Targeted neuromodulation to improve neuropsychiatric function

Abstract

The present invention provides systems and methods for targeted neuromodulation. The method includes receiving a first image showing brain structure from a first imaging system and a second image showing brain connectivity from one of the first and second imaging systems. For each voxel of a plurality of voxels within a region of interest, a first utility value associated with directly modulating tissue within the region of interest is identified from the first image. For each voxel of the plurality of voxels, a second utility value associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest is identified from the second image. Based on the first utility value and the second utility value, an overall utility value for each voxel of the plurality of voxels is identified, and an optimal location is identified based on the overall utility value.

Description

Detailed Description of the Invention

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This application claims priority to U.S. provisional application Ser. No. 63,400,960, entitled "Focused Neuromodulation for Improving Neuropsychiatric Function," filed Aug. 25, 2022, U.S. patent application Ser. No. 18 / 310,465, entitled "Screening, Monitoring, and Treatment Framework for Focused Ultrasound," filed May 1, 2023, and U.S. patent application Ser. No. 18 / 227,270, entitled "Analysis Framework for Assessing Human Health," filed July 27, 2023, the entire contents of which are incorporated herein by reference. [Technical Field]

[0002] The present invention relates to the field of medical systems, and more particularly to targeting neuromodulatory treatment of neuropsychiatric functions. [Background technology]

[0003] A bronchogram was performed using diffusion MRI data. Free water diffusion is called "isotropic" diffusion. When water diffuses in a medium with a barrier, the diffusion is non-uniform, which is called anisotropic diffusion. In this case, the relative mobility of the molecules from their origin has a shape different from a sphere. This shape is commonly modeled as an ellipsoid, and this technique is called diffusion tensor imaging. The barrier can be many things, such as cell membranes, axons, or myelin. However, in white matter, the main barrier is the myelin sheath of the axon. Axon bundles provide a barrier to perpendicular diffusion and a pathway for parallel diffusion along the fiber direction. Summary of the Invention

[0004] One aspect of the present invention provides a method for targeting neuromodulation in a patient's brain for use in one of improving, diagnosing, and managing neuropsychiatric function. A first image showing brain structure is acquired from a first imaging system, and a second image showing brain connectivity is acquired from one of the first imaging system and a second imaging system. For each voxel of a plurality of voxels within a region of interest, a first utility value associated with directly modulating tissue within the region of interest is identified from the first image. For each voxel of the plurality of voxels, a second utility value associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest is identified from the second image. An overall utility value for each voxel of the plurality of voxels is determined based on at least the first utility value and the second utility value. An optimal location for neuromodulation is identified based on the overall utility value for each voxel of the plurality of voxels.

[0005] Another aspect of the present invention provides a system. An imaging interface receives a first image showing brain structure from a first imaging system and a second image showing brain connectivity from one of the first imaging system and a second imaging system. A targeting component identifies, for each voxel of a plurality of voxels within a region of interest, a first utility value from the first image associated with directly modulating tissue within the region of interest, and for each voxel of the plurality of voxels, a second utility value from the second image associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest, determines an overall utility value for each voxel of the plurality of voxels based on at least the first utility value and the second utility value, and identifies an optimal location for neuromodulation based on the overall utility value for each voxel of the plurality of voxels. A neuromodulation system delivers neuromodulation to the optimal location.

[0006] Yet another aspect of the present invention provides a method for improving neuropsychiatric function in a patient by selecting an influence volume having a center point within a target region including the adenoseptal nucleus and the ventral internal capsule, the center point being located between 7 mm and 12 mm from the midline of the brain on both sides, between 1 mm and 6 mm anterior to the anterior cortex (AC), and between 2 mm superior and 2 mm inferior to the AC. Neuromodulation is delivered to the selected influence volume.

[0007] These and other features of the present invention will become apparent to those skilled in the art to which the invention pertains upon reading the following description in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0008] [Figure 1] 1 illustrates a system for targeting neuromodulation used to treat or diagnose neuropsychiatric functions in specific regions of a patient's brain. [Figure 2] 1 illustrates a system for assessing the effects of neuromodulation on one or more patients. [Figure 3] A system is presented that identifies risk or diagnoses neuropsychiatric function based on patient brain imaging, facilitating, for example, who to select for treatment, how often to treat, and other treatment decision-making uses. [Figure 4] 1 illustrates a first method for targeting neuromodulation in the brain of a patient for use in the treatment or diagnosis or management of neuropsychiatric functions. [Figure 5] A second method is presented for targeting neuromodulation in the brain of a patient for use in the treatment or diagnosis or management of neuropsychiatric functions. [Figure 6] A method for identifying risk of neuropsychiatric function based on brain imaging of a patient is presented. [Figure 7] FIG. 1 shows a schematic block diagram of an example system of example hardware components capable of implementing the systems and methods shown in FIGS. 1-6. [Figure 8] Coronal view of a brain MRI scan slice identifying the adenoseptal nucleus and internal capsule. [Figure 9]Figure 1 Axial view of a brain MRI scan slice identifying the anterior, posterior, and middle association points and midline of the brain. [Figure 10] FIG. 1 is a schematic diagram of exemplary treatment sites and volumes of the impact of neuromodulation on addiction. [Figure 11] FIG. 1 is a schematic coronal view of a brain MRI scan slice showing exemplary treatment sites. [Figure 12] 1 is a diagram of an MRI image of a brain with implanted DBS electrodes. [Figure 13] FIG. 13 shows a schematic block diagram of an example system of example hardware components capable of implementing the systems and methods shown in FIGS. 1-12. DETAILED DESCRIPTION OF THE INVENTION

