specific disease, it may become necessary or otherwise beneficial to know the chemical biomarker content (or other biomarkers) on an ongoing basis. For the purposes of understanding the trends in the chemical biomarker content (or other biomarker content), it may also be necessary or otherwise beneficial to open the BBB at the same location on an ongoing basis. For the initial or first sonication session for any particular location for a subject, the region of the opening of the BBB may be specified manually by a user (e.g., physician) or automatically using tools, such as those based on machine learning and / or artificial intelligence. The region of opening of the BBB can be specified in the coordinate system associated with the MR image of the subject or other images obtained with other modalities. The location of the BBB opening can be stored along with other subject data in databases maintained by the medical institution, such as via hospital network 247 of the system 230 of Fig. 2A. For subsequent sessions, if the same region of the brain is monitored, the location of the BBB opening can be provided to the ultrasound device by retrieving the opening location from the database. Since the ultrasound device 246 may be connected to the hospital network 247, as indicated in Fig.2A, the location of the BBB opening can be retrieved through the use of the network 247. Thus, for a single subject, multiple BBB opening procedures can be performed at the same1690.105.111location within the brain, and the process outlined in Fig. 5A can be repeated each time the BBB is opened. In some embodiments, the transformation matrix described at 585 in Fig. 5A may be stored in memory, such as at 375 in Fig. 3 in the atlas builder system 370 for each subject for that specific location and is used at 585 instead of recalculation of the matrix each time the process is applied.

[0083] In some embodiments, the chemical biomarker content obtained over multiple sonication sessions for the same location for the same subject, provides a way to determine the changes to the chemical biomarkers associated with a specific location. If the subject is in therapy and, for example, undergoing radiation or taking medications or both, the changes in the chemical biomarker content may provide an indication of whether the therapy is effective. If the therapy is effective, the chemical biomarker content towards the end of the therapy or when the disease is stable may indicate what “normal” chemical biomarker content is.Building the atlas over single or multiple sonication sessions over multiple subjects

[0084] The concept of building the composite biomarker atlas for a single subject for a specific location for a specific disease can be extended to adding data from multiple subjects for the same disease at the same or different locations.Building the composite brain atlas

[0085] Over multiple subjects, over multiple sonication sessions, and / or over different diseases, comprehensive chemical biomarker maps and / or atlas may be generated in some embodiments. As noted above, embodiments may be directed to determining what “normal” chemical biomarker content for every location of the brain may be. In some embodiments, a wide variety of such composite biomarker atlases can be created. As an example, a composite chemical biomarker atlas can be created by using a standardized MR image, as described above, and associating a location (e.g., one or more) within the image with the “normal” biomarker content for that location. The “normal” biomarker content may be the common set of chemical biomarkers detected for that location over multiple subjects.Analysis system

[0086] The analysis system 376 in Fig. 3 may be or form part of a computing device and / or system with storage. Various different analysis can be performed to build the atlas and to provide data once an atlas is built. For example, as mentioned above, a composite chemical biomarker atlas of a normal, healthy brain can be built. Over several subjects, an age-specific1690.105.111composite chemical biomarker atlas may be created. Other examples of composite biomarker atlases include those that can be built for different diseases. Once built, the composite biomarker atlases can be used as a reference to compare the specific biomarker content from a specific subject to the various atlases; this type of analysis may be useful to determine the exact nature of the disease.Use of the atlas builder system

[0087] Users of the atlas builder system 370 may include physicians, researchers, insurance companies, data analysts, subjects, academic institutions, nurses, physician assistants, etc. The atlas builder system 370 can be used in various ways. As an example, after the composite biomarker atlas is created, a user can recall a volumetric chemical biomarker atlas for a particular disease occurring at a particular location from memory; if a graphical user interface is utilized, the volumetric chemical biomarker atlas may be displayed in a multiplanar reconstruction (MPR) or MPR mode. The user can scroll through the various planes and click at a location and the chemical biomarker content at that location may be displayed as a table on the graphical user interface. The atlas builder system 370 can have a search function that the user can use to recall the appropriate chemical biomarker atlas or other composite biomarker atlases.

[0088] As another example, a user can compare the chemical biomarker content associated with a particular subject from a particular BBB opening location to chemical biomarker content in the composite biomarker atlas at that location from healthy brains. As a further example, the comparison may be made to chemical biomarker content from brains that have the same disease. In some embodiments, if the data in the composite biomarker atlas has chemical biomarker content as a function of disease progression, a more accurate picture of the disease status of that particular subject may be determined. Similarly, if data exists in the composite biomarker atlas in relation to responders and non-responders of therapy to that particular disease, then the chemical biomarker status of the subject may indicate how the particular subject may be responding if the subject is on the same therapy. These are only some examples of how the composite biomarker atlas may be used.Use of the composite atlas builder to determine the best parameters for BBB opening