[0009] Various examples of the systems and methods described herein use brain connections registered with anatomical imaging to identify optimal locations and geometries for neuromodulation application. Neuromodulation techniques typically involve transmitting energy to specific brain regions to alter neurons, axons, entire nerves, nervous systems, or neural activity. Targeting of neuromodulation has traditionally been based on anatomical structures. Unfortunately, this requires coarser modulation settings and timely instillation periods to achieve the expected effect. The present invention relates to identifying precise brain regions using information based on individual nervous system pathways and adjusting the corresponding pathways to optimize the desired effect. Furthermore, the system complements this data with comprehensive bioparameter-guided detection and prediction of functional status, symptoms, and disease states, networks, and biobehavioral, cognitive feedback, virtual reality presentation, and imaging, allowing functional status, symptoms, and disease states to be quantified and monitored through neuromodulation screening and other treatments. In one example, focused ultrasound microtherapy can be used to screen and evaluate patients to determine whether treatment is effective. If not, higher frequencies and doses can be used. This screening can be used to diagnose disease, identify appropriate treatments (eg, drugs, neuromodulatory, or other treatments), and determine the need, frequency, and dosage of neuromodulatory treatment.

[0010] As used herein, a "neuromodulation technique or mode" or "neuromodulation" refers to any suitable technique for applying local energy (or other modes of neuromodulation) to the brain to modulate (e.g., activate, inhibit, modulate, reset, normalize) neural activity and neural networks, or for altering blood-brain barrier permeability to apply therapy to a specific location. Non-relaxation neuromodulation techniques or modes include ultrasound, e.g., focused ultrasound; electrical stimulation, e.g., superficial (including transcutaneous, transdermal, or subcutaneous) or cortical or deep brain stimulation; magnetic stimulation, including transcranial magnetic stimulation; optogenetic stimulation; or pulsed electromagnetic radiation. Other modes of neuromodulation include the use of light, pressure, and heat / cold. As used herein, the term "modulation" refers to inhibiting, exciting, modulating, normalizing, resetting, or normalizing neural activity.

[0011] As used herein, a "predictive model" is a mathematical or machine learning model for predicting the future state of a parameter or estimating the current state of a parameter that cannot be directly measured.

[0012] As used herein, a "voxel" is an arbitrarily defined volume within a region within an image or model generated from an image that represents the smallest unit of analysis within the image or model.

[0013] "Neuropsychiatric functions" include, for example, behavioral manifestations of brain functions such as cognition, activation, motivation, emotion regulation, behavioral control, and perception and / or understanding of emotional stimulus features, including processing speed. Other examples of neuropsychiatric functions include general intellectual function, basic attention, complex attention (working memory), executive function, memory (visual and verbal), language, visuo-structural function, and visuo-spatial organization. Thus, non-limiting examples of neuropsychiatric functions include general intellectual function, basic attention, the ability to monitor and guide, and attentional functions such as the flexible allocation of attentional resources, working memory, and divided attention. Divided attention refers to a limited-capacity memory system that can temporarily store and manipulate information that is the direct focus of attention (e.g., the ability to simultaneously maintain two mindsets in a flexible manner). Executive functions include, for example, planning, problem-solving abilities, purposeful and self-directed action, organizational skills, goal-directed action, the ability to generate multiple responses, and concept collection (i.e., the ability to maintain (or not lose) one activity), the ability to evaluate and modify behavior based on feedback, verbal and visual memory, or the ability to retain and remember new information for future use, visuospatial skills such as judging the direction of lines and discerning spatial relationships and patterns, two-dimensional construction skills (such as drawing a picture or completing a puzzle) and three-dimensional construction skills (such as arranging blocks to fit a design). visual-constructive skills including anticipatory naming (e.g., naming a specific word on demand, e.g., when viewing a picture of an object), word fluency, or generating a non-redundant word list belonging to a specific category; motivation / drive / activation in human, cognitive, or behavioral domains; emotion regulation (e.g., the ability to regulate and direct emotions and feelings in an appropriate manner), and interpretation of emotional stimuli (e.g., the ability to interpret emotional facial expressions, posture, body language, rhythm, and contextual information to infer the emotional state of others or to help identify an appropriate emotional response). Neuropsychiatric dysfunction refers to abnormal neuropsychiatric functioning compared to normal, healthy people.

[0014] Impaired neuropsychiatric function may include, for example, abnormal anxiety, stress, fear, emotion, depression, obsessive-compulsive disorder, or compulsive behavior. Various patients exhibiting impaired neuropsychiatric function can be improved, including those with neuropsychiatric, psychiatric, neurodevelopmental, or neurodegenerative disorders. Non-limiting examples of such disorders include anxiety, addiction, emotional disorders, obsessive-compulsive disorder (OCD), depression, post-traumatic stress disorder, biphasic affective disorder, autism, autism-related disorders, reading comprehension disorders, attention deficit disorder, acquired brain injury, stroke, schizophrenia, other forms of dementia, and Parkinson's disease. Addiction includes addiction to addictive behaviors, addictive chemicals, or combinations thereof. Non-limiting examples of addictive behavior addictions include gambling addiction, food addiction (including obesity and / or eating disorders), sex, shopping, exercise, sports training, electronic gaming, media / internet use, pathological work, compulsive sexual criminal behavior, and combinations thereof. Non-limiting examples of addictions to addictive substances include addiction to nicotine, alcohol, marijuana, painkillers, opioid drugs, heroin, benzodiazepines, doping agents such as toluenepropylamine, dextrobenzenepropylamine, and benzenepropylamines including piperate methyl acetate, gasoline, household cleaning products, inhalants such as aerosols, sedative / hypnotic drugs such as barbiturates, zopiratan tartrate, and ezopirone, and combinations thereof. These situations are examples of situations where neuropsychological function may be impaired and therefore require improvement. While the present invention is primarily described in the context of addiction, these methods and systems can be used in other situations where patients exhibit impaired neuropsychiatric function, such as Alzheimer's disease and related dementias and cognitive disorders, including, but not limited to, mild cognitive impairment (MCI), Louisiana dementia, frontal lobe dementia, vascular dementia, Parkinson's dementia, chronic traumatic encephalopathy, Huntington's disease, and multiple system atrophy.

[0015] "Registration" of two or more images involves the process of assigning relative positions between pixel or multi-pixel features in two or more images. For example, this assignment can be represented by an explicit transformation model between the two or more images, or by feature matching techniques that identify common structural features in the two images.

[0016] As used herein, "intensity distribution" refers to the spatial variation of local energy intensity provided to the brain. The intensity distribution can include the variance between the two sides of the region providing the local energy or more complex spatial variations of energy.

[0017] FIG. 1 illustrates a system 100 for targeted neuromodulation to treat or diagnose impaired neuropsychiatric function in specific regions of a patient's brain. The system 100 includes a processor 102 storing executable instructions for targeted neuromodulation to treat or diagnose impaired neuropsychiatric function and a non-transitory computer-readable medium 110. The executable instructions include an imaging interface 112 that receives a first image showing the patient's brain structure and a second image showing brain connectivity from one or more associated imaging systems (not shown). In one embodiment, the first image is a T1 magnetic resonance imaging (MRI) image and the second image is a diffusion tensor imaging (DTI) image generated using the MRI imager. The imaging interface 112 can include appropriate software components for communicating with the imaging system (not shown) or a repository (not shown) that stores images over a network via a network interface (not shown) or bus.

[0018] In some implementations, the imaging interface 112 can segment the first image into multiple subregions of the brain. The identified subregions can include, for example, the forehead pole, the temporal pole, the superior forehead region, the medial marginal forehead region, the caudal frontal buckle, the rostral frontal buckle, the entorhinal region, the parahippocampal region, the pericacciliary region, the lingual region, the cuneus region, the isthmus region, the precuneus region, the paracentral lobule, and the fusiform region. For example, the first image can be placed in a standard gallery to provide the segmentation. In another embodiment, the imaging interface 112 can include a convolutional neural network that trains on multiple annotated image samples that can be used to provide the segmented image. An example of such a system can be found in Huo et al., "3D Whole Brain Segmentation Using Spatially Positioned Atlas Network Blocks," accessible via https: / / doi.org / 10.48550 / arxiv.1903.12152, the entire contents of which are incorporated herein by reference. The imaging interface 112 can also register the second image with the first image to know the locations in the brain that connect the nodes in the groups.

[0019] Each of the first and second images can be provided to a targeting component 116, which selects the location and intensity distribution of neuromodulation. The targeting component 116 generates brain connectivity from the second image, which indicates neural connections within the brain. For example, a region of interest can be defined within the first image based on segmentation of the first image, and the location and intensity distribution of neuromodulation within the region of interest can be selected based on the generated connectivity. It should be understood that the connectivity can be identified as passive connectivity, which represents physical connections between brain parts, or as passive connectivity, which represents activity induced by brain parts in response to energy provided by specific locations.

[0020] The targeting component 116 can divide the defined region of interest into multiple voxels and can assign a utility value to each voxel based on the voxel's location within the region of interest and its connections to other regions of the brain within the connectivity to form a utility map. The term "utility value" is used broadly herein to cover any method for assigning values ​​related to neuromodulation to tissue regions, and is expressly intended to cover cost methods that assign positive values ​​to regions where energy is not desired, and utility methods that assign positive values ​​to regions where energy is desired. In one example, the utility value for each voxel can be determined based on at least a first utility value determined from a first image and a second utility value determined from a second image.