[0089] In some embodiments, and referring to back to Fig 2A, in the process of building the composite biomarker atlas, the parameters of sonication used by the ultrasound device 246, such as in-situ (estimated) pressure level, frequency of sonication, sonication duration, pulse1690.105.111length, pulse repetition frequency, etc., can be sent by the ultrasound device 246 to the atlas builder system 250 along with the planned or confirmed BBB opening location. As data from multiple subjects can be are entered, the atlas builder system 250 can be queried for a set of sonication parameters that are most commonly used for opening the BBB at a specific location. As a safe opening of BBB is desirable, the parameter set that is most commonly used to obtain such an opening can be beneficial to know. As described above, a search function can help to define a search space. For example, a user may want to know the parameters of sonication to open the BBB safely at the temporal lobe in adults aged 60 years. The parameter set may automatically be sent to the ultrasound device 246 from the atlas builder system 250 and may be used for BBB opening. When the most common parameter set for a particular situation is received by the ultrasound device 246, the parameters of sonication may be adjusted to accommodate the particular characteristics of the subject 210. For example, when the ultrasound device 246 estimates the in-situ acoustic pressure and sends this value to the atlas builder system 250, the attenuation of the skull, scalp, and brain tissue can be assumed or estimated. The acoustic pressure at the surface for each of the transducers for a given applied voltage can be known through measurements, simulation, or both. The acoustic path from the transducers to the region where the BBB is to be opened can be known during the process of registration, as previously described. With the knowledge of the acoustic pressure at the surface, the acoustic path, and the attenuation of the intervening tissue (skull, scalp, and brain tissue), the in-situ acoustic pressure can be calculated as the attenuation is specified in dB / unit length. In general, the in-situ acoustic pressure that opens the BBB safely across humans is the same within a range; for example, at 250 kHz, the range is approximately 0.3-0.4 MPa. Thus, over multiple subjects, better estimates of this range can be calculated, and an average value can be calculated. Then, for a particular subject, this average value may be sent to the ultrasound device 246; however, to calculate the voltage to be applied to the transducers to develop that average in-situ pressure (for that subject), the particular characteristics of the subject 210, such as thickness and attenuation of the skull, can be incorporated in the calculation. As described above, in some embodiments, the CT image of the subject can be available to the ultrasound device 246. The CT image can indicate how thick the skull is. The conversion between the brightness of CT image and sound speed in the skull (needed to calculate the acoustic path and, thus, the attenuation) has been established in the literature. Thus, this is an example where the average parameters of sonication are adjusted for a particular subject.1690.105.111

[0090] In various embodiments, a transformation matrix can be calculated to determine where to apply sonication on an actual subject from the information in the standardized image using a process called inverse transformation. Fig. 4B illustrates this process. Here, an inverse of the matrix computed in Fig. 4A is calculated as defined below.M x M-1= AT1X M = I ... Eqn. 3M~l= — Adj M... Eqn. 4\M |J MWhere M~ 483 is the inverse of M, \M\ is the determinant of M, and Adj M is the adjoint of M and I is the identity matrix. The determinant of any matrix is calculated as the sum of the products of the elements of any row or column and their respective cofactors. The adjoint of any matrix is the transpose of its cofactor matrix. The cofactor of an element is the minor of that element multiplied by (~l)i+-’ , where i and j are the row and column indices of the element, The minor of an element in a matrix 483 is the determinant of the matrix 483 formed by removing the row and column of that element.

[0091] Thus, using this process, the coordinates in the input image 480 of a subject can be calculated to correspond to the coordinates in the standardized image 482. The coordinates of a region or location in the standardized image 482, such as Pj shown in the standardized image in Fig. 4B, can be transformed into the coordinates of the corresponding region in the image of the subject, such as PR'. When the corresponding coordinates in the input image 480 of the subject are calculated, the location of how to guide the ultrasound to that location from outside the skull can be calculated, as further described below.

[0092] In some embodiments, the composite biomarker atlas can be used to calculate the parameters of sonication, which can include where sonication is to be applied, based on the information in the atlas.

[0093] Some embodiments include or allows for the comparison of results of BBB opening from subject-to-subject as the same region of the brain (as defined by the corresponding region to the standard image) is sonicated across subjects. In particular, if the same parameters are used for excitation across subjects, the comparison may be more accurate.

[0094] In some embodiments, the parameters of sonication may be adjusted on a subject-to-subject basis based on the MR and CT images of the respective subject, such that the estimated acoustic pressure experienced by the specific location in the brain of the subject is the same within a certain predefined range. The CT and MR images can help determine the attenuation that the ultrasound may experience as it travels from the ultrasound cap (232 as shown in Fig. 2A) to the specific location within the brain of the subject. The brightness of1690.105.111the CT images provides information about the speed of sound through the skull. The brightness also provides the thickness of the bone. Thus, using the CT and MR images, the acoustic path between the transducers in the ultrasound cap at the region may be determined; the attenuation of sound as it travels through the bone and soft tissue may also be assumed and preprogrammed in the ultrasound device 246 in Fig. 2A.Building or using the composite chemical biomarker atlas, including preplanning and sonication steps and use of the composite biomarker atlas