[0021] For example, a first utility value can be determined for each tissue type represented by the voxel. In one embodiment, regions of interest from the first image can be arranged in a histology-based organized atlas containing values ​​for each region associated with the region of interest. In one example, values ​​in the histology-based atlas can be assigned based on the expected concentration of excitatory or inhibitory neurons in the population at each location, e.g., by analyzing histological samples representative of the region of interest in a population of individuals. In one example, information from the second image can be used to adjust the value of the first utility value. For example, depending on the specific disorder being treated, values ​​in hemispheres or regions with low connectivity can be assigned a higher value, indicating the potential to increase overall brain connectivity, or a lower value, representing a reduced secondary effect of neuromodulation. In the example of addiction, the region of interest can be all or part of the afferent nucleus, and the ventral internal capsule and neurons of interest can include, for example, GABA-energy neurons, such as mesospinal projection neurons (MNS). For brain injury and stroke, these values ​​can be determined based on proximity or connectivity to damaged brain tissue. For neurodevelopmental disorders, greater value may be given to tissues with high concentrations of excitatory or inhibitory neurons and networks within the septal nucleus to reduce anxiety and craving. For epilepsy and pain, regions of interest may be located within the epileptic focus, surrounding cortical and subcortical brain tissue, transmitting and propagating fibers in the damaged area, and the thalamus, hippocampus, temporal lobe, insula, and corpora thoria.

[0022] In the case of stroke, the region of interest can be located within the stroke area, surrounding cortical and subcortical brain tissue, input and output fibers of the damaged area, and one or more of the thalamus, aphenoseptal nucleus, motor cortex, sensory cortex, visual cortex, or language cortex. In the case of stroke, the region of interest can be located, for example, within the brain damage area, surrounding cortical and subcortical brain tissue, input and output fibers of the damaged area, and one or more of the thalamus, aphenoseptal nucleus, motor cortex, sensory cortex, visual cortex, or language cortex. In the case of Parkinson's disease, the region of interest can be located, for example, in one or more of the globus pallidus, thalamic neurons, thalamus, pulvinar, thalamic input and output. In the case of neurodevelopmental disorders, the region of interest can be located, for example, in one or more of the thalamus and associated nuclei, sensory cortex, visual cortex, frontal lobe, and input and output neurons associated with the sensory, visual, and frontal lobe. For neurodevelopmental disorders, the area of ​​interest may be located in one or more of the following: the hippocampus, basal ganglia, septum nucleus and internal capsule, sensory cortex, superior lobe, frontal lobe, calcitonin cortex, and temporal lobe. For each of these disorders, a high concentration of excitatory or inhibitory neurons and networks may provide greater value, depending on the specific disorder. Similarly, if a patient exhibits other impaired neuropsychiatric functions, appropriate areas of interest may be identified and treated as specified herein.

[0023] A second utility value can be determined from the second image and represents the indirect modulation of neurons connected to a given voxel. While brain regions, such as the frontal lobe, typically contribute to indirect modulation, modulation of other regions, such as the amygdala, can lead to undesirable side effects, such as anxiety. Using connectivity data from the second image, a set of locations connected to each voxel and the connection strength for each set of locations can be determined. A second utility value for each voxel can be determined as a weighted linear combination of these location values, with the weights derived from the connection strengths of each location.

[0024] A neuromodulation-related volume of influence has an associated intensity distribution around a reference point (e.g., a center point) that represents the amount of energy delivered to that region for a given location of the reference point. In one example, each voxel can be assigned a value normalized by the maximum intensity, which can be used to weight the voxel's contribution to the total cost associated with the location and intensity distribution. In other embodiments, the intensity throughout the volume of influence can be assumed to be substantially uniform, thus eliminating the need for weighting. In addition to the utility value assigned to each voxel, each voxel can also have one or more values ​​that indicate the application of a utility affecting a volume centered at that voxel, e.g., as a sum or weighted sum affecting all voxels within the volume. It should be understood that the volume itself can be affected by adding additional foci to focused ultrasound, selectively activating electrodes, or altering the field direction and strength generated by electrodes in deep brain stimulation to adjust the intensity distribution and shape affecting the volume. An optimization process, such as gradient descent, can be used to explore the region of interest to obtain an optimal or near-optimal center point location and shape affecting the volume, and the resulting values ​​can be provided to the treatment planning system 118 to generate a treatment plan for the patient. In one example, the treatment planning system 118 can be limited to select an influence volume having a center point located between 7 millimeters (mm) and 12 mm from the midline on both the left and right sides of the brain, between 1 mm and 6 mm anterior to the anterior cortex (AC), and between 2 mm above and 2 mm below the AC.

[0025] In one example, the targeting component 116 can operate in conjunction with the neuromodulation system 120 to refine the atlas of brain connectivity. Specifically, energy can be applied to various locations within the region of interest, activity within the brain can be identified via appropriate functional imaging modes, and whether the expected changes in brain activity from the second image after neuromodulation are realized can be determined. Typically, these images can be acquired while the patient is performing a task or experiencing cues related to the patient's illness. Additionally or alternatively, changes in brain connectivity after treatment can be used to measure brain connectivity and optimize parameters within the targeting component. In one embodiment, as a result of neuromodulation at a given location, the activity measured in response to neuromodulation at that location within the brain is recorded. This can be used to modify the weights for each connection set applied to each voxel, or to modify values ​​applied to the patient voxel itself. Additionally, the targeting component can generally be updated based on feedback to modify parameters related to other patient targets, such as region of interest values ​​stored in the histology-based atlas or assigning connection strengths using connectivity data in the second image. When providing energy, multiple locations may be affected, and detected activity can be attributed to each location, e.g., represented as voxels, based on the proportion of energy each voxel received. Alternatively, multiple measurements can be compared, e.g., by solving one or more n-dimensional linear systems, where n is the number of voxels in the region of interest. Thus, the active connectivity associated with each location within the region of interest can be determined, and the values ​​and connection weights associated with each of the multiple voxels can be adjusted.