[0095] Fig. 6 A illustrates an example flow chart for building or using the composite chemical biomarker atlas, including preplanning and sonication steps and use of the composite chemical biomarker atlas. As previously described, a composite biomarker atlas may include or refer to an atlas that uses a standardized image that represents a brain to which response information is added on a location-by-location basis for a location. The information may be based on one or multiple subject. At 603 of the flow chart 601, a user (e.g., physician) orders scans for a subject, such as based on symptoms. For brain-related issues, MRI scans or CT scans or other types of scans may be ordered. At 605, the user may order a liquid biopsy of the brain of the subject to determine the disease status based on the chemical biomarker content. The liquid biopsy may be performed with a focused ultrasound approach with microbubbles. With the ultrasound device described in Fig. 2A, a non-invasive liquid biopsy may be performed. At 607, the preplanning step, in sub-step 1, the same or different user may specify the region where the BBB is to be opened. This sub-step may be performed on the ultrasound device, or it may be done on a computing device attached to the hospital network where programs for performing this sub-step are installed. The MRI and CT data can be used for this step and can be transferred from the hospital network.

[0096] In sub-step 2 at 607, the ultrasound device can calculate the initial parameter set (e.g., sonication parameters), including which transducers of the ultrasound cap are to be activated and the amplitude and phase or timing of activation according to functions programmed within the ultrasound device. In sub-step 3 at 607, if the BBB is being opened at the same location of the brain in a repeated procedure (i.e. , not for the first time), the initial parameter set of sub-step 2 may not be calculated. In sub-step 4 at 607, if the composite biomarker atlas is already built, the parameters of sonication for the BBB opening, such as in-situ (estimated) pressure level, frequency of sonication, sonication duration, pulse length, pulse repetition frequency, can be requested by the ultrasound device from the atlas builder system and adjusted for the subject, as described above.1690.105.111

[0097] The flow chart 601 at 609 includes various sub-steps involved in sonication. In substep 1 at 609, a pre-sonication blood draw may be obtained through a needle or a catheter placed in a vein, e.g., in the arm (peripheral site). This can establish a baseline level of chemical biomarkers already present in the blood. In sub-step 2 at 609, the subject can be situated in a chair or on a bed, and the ultrasound cap is arranged on the head of the subject. In sub-step 3 at 609, the ultrasound device can perform additional calculations, including the registration step, as described above. The subject can be injected with microbubbles in this sub-step. This step may also include the calculations related to confirming the location of ultrasound focus as described in US Patent No. 12,017,093. In sub-step 4 at 609, the sonication to open the BBB is commenced. As described above, this sub-step may involve compensating for the motion of the subject so that the sonication can continue despite the motion. In sub-step 5 at 609, after sonication, a post-sonication blood draw may be obtained, such as from or near the same peripheral site as used in the pre-sonication blood draw. The blood samples can be sent to laboratories and / or to a chemical biomarker analysis system for chemical biomarker analysis. In sub-step 6 at 609, the ultrasound device sends information to the atlas builder system, including the parameters of sonication and confirmed or planned BBB opening locations.

[0098] The processing involved at 611 is described above in connection to Fig. 5A.

[0099] At 613, the results of the chemical biomarker analysis for the subject may be compared to other reference values stored in the composite biomarker atlas associated with the specific disease that the subject has and associated with the specific location of BBB opening. As described above, the results may be accessed via a computer display. The results may be displayed graphically or in a textual form.Application to Neuromodulation

[0100] In ultrasound neuromodulation, the BBB is not opened with focused ultrasound. Rather, low-intensity focused ultrasound is used to non-invasively stimulate portions of the brain to stimulate or suppress neural activity in the brain. The study of ultrasound neuromodulation is complex because of the number of variables that can affect the neuromodulatory response given an ultrasound stimulus. Variables can include disease states, the location of the brain that is stimulated, the parameters of the sonication for the stimulation, the type of response being studied, etc. With the vast array of variables, a composite biomarker atlas may be useful so users can refer to the composite biomarker atlas for reference and / or for commonly observed responses for a given stimulus.1690.105.111

[0101] Fig. IB illustrates an example of a composite biomarker atlas 124 for a neuromodulation application. The composite biomarker atlas 124 shown in Fig. IB shares at least some of the same characteristics and features as the composite biomarker atlas 104 shown in Fig. 1 A, except that the configuration of the data and the data itself may be different. While in Fig. 1 A, the biomarker data can be chemical in nature, in Fig. IB, the biomarker data is electrical in nature. In some embodiments, the specific electrical signals of interest for the neuromodulatory application can include the electrical activity within the brain and the signals from the brain to various muscles, such as the muscles to the arm. To capture and analyze the electrical activity, Fig. IB shows two physiology sensors 128-1, 128-2 (SI and S2). These sensors 128-1, 128-2 can be electrodes, in some embodiments. In some embodiments, there may be only one sensor or more than two sensors 128-1, 128-2. While Fig. IB shows the sensors 128-1, 128-2 coupled to the head of the subject 110, the sensors 128-1, 128-2 may be coupled to other parts of the body, as shown in Figs. 2A-2B. The sensors 128-1, 128-2 can record the electrical activity as experienced by the portion of tissue the sensor 128-1, 128-2 is coupled or attached to (typically by some adhesive material and can be detached).

[0102] Map 1 122-1 is an example biomarker map for an ultrasound stimulus applied to the location Pl. The leftmost column 126-1 (column 1) contains a description of the various stimulations. Thus, stimi(ti) depicts the ultrasound stimulation one (1) applied at time ti at location Pi. The details of the stimulation (e.g., sonication) may be stored in a graphical form or in a textual form, and can include estimated in-situ pressure, frequency, pulse length, pulse repetition frequency, sonication time, etc. Similarly, stintzfe) is another ultrasound stimulation applied at a different time t2. Stim2(t2) may have the same or different parameters of sonication as stimi(ti). Columns 2 through m 126-2, 126-M contain the response S l(pi,ti), S2(pi,ti)... Sm(pi.ti) as measured by the sensors 128-1, 128-2 (e.g., SI, S2... Sm). These may be highly sampled waveform data as received by the sensors 128-1, 128-2. Metadata about the waveforms, such as the sampling rate, may be stored with the sampled waveform data so that the waveform can be reconstructed later.

[0103] In addition, in some embodiments, although the notation of the responses indicates that the responses are measured at specific instances such as ti, each entry in the columns 126-1, 126-2...126-M may be sampled time series extending over a period of time that starts at time ti. Thus, a stimulus can be applied at ti at location Pi, and the response is measured, digitized over a period of time (which may be predefined or programmable), and stored as a sampled waveform.1690.105.111

[0104] In some embodiments, the data (or waveform) from each sensor 128-1, 128-2 can be stored in different columns. Thus, the composite biomarker atlas 124 comprising electrical biomarkers can be built starting with one subject and one brain location but can be extended over multiple subjects and multiple brain locations.

[0105] The composite biomarker atlas 124 can be built for specific conditions, such as disease conditions, age, etc. As with Fig. 1 A, the composite biomarker atlas 124 can include biomarker maps 122-1, 122-2 for multiple locations (Pl, P2) which can be constructed in a similar manner (i.e., the axes, parameters, and size of the matrix are all identical while the values in each cell may be different), but in other examples, the biomarker maps 122-1, 122-2 may be different from each other.

[0106] Fig. 2B illustrates another example system that includes an ultrasound device configured to apply sonication to a specific location within the brain of a subject. Many aspects and features of the system 260 illustrated in Fig. 2B are the same as the system 230 illustrated in Fig. 2A, but some differences now described. Fig. 2B shows physiologic sensors 268-1, 268-2 coupled or attached to the subject 210. In some embodiments, the sensors 268-1, 268-2 can be removably coupled with a biocompatible adhesive. The physiologic sensors 268-1, 268-2 can be electrically connected to a physiologic waveform processing system 262, which can sample the waveforms using the sensors 268-1, 268-2, store the waveforms, and send the waveforms along with other metadata (such as sampling rate) to the atlas builder system 250, at 263. Fig. 2B does not show a blood draw component, although, in some embodiments, a blood or other fluid draw step may be present. The ultrasound cap 232 in Fig.2B can provide the ultrasound stimulus, but typically, for neuromodulation studies, a BBB opening is not necessary; hence, the microbubble delivery system is not shown.

[0107] The physiological waveform processing system 262 can be configured to capture a physiologic response. In some embodiments, the physiological waveform processing system 262 can include a processor and memory (such as described above) configured to detect and analyze electrical biomarkers which include or are indicative of a physiological waveform, such as an electroencephalogram (EEG) and / or magnetoencephalogram (MEG). The electrical biomarkers can thus be detected using physiological sensors 268-1, 268-2.

[0108] A registration device 248 is shown in Fig. 2B. In cases where the MR and / or CT images are present for a subject 210, the registration step can be the same as described in the context of Fig. 2A. In addition to the registration of the face to the CT image, in Fig. 2B, the location of the sensors 268-1, 268-2 may also be registered. Thus, for example, a EM pointer sensor 238 can be used to point to and enter the coordinates of each of the sensors 268-1,1690.105.111268-2 in the physical world with respect to the home base of the registration device 248. Once the coordinate pointing process is complete, as described above in connection with Figs. 4A-4B, a transformation matrix can calculated between the MR image of the subject 210 and a standardized MR image. The physical world coordinates of the sensors 268-1, 268-2 can also be transformed using the same transformation matrix to the coordinates in the frame of reference of the standardized image.

[0109] Another technique that does not use the EM pointing technique may be used to enter the coordinates of the sensors 268-1 , 268-2 in the coordinate frame of the standardized image. In some embodiments, a synthetic human head derived from the standardized MR image can be displayed on a computer display (e.g., a screen). Software such as “mri-reface” can be used for this derivation. A 3D-rendered image of the synthetic human head may be displayed on the computer display. Next, an user may place the sensors 268-1, 268-2 on the subject 210 and, using computer tools, indicate on the synthetic head where the sensors 268-1, 268-2 are placed. A standardized image of the full body of the subject 210 may also be displayed, and the same technique may be used to enter the coordinates of the sensors 268-1, 268-2 that are placed on other parts of the body other than the head. Thus, with these steps, the position of the physiologic sensors 268-1, 268-2 may be entered into the atlas builder system 250.