[0026] FIG. 2 illustrates a system 1300 for assessing the effects of neuromodulation on one or more patients. The system 1300 includes a processor 202 storing executable instructions for targeting neuromodulation to treat and diagnose impaired neuropsychiatric function and a non-transitory computer-readable medium 210. The executable instructions include an imaging interface 212 that receives a first image showing the patient's brain structure and a second image showing brain connectivity. In one embodiment, the first image is a T1 magnetic resonance imaging (MRI) image and the second image is a diffusion tensor imaging (DTI) image generated using an MRI imager. The imaging interface 212 can also receive functional images from the same or a different MRI imager, for example, representing activity within the brain. The imaging interface 212 can include appropriate software components for communicating with an imaging system (not shown) or a repository for storing images (not shown) over a network via a network interface (not shown) or bus.

[0027] The first image is provided to a registration component 214, which segments the first image into multiple brain subregions. The identified subregions may include, for example, the forehead pole, the temporal pole, the superior forehead region, the medial border forehead region, the caudal frontal buckle, the rostral frontal buckle, the entorhinal region, the parahippocampal region, the pericacciliary region, the lingual region, the cuneus region, the isthmus region, the precuneus region, the paracentral lobule, and the fusiform region. In one example, the registration component 214 registers the first image to a standard gallery to provide the segmentation. In another embodiment, a convolutional neural network trained on multiple annotated image samples can be used to provide the segmented image. The registration component 214 can also register a second image with the first image to determine the locations of connecting nodes in groups within the brain. In addition to performing registration using the first image, various biological feature parameters can be extracted from the image, such as parameters related to connectivity, networks, gray-to-white ratio, and volume.

[0028] Each of the segmented first and second images can be provided to a feedback component 216 that identifies the effectiveness of treatment for the patient based on patient data collected in response to the stimuli. The stimuli can be applied to neuromodulation in a given area or suggestions related to the patient's disorder can be provided to the patient. An example of suggestion suggestions can be found in U.S. Patent Publication No. US2021 / 0162217, filed December 2, 2020, entitled "Method and System for Hint-Based Responsiveness Improvement and Addiction Monitoring," the entire contents of which are incorporated herein by reference. During the process of presenting a given stimulus, brain electrical activity or other physiological activity can be recorded to identify whether neural activity increases, decreases, or otherwise changes in response to the stimulus. The collected feedback can also include patient self-reports, clinician observations, measured electrical activity, measured biometric parameters such as heart rate variability and blood pressure, and other relevant parameters.

[0029] In one embodiment, the identified activity can be used as feedback to determine the success of treating a patient with neuromodulation. This can occur during or shortly after treatment ("acute feedback"), shortly (e.g., hours to days) after treatment ("semi-acute feedback"), or longer (e.g., weeks to months) after treatment ("chronic feedback"). Cues can be selected to measure the impact of treatment on the patient, and can include addiction-related cues (e.g., images, scents, tastes, sounds associated with addictive activities), activities that measure the patient's performance (e.g., memory and attention tasks), or other cues suitable for assessing the impact of the illness on the patient.

[0030] In some embodiments, a patient's baseline craving level for the addictive behavior or addictive chemical can be measured. These baseline levels can be taken before or at different times during treatment, and can be taken by the patient under the Addiction Care Standards outlined by the American Society of Addiction Medicine (ASAM 2013). These baseline levels can be taken at the time of ingestion, during medication-assisted and / or behavioral treatment. This baseline craving level can be measured in a clinical / laboratory setting. Furthermore, during treatment, these baseline levels may change over time. For example, the patient can be asked to rate their craving for the treatment-seeking substance or behavior using a 100-point visual analog scale (VAS), with 100 representing maximum craving and 0 representing no craving. After the baseline craving assessment, one method can include exposing the patient to cues related to the addictive behavior or addictive chemical and then assessing the craving level produced by the patient during or after cue exposure to identify changes in craving. After the presentation period or exposure to the presentation period, the patient's final or subsequent craving level is measured proximate in time, and the patient's final craving level is correlated with the patient's response to the presentation. For example, the resulting craving level can be measured during exposure to the cue, within 5 minutes of exposure to the cue, within 10 minutes of suggestion, or any measurable period in between. More specifically, one method can include obtaining a timely measurement of the patient's craving level for the addictive behavior or addictive chemical during or after exposure to the cue, and determining whether the patient's craving level is increased, or whether the patient's craving level is substantially the same or decreased compared to the baseline craving level. The method can then include providing or adjusting neuromodulation to improve the patient's addiction based on a comparison of the baseline craving level and the resulting craving level. For example, neuromodulation can be provided or adjusted upon determining that the generated craving level is higher than the patient's baseline craving level.