[0110] For the system 260, motion-compensated stimulation can be performed in a similar manner as described above. The difference is that the BBB may not be opened but the technique to compensate for subject motion and continue to sonicate the same region of the brain can be implemented, as described above.

[0111] The composite biomarker atlas for neuromodulation can be built with a similar process as described above for a single subject for a specific brain location over one or multiple sonication sessions (without BBB opening). Also, the same techniques can be used to build the composite biomarker atlas over multiple subjects for the same location in the coordinate frame of reference in the standardized image over one or multiple sonication sessions.Building the composite biomarker atlas for electrical biomarkers for the neuromodulation application, including preplanning and sonication steps and use of composite atlas

[0112] Fig. 5B is a flow chart illustrating an example method for building a composite electrical biomarker atlas. The flow chart 591 to build the composite biomarker atlas for the neuromodulation application with electrical biomarkers is shown in Fig. 5B and is similar to the atlas building process for the chemical biomarkers shown in Fig. 5A. While the steps1690.105.111remain the same in Fig. 5A and Fig. 5B, some details with some of the steps may be different, and are described below.

[0113] At 592 in Fig. 5B, the MR image of the subject or other images are received by the atlas builder system. The MR image or other images may be sent to the atlas builder system over the hospital network and / or by the ultrasound device. At 593, a transformation matrix is generated that transforms the image of the subject (input) to a stored standardized image (target image). While Fig. 5B shows only one transformation, embodiments include generating multiple transformation matrices, each associated with a different standardized image. At 594 in Fig. 5B (as compared to 586 in Fig. 5A), the information sent by the ultrasound device is related to where the simulation is being applied within the brain of the subject. At 595 in Fig. 5B (as compared to 587 in Fig. 5A), the information received by the atlas builder system is from the physiologic waveform processing system. For example, this information can be the electrical waveforms as sensed by the sensor (268-1, 268-2). At 596, the coordinates of the ultrasound stimulation are calculated in the frame of reference of the standardized image using the transformation matrix calculated at 593. At 597, the electrical biomarker information (or the waveforms for the various physiologic sensors) is associated with the stimulation location in the standardized image. Thus, at the end of this process, the electrical biomarker and other information, whether entered by a user or sent by the ultrasound device, is associated with the stimulation location. In some embodiments, at 593, a user can add labels or tags to the biomarker information, as described above.Building the composite biomarker atlas over multiple time points and over multiple subjects

[0114] Similar to the process for the chemical biomarkers, the composite electrical biomarker atlas can be built for the same subject at one or multiple locations of the brain over one or multiple sonication sessions. Then, the atlas can be built over multiple subjects (with the same or different locations relative to the standard image), each with one or multiple sonication sessions.Building or using the composite electrical biomarker atlas

[0115] Fig. 6B illustrates an example flow chart for building or using the composite electrical biomarker atlas, including preplanning and sonication steps and use of the composite electrical biomarker atlas. The flow chart 615 for building or using the composite electrical biomarker atlas is similar to that described for the chemical biomarker atlas of FIG.6A. Fig. 6B illustrates the process of electrical biomarkers - only some of the differences are1690.105.111described. At 617, a user orders scans for a subject based on symptoms. For brain-related issues, MR images or CT scans or other types of scans may be ordered. At 619, the user orders an ultrasound stimulation procedure. At 621 (which is similar to 607 in Fig. 6A), in sub-step 2, the parameters of sonication (for the ultrasound stimulation) are calculated. At 623 (which is similar to 609 in Fig. 6A), in sub-step 1, a presonication set of electrical waveforms are captured and stored. In sub-step 4 at 623, the focused ultrasound stimulation is applied (with or without microbubbles). In sub-step 5 at 623, a post-sonication set of electrical waveforms from the physiologic sensors is captured and stored. At 625, the processing illustrated in Fig. 5B is carried out. At 627, while in the chemical biomarker atlas, a user can click a location with the standardized image and obtain the chemical biomarker content; for the neuromodulation application, electrical waveforms may be displayed.Use of the composite atlas builder system to determine parameters for neuromodulation

[0116] Similar to how the parameters of sonication are calculated in relation to the BBB opening application, the same technique can be used to calculate the parameters of sonication and the location where the sonication is applied for the neuromodulation application. This allows for the comparison of results of neuromodulation from subject-to-subject as the same region of the brain (as defined by the corresponding region to the standardized image) is sonicated across subjects. In some embodiments, if the same parameters of sonication are used for excitation across subjects, the comparison may be more accurate. In addition, if the parameters are adjusted so that the same in-situ estimated acoustic pressure is experienced by the same location in the brain across different subjects corresponding to the same location in the standard atlas, the comparisons may be more accurate.Multi-modal composite biomarker atlas and atlas operations

[0117] Fig. 7 illustrates an example use of a composite biomarker atlas. As described above, the composite biomarker atlas may contain only one type of biomarker information, e.g., is unimodal, or it may contain multiple types of biomarker information, e.g., is multimodal. Since the composite biomarker atlas or atlases may be using the same standardized image, logical operations may be performed on the content of the composite biomarker atlases. As described above, a user using the atlas builder system can scroll through various planes and click at a location to display the biomarker content. Clicking at a location may provide the user with the choice of displaying one or multiple biomarkers. Fig. 7 illustrates an example user interface 731 of the atlas builder system used to build an composite biomarker atlas that1690.105.111is multi-modal. A composite biomarker atlas or response being multi-modal, as used herein, refers to and / or includes a response to focused ultrasound which includes more than one type of biomarker, such as different combinations of chemical, mechanical, and / or electrical biomarkers. On the left side of the user interface 731 , a multiplanar image 733 of the standardized image is illustrated. Panels A, B, C are orthogonal planes through the standardized image (which may be volumetric). The volume-rendered image is shown in panel D. A user may click on any of the panels and scroll through the planes. A cursor is shown with the plus symbol in the panels. The right side of the user interface 731 illustrates the biomarker information 735 for the selected point (selected by the cursor). The atlas builder system may allow Boolean and other operations to let a user select the information that is displayed. As an example, if the biomarker information 735 associated with the standardized image 733 is based on population data (i.e., based on biomarkers seen over a large number of subjects), the user may want to display only the chemical biomarker content. In another example, the user may want to display the chemical biomarker and the electrical biomarker (e.g., EEG). As another example, a user may want to display the EEGs only from sensors placed on the frontal lobe. These examples demonstrate the ability of the atlas builder system to allow the user to customize the display.Architecture of the atlas builder system

[0118] Fig. 8 illustrates example blocks of an atlas builder system. In particular, Fig. 8 shows blocks that can form the basis of computer-readable instructions (e.g., software) and hardware architecture 841 of the atlas builder system. This architecture 841 can enable artificial intelligence (Al) and / or machine learning (ML) using the biomarker information as the basis. The architecture 841 may support a data services pipeline 843 that may include a data service block 845 that can exchange data from various devices and systems such as the ultrasound device 846 and external laboratories 848 or other biomarker analysis systems. The data services pipeline 843 may also include an analysis capability such as the biomarker panel analysis block 847, where the results from the analysis of blood or other fluids may be accepted and further analyzed. The image processing and labeling block 849 may exchange images and data between the atlas builder system and external image labeling services 850. For example, this pathway may be used to accept MR and CT images of a patient from the hospital network. The atlas builder system can also interact with the hospital network, such as with the Electronic Health Record (EHR) systems 852 and can pass messages and data back and forth and via the EHR ingesting and processing block 851. The data services pipeline 8431690.105.111may interact with a data platform 853 which may receive, store, and associate various pieces of data 855, 857, 859, 861, such as the ultrasound data reflecting from the microbubbles, location of the BBB opening or sonication, data from sensors, MR images, CT images, biomarker data, and deidentified patient data, which may be received from various external systems 875. The data platform 853 may interact with an Al block 863, 869. In some embodiments, at least one Al block 863, 869 may be a proprietary Al block 869 which may include the composite biomarker atlas 873 as described above. The Al block 869 may include tools that have been trained on the composite biomarker atlas, such as prognostic and clinical data support (CDS) tools 871. Interactions between the data services pipeline 843 and the proprietary Al block 869 may also occur to external Al block 863 with build Al models 865, 867 using the data in the data platform 853. Some portions or all of the architecture 841 of the atlas builder system may be hosted in a cloud-based service tool 877. Users 879 such as physicians, hospitals, insurance representatives, data analysts, patients may interact with the architecture directly or through a cloud-based service tool 877. Additional tools that the cloud based services offer may be used to further access or analyze the data or the atlas.

[0119] Machine learning models can encompass predictive data models that estimate or generate outputs based on input data. Various machine learning frameworks are available from multiple sources, offering open-source machine learning datasets and tools to facilitate the design, training, validation, and deployment of machine learning models. These models can be executed on artificial intelligence (AI) / machine learning (ML) processors, which are specialized hardware accelerators, such as Neural Processing Units (NPUs) or Machine Learning accelerators (MLAs). These AI / ML processors are integrated circuits (ASICs) that can have multi -core architectures that employ precision processing, optimized dataflow, and efficient memory use to accelerate computations and enhance throughput in processing machine learning models.

[0120] Example machine learning models include artificial neural network, support vector machine (SVM), deep learning, cluster, and / or other models. An artificial neural network can estimate a function(s) that depends on inputs. In some embodiments, one or more layers of artificial neurons can receive input data and generate output data. Neural networks can include networks such as, but not limited to, learning networks (e.g., deep, deep structured, hierarchical, and the like), convolutional, auto-type networks (e.g., autoencoder, auto-associator), Diablo networks, and neural network models (e.g., feedforward, recurrent).