[0031] In some embodiments, rather than measuring the patient's craving level, physiological, cognitive, psychosocial, or behavioral parameters related to the patient's addictive behavior or the patient's addictive chemical are measured. Specifically, the method can include obtaining a baseline measurement of the patient's physiological, cognitive, psychosocial, or behavioral parameter. As described above, these baseline levels change over time during treatment. After this initial assessment, one method can include exposing the patient to cues related to the addictive behavior or the addictive chemical and then evaluating the patient's final parameter values ​​to identify changes in the parameter values ​​during or after cue exposure. After the presentation period or exposure to the presentation period, the patient's outcome or subsequent parameter values ​​are measured proximate in time so that the patient's outcome parameter measurements are related to the patient's response to the presentation. More specifically, the method can include obtaining a temporally proximate measurement of the parameter's outcome value during or after exposure to the cue to identify whether the parameter's outcome value is increased, or whether the outcome parameter value is substantially the same or decreased compared to the baseline parameter value. The method can then provide or adjust neuromodulation to the patient based on a comparison of the obtained parameter value with the baseline parameter value. For example, neuromodulation may be provided or adjusted based on identifying that the obtained parameter value is elevated above the patient's baseline parameter value.

[0032] The cues the patient encounters may be visual, taste, auditory, tactile, olfactory, or a combination thereof. For example, if the patient is in a natural non-clinical environment, such as at home, work, or other non-clinical environment, the patient may be exposed to these cues via a smartphone, tablet, personal computer, or laptop. The patient may also be exposed to cues through virtual reality, augmented reality, or mixed reality. During any assessment period, the patient may be exposed to a variety of cues, and if the patient uses a variety of substances or behaviors, the patient may be exposed to cues related to different addictive substances or behaviors. If the patient is addicted to chemicals, the cues may be, for example, images of drugs, drug paraphernalia, or drug users. This cue may be specific to the specific addictive behavior or addictive chemical for which the patient is seeking treatment and may include multiple cues, including multiple different types of cues. For example, if the patient is addicted to alcohol, the cues may be the smell of alcohol, a visual image of a bar, or the sound of an alcoholic beverage container being opened. For example, if the patient is addicted to heroin, the cues may be visual images of heroin, a hypodermic needle, a spoon, or a lighter. For example, if a patient is addicted to gambling, the cue could be a visual image of a gambling hall or gambling chips. The above examples are merely illustrative and are intended to demonstrate that cues can be addiction-specific and stimulate different senses. These cues could also resemble characteristics of the patient, such as age, gender, race, preferred chemical, or route of administration. In other words, the cues that a patient encounters can be personalized to the specific patient seeking treatment.

[0033] In aspects where a patient's physiological parameters are measured, the physiological parameters are responses of the patient's autonomic nervous system to the presented exposure, and multiple physiological parameters can be measured during any given assessment session. The physiological parameters can be measured via a wearable device, such as a ring, watch, or belt, or via a smartphone or tablet, and can be measured in a natural, non-clinical environment, such as when the patient is at home, at work, or other non-clinical environment. Exemplary physiological parameters include heart rate, heart rate variability, sweating, salivation, blood pressure, pupil size, brain activity, electrodermal activity, body temperature, and blood oxygen saturation. Table I provides non-limiting examples of measurable physiological parameters and exemplary tests for measuring the physiological parameters.

[0034] (Table I) TIFF2025528428000002.tif174156

[0035] Physiological parameters can be measured using appropriate devices in clinical settings and wearable, implantable, or portable devices in non-clinical settings. Some information can also be determined from self-reports made by users through applications in or through interactions with applications on mobile devices. For example, a smartwatch, ring, or patch can be used to measure a user's heart rate, heart rate variability, body temperature, blood oxygen saturation, movement, and sleep. In non-clinical settings, these values ​​can also be analyzed intraday to estimate variability. For example, eye tracking can be performed using a camera and dedicated software on the mobile device.

[0036] Table II provides non-limiting examples of cognitive parameters that can be gamified and measured, as well as exemplary methods and tests / tasks for measuring such cognitive parameters. Cognitive parameters can be assessed through a series of cognitive tests that measure, for example, executive function, decision making, working memory, attention, and fatigue.

[0037] (Table II) TIFF2025528428000003.tif195156

[0038] These cognitive tests can be administered in a clinical / laboratory setting or in a natural or non-clinical setting (e.g., when the user is at home, at work, or in other non-clinical settings). Smart devices such as smartphones, tablet computers, or smartwatches can facilitate measuring these cognitive parameters in natural or non-clinical settings. For example, the Eriksen Flanker, N-Back, and psychomotor vigilance tasks can be administered by applications on a smartphone, tablet computer, or smartwatch. In one example, a patient can explore a virtual reality environment and collect objects in the environment. The patient is then asked to recall the location of each item in the virtual environment and its relationship to a point of origin, testing the patient's ability to recall the spatial relationships between the virtual locations.

[0039] Table III provides non-limiting examples and exemplary tests, devices, and methods of parameters related to a user's movement and activity; for ease of reference, these parameters are alternatively referred to herein as measurable "motion parameters." The use of portable monitoring, physiological sensing, and portable computing devices allows for the measurement of motion parameters. The use of embedded accelerometers, GPS, and cameras allows for the capture and quantification of a user's movements, thereby ascertaining how their health is affected and the associated health-related parameters. Range of motion and gait analysis can be assessed in a clinical setting using appropriate motion capture and camera devices.