[0121] An SVM can utilize a linear classification. This classification can act to separate the data points into classes based on distance of the data points from a hyperplane. In some1690.105.111embodiments, the hyperplane is arranged to maximize the distances from the hyperplane to the nearest data points on either side of the hyperplane. This arrangement can group points located on opposite sides of the hyperplane into different classes. However, in some embodiments, the SVM can include a nonlinear classification that separates the data points with a hyperplane in a transformed feature space. The transformed feature space can be determined by one or more kernel functions, including nonlinear kernel functions. In some embodiments, the SVM is a multiclass SVM that separates data points into more than two classes, which can reduce a multiclass problem into multiple binary classification problems.

[0122] In some embodiments, a deep learning model can include models such as, but not limited to, convolutional networks (e.g., deep belief, neural), belief networks, Boltzmann machines, deep coding networks, stacked autoencoders, stacking networks (e.g., deep or tensor deep), hierarchical-deep models, deep kernel machines, and the like. It will be understood that such embodiments can include variants and / or combinations of the abovenoted example networks.

[0123] In some embodiments, the machine learning model(s) can include a clustering method(s), which can include hierarchical clustering, k-means clustering, density-based clustering, and the like. In some embodiments, the hierarchical clustering can be used to construct a hierarchy of clusters of the set of features. In some embodiments, the hierarchical clustering utilizes a “bottom up” approach (e.g., agglomerative) wherein each data point starts in its own cluster, and pairs of clusters are merged at progressively higher levels of the hierarchy. However, in some embodiments, the hierarchical clustering utilizes a top-down approach in which all data points start in one cluster, and then clusters are split at progressively lower levels of the hierarchy.

[0124] In some embodiments, the k-means clustering implementation can include placing the set of features into k clusters, where k is an integer equal or greater than two. Via such clustering, each data point belongs to a cluster having a mean that is closer to the data point than any means of the other clusters. However, in some embodiments, a machine learning model can include density-based clustering, which can be used to group together data points that are close to one another, while identifying as outliers any data points that are far away from other data points.

[0125] In some embodiments, a machine learning model can include a mean-shift analysis that can be used to determine the maxima of a density function based on discrete data sampled from that function.1690.105.111

[0126] In some embodiments, a machine learning model can include structured prediction techniques and / or structured learning techniques. Such techniques can be used to predict structured objects and / or structured data, such as structured sets of features and / or sensor data. In some embodiments, the structured prediction and / or structured learning techniques can include graphical models, probabilistic graphical models, sequence labeling, conditional random fields, parsing, collective classification, bipartite matching, Bayesian networks or models, and the like. It will be understood that such examples include variants and / or combinations of the example techniques.

[0127] In some embodiments, a machine learning model can include anomaly detection and / or outlier detection that can be used to identify data that does not conform to an expected pattern or are otherwise distinct from other data in a dataset.

Claims

1690.105.111CLAIMS1. A method comprising :processing a response of a subject to focused ultrasound applied to a specific region within a brain of the subject, wherein the response includes or corresponds to a biomarker;associating information from the focused ultrasound to coordinates of the specific region within the brain of the subject; andbased thereon, building a composite biomarker atlas of the brain of the subject.

2. The method of claim 1, wherein associating the information from the focused ultrasound to coordinates of the specific region within the brain includes associating parameters of sonication and / or the response of the subject to the coordinates.

3. The method of claim 1, wherein associating the information from the focused ultrasound to coordinates of the specific region within the brain includes associating the information from coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain.

4. The method of claim 1, comprising:applying the focused ultrasound to the specific region within the brain of a subject; andprocessing and analyzing the response to the focused ultrasound and, in response, associating the information including results of the analysis to the coordinates of the specific region.

5. The method of claim 4, wherein the focused ultrasound is applied non-invasively.

6. The method of claim 1, comprising applying focused ultrasound to the specific region within the brain of the subject via a sonication procedure which is uninterrupted in response to subject motion such that the specific region of the brain is sonicated independent of the subject motion.

7. The method of any one of claims 1-6, wherein parameters of sonication are obtained from the composite biomarker atlas and optionally are modified on a subject-to-subject basis.1690.105.1118. The method of any one of claims 1-6, comprising building the composite biomarker atlas for the brain over multiple subjects, including the subject, and over a single sonication session or multiple sonication sessions for each of the multiple subjects.

9. The method of any one of claims 1-6, wherein the biomarker is a chemical biomarker.

10. The method of any one of claims 1 -6, wherein the biomarker is an electrical biomarker.

11. The method of any one of claims 1-6, wherein the response is multi-modal.

12. The method of any one of claims 1-6, comprising processing and analyzing the response by comparing the response to sonication applied to the specific region of the brain of the subject and a response to sonication applied to a specific region of a brain of another subject.

13. The method of any one of claims 1-6, comprising calculating a location of the sonication in the subject based on a corresponding location in a standardized image of a brain.

14. The method of any one of claims 1-6, comprising obtaining parameters of sonication for a particular location of the specific region in the brain of the subject from the information associated with a location in the composite biomarker atlas.

15. A device comprising:memory- that stores a set of instructions; anda processor coupled to the memory and configured to execute the instructions to: obtain and process a response of a subject to focused ultrasound applied to a specific region within a brain of the subject, wherein the response includes or corresponds to a biomarker;associate information from the focused ultrasound to coordinates of the specific region within the brain of the subject; andbased thereon, build a composite biomarker atlas of the brain of the subject.1690.105.11116. The device of claim 15, wherein the processor is configured to execute the instructions to associate the information from the focused ultrasound to coordinates of the specific region within the brain includes associating parameters of sonication and / or the response of the subject to the coordinates.