[0040] (Table III) TIFF2025528428000004.tif74157

[0041] Table IV provides non-limiting examples and exemplary tests, devices, and methods of parameters related to a user's sensory sensitivity, and for ease of reference, these parameters are alternatively referred to herein as measurable "sensory parameters."

[0042] (Table IV) TIFF2025528428000005.tif46157

[0043] Table V provides non-limiting examples and exemplary tests, devices, and methods of parameters related to a user's sleep quantity, progress, and sleep quality; for ease of reference, these parameters are alternatively referred to herein as measurable "sleep parameters."

[0044] (Table V) TIFF2025528428000006.tif73156

[0045] Table VI provides non-limiting examples of parameters extracted by locating biomarkers associated with a user, and exemplary tests, devices, and methods; for ease of reference, these parameters are alternatively referred to herein as measurable "biomarker parameters." Biomarkers may further include imaging and physiological biomarkers related to chronic health conditions and improvement or worsening of chronic health conditions.

[0046] (Table VI) TIFF2025528428000007.tif59156

[0047] Table VII provides non-limiting examples and exemplary tests, devices, and methods of socio-psychological and behavioral parameters, which for ease of reference are referred to herein as alternatively measurable "social-psychological parameters."

[0048] (Table VII) TIFF2025528428000008.tif188156

[0049] In addition to one or a combination of physiological, cognitive, psychosocial, and behavioral parameters, clinical data may also be part of a multidimensional feedback method for assessing the effectiveness of treatment. Such clinical data may include, for example, the user's clinical condition, the user's medical history (including family history), employment information, and housing status. In particular, functional imaging can be used to assess the effects of neuromodulation on brain activity. For example, decreased activity in inhibitory neuron target regions or increased activity in excitatory neuron target regions can indicate successful neuromodulation. Similarly, connectivity information identified by fibril beam imaging can be used to indicate temporal changes in brain connectivity with treatment. In particular, the mass, volume, and density of connections shown by fiber bundle imaging, as well as changes in these values ​​over time, can indicate the effectiveness of treatment. This can then be used to guide changes in treatment frequency and parameters and the selection of treated patients. This is particularly interesting in addiction, where increased connectivity from the septum nucleus to the frontal lobe can indicate successful treatment. In neurodevelopmental disorders, increased overall brain connectivity can indicate response to treatment. Increased connectivity is also an indicator of increased neuroplasticity, which indicates improvement in traumatic brain injury and stroke.

[0050] The type of change in the patient's physiological parameter measurements or craving level during or after cue exposure can influence whether to provide neuromodulation or adjust existing neuromodulation. For example, to provide neuromodulation, a method can include initiating neuromodulation if the patient's physiological parameter measurements or craving level increase during or after cue exposure compared to baseline physiological parameter measurements or baseline craving levels. In contrast, if the physiological parameter measurements or craving level during or after cue exposure are essentially the same as the baseline physiological parameter measurements or baseline craving levels, neuromodulation may not be applied. To adjust treatment against the backdrop of neuromodulation, a method can include adjusting neuromodulation parameters or doses, such as the duration, frequency, or intensity of neuromodulation. If the patient's physiological parameter measurements or craving level increase during or after cue exposure compared to baseline physiological parameter measurements or baseline craving levels, one method can include adjusting neuromodulation to make neuromodulation more effective. For example, if the patient was performing FUS for 5 minutes during the treatment period, the patient could perform FUS for 20 minutes each course, or if the patient was performing FUS every 30 days, the patient could perform it every two weeks. In contrast, if the physiological parameter measurements or craving levels after cue exposure are essentially the same as the baseline physiological parameter measurements or baseline craving levels, then the neuromodulation parameters may not need to be adjusted, and subsequent neuromodulation sessions can be used primarily as maintenance sessions, or the intensity, frequency, or duration of neuromodulation can be reduced, for example. Alternatively, if the physiological parameter measurements or craving levels during or after cue exposure are approximately the same as the baseline physiological parameter measurements or baseline craving levels, the patient can stop receiving any subsequent neuromodulation. The above scenarios are merely exemplary and are intended to illustrate whether the presence and type of changes in the patient's physiological parameter measurements and craving levels during and after cue exposure may suggest whether treatment should be provided, or whether existing treatment should be adjusted or terminated.