17. The device of claim 15, wherein the processor is configured to execute the instructions to associate the information from the focused ultrasound to coordinates of the specific region within the brain includes associating the information from coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain.

18. The device of claim 15, wherein the processor is configured to execute the instructions to obtain the response of the subject to the focused ultrasound via communication with an ultrasound device and at least one of:a chemical biomarker analysis system configured to capture a chemical response; a physiological waveform processing system configured to capture a physiologic response;an imaging device configured to capture an image of the brain of the subj ect; and a computing device in communication with the imaging device and / or the ultrasound device.

19. The device of claim 15, wherein the processor is configured to execute the instructions to process and analyze the response to the focused ultrasound and, in response, associate the information including results of the analysis to the coordinates of the specific region.

20. The device of any one of claims 15-19, wherein the processor is configured to execute the instractions to obtain parameters of sonication from the composite biomarker atlas and which optionally are modified on a subject-to-subject basis.

21. The device of any one of claims 15-19, wherein the processor is configured to execute the instructions to build the composite biomarker atlas for the brain over multiple subjects,1690.105.111including the subject, and over a single sonication session or multiple sonication sessions for each of the multiple subjects.

22. The device of any one of claims 15-19, wherein the biomarker is a chemical biomarker and / or an electrical biomarker.

23. The device of any one of claims 15-19, wherein the response is multi-modal.

24. The device of any one of claims 15-19, wherein the processor is configured to execute the instructions to process and analyze the response by comparing the response to sonication applied to the specific region of the brain of the subject and a response to sonication applied to a specific region of a brain of another subject.

25. The device of any one of claims 15-19, wherein the processor is configured to execute the instructions to calculate a location of the sonication in the subject based on a corresponding location in a standardized image of a brain.

26. The device of any one of claims 15-19, wherein the processor is configured to execute the instructions to obtain parameters of sonication for a particular location of the specific region in the brain of the subject from the information associated with a location in the composite biomarker atlas.

27. A system comprising:an ultrasound device configured to applying focused ultrasound to a specific region within a brain of a subject, and optionally in response, open a blood-brain barrier of the subj ect;a processor configured to:obtain and process a response of the subject to the focused ultrasound applied to the specific region within the brain of the subject, wherein the response includes or corresponds to a biomarker;associate information from the focused ultrasound, to coordinates of the specific region within the brain of the subject; andbased thereon, build a composite biomarker atlas of the brain of the subject.1690.105.11128. The system of claim 27, wherein the processor is configured to associate the information from the focused ultrasound to coordinates of the specific region within the brain by associating parameters of sonication and / or the response of the subject to the coordinates.

29. The system of claim 27, wherein the processor is configured to associate the information from the focused ultrasound to coordinates of the specific region within the brain by associating the information from coordinates of the specific region within the brain of the subject to coordinates of a specific region in a standardized image of the brain.

30. The system of claim 27, wherein the processor is configured to obtain the response of the subject to focused ultrasound via communication with the ultrasound device and at least one of:a chemical biomarker analysis system configured to capture a chemical response; a physiological waveform processing system configured to capture a physiologic response;an imaging device configured to capture an image of the brain of the subject; and a computing device in communication with the imaging device.

31. The system of claim 27, wherein the ultrasound device is configured to apply the focused ultrasound non-invasively.

32. The system of claim 27, wherein the ultrasound device is configured to apply the focused ultrasound to the specific region within the brain via a sonication procedure which is uninterrupted in response to subject motion such that the specific region of the brain is sonicated independent of the subject motion.

33. The system of claim 27, wherein the processor is configured to process and analyze the response to the focused ultrasound and, in response, associate the information including results of the analysis to the coordinates of the specific region.

34. The system of any one of claims 27-33, wherein the processor is configured to obtain parameters of sonication from the composite biomarker atlas and which optionally are modified on a subject-to-subject basis.1690.105.11135. The system of any one of claims 27-33, wherein the processor is configured to build the composite biomarker atlas for the brain over multiple subjects, including the subject, and over a single sonication session or multiple sonication sessions for each of the multiple subjects.

36. The system of any one of claims 27-33, wherein the biomarker is a chemical biomarker and / or an electrical biomarker.

37. The system of any one of claims 27-33, wherein the response is multi-modal.

38. The system of any one of claims 27-33, wherein the processor is configured to process and analyze the response by comparing the response to sonication applied to the specific region of the brain of the subject and a response to sonication applied to a specific region of a brain of another subject.

39. The system of any one of claims 27-33, wherein the processor is configured to calculate a location of the sonication in the subject based on a corresponding location in a standardized image of a brain.

40. The system of any one of claims 27-33, wherein the processor is configured to obtain parameters of sonication for a particular location of the specific region in the brain of the subject from the information associated with a location in the composite biomarker atlas.

Images