[0051] Furthermore, the degree of a patient's physiological, cognitive, psychological, or behavioral parameter measurements during or after cue exposure, and the degree of the patient's craving level during or after cue exposure, can affect the parameters of initial or subsequent neuromodulation. For example, if a particular patient seeking treatment has a higher craving level during or after cue exposure than the average craving level of the same patient population (patients with the same addiction), treatment may be more aggressive at the beginning or thereafter (e.g., in the case of neuromodulation, the duration, frequency, or intensity of neuromodulation may be greater than the duration, frequency, or intensity of patients provided to the same patient population). Similarly, if a particular patient seeking treatment has a higher physiological, cognitive, psychosocial, or behavioral parameter measurement during or after cue exposure than the average parameter measurement of the same patient population, treatment may be more aggressive at the beginning or thereafter. In contrast, if a particular patient's craving level or parameter measurement during or after cue exposure is lower than the average craving level or parameter value of the same patient population, treatment may be less aggressive at the beginning or thereafter. In other words, the severity or degree (and baseline values ​​and levels) of a patient's final craving level or final physiological, cognitive, psychosocial, or behavioral parameter measurements during or after cue exposure can be related to the degree of neuromodulation or aggression. The above scenarios are exemplary and are intended to illustrate that the degree of change in a patient's physiological parameter measurements and craving levels during and after cue exposure impacts initial and subsequent treatment parameters.

[0052] In some embodiments, the feedback is not a response to a presentation, but rather a comparison of one or more combinations of the patient's physiological, cognitive, psychosocial, and behavioral parameters. For example, a baseline measurement of one or more combinations of the patient's physiological, cognitive, psychosocial, and behavioral parameters can be obtained. The patient can then be exposed to neuromodulation, such as an initial focused ultrasound signal, an initial deep brain stimulation signal, or an initial transcranial magnetic stimulation signal, to the patient's neural target site. During or after application of the initial focused ultrasound signal, the initial deep brain stimulation signal, or the initial transcranial magnetic stimulation signal, a subsequent measurement of a resultant value of one or more combinations of the patient's physiological, cognitive, psychosocial, and behavioral parameters can be obtained. The resultant value can be compared to the baseline value to determine whether the patient's addiction has improved. The neuromodulation can be adjusted after determining that the patient's addiction has not improved.

[0053] As discussed above, if it is determined that neuromodulation is not successful, neuromodulation can be provided to a different target location. It should be understood that this system can operate in combination with the system of FIG. 1 to direct the selection of a new target location for neuromodulation. In particular, a new target for neuromodulation can be selected, for example, via targeting component 114 shown in FIG. 1.

[0054] Stimulus presentations can be repeated to accumulate functional imaging information associated with each of the multiple stimuli, which can be accumulated and stored in memory. If neuromodulation is applied to the stimuli, the functional imaging information can be used to record the location of the applied neuromodulation. The accumulated imaging information and connectivity groups can also be provided to an expert system 218, which identifies brain regions and nodes that respond to the stimuli and correlations between region and node activity, referred to herein as co-activation locations. The identified co-activation locations can be informed by the connectivity groups, such that only connected regions and nodes within the connectivity group are identified as co-activated. It is understood that functional imaging information can be accumulated for a single patient, a set of patients with a specific type of neuropsychiatric function, a set of patients with a general class of neuropsychiatric function, patients with no neuropsychiatric function, or two sets of patients with neuropsychiatric function. Based on the patient or set of patients used and the stimuli used, the results can represent the patient's unique brain circuitry, biomarkers of general neuropsychiatric function or a specific type of neuropsychiatric function, or biomarkers associated with response to neuromodulation or other treatment. For example, the identified patterns can be used to guide treatment of a particular patient or for biomarkers to identify neuropsychiatric function in new patients.

[0055] In one example, information collected on a set of patients with general neuropsychiatric function or a specific disorder can be used to direct the selection of an initial location for applying neuromodulation. For example, if a particular location defined for one or more marker structures in the brain has consistently been successful in a set of patients with the same or similar disorder, this location may be the default location for initial neuromodulation treatment. As described above, feedback describing the patient's response to treatment can be collected, and a new location for the patient can be selected if necessary. The new location can be identified based on connection information and information collected for the set of patients. In one embodiment, a similarity inference system can be used to identify past patients with characteristics similar to the current patient, and the successful location for similar patients can be used. Characteristics for matching patients with similar patients include demographic characteristics, medical history, measured biometric parameters such as blood pressure and heart rate variability, observed patient responses to presentations, and brain electrical activity measured during and after neuromodulation.

[0056] In another example, measurement feedback from the patient, such as measured electrical activity or biometric parameters, can be used to automatically detect cues presented to the patient. In this embodiment, neuromodulation and measurements of electrical activity and biometric parameters can be provided by one or two of an implanted device and a wearable device, and periodic measurements of electrical activity can be used to assess the patient. These measurements can be provided to an expert system 218 that can be trained on data representing the patient's responses to presented cues to determine whether the patient has encountered a cue related to the patient's neuropsychiatric function. If it is determined that the patient is responding to a prompt in the environment, a device providing neuromodulation can be activated to provide immediate treatment to the patient, thereby automatically presenting neuromodulation in response to the environmental prompt.

[0057] FIG. 3 illustrates a system 300 for identifying a patient's risk of impairment of neuropsychiatric function based on brain imaging. The system 300 includes a processor 302 storing executable instructions for identifying a patient's risk of neuropsychiatric function based on brain imaging and a non-transitory computer-readable medium 310. The executable instructions include an imaging interface 312 that receives a first image representing the patient's brain structure and a second image representing brain connectivity. In one embodiment, the first image is a T1 magnetic resonance imaging (MRI) image and the second image is a diffusion tensor imaging (DTI) image generated using an MRI imager. The imaging interface 312 can also receive functional images from the same or a different MRI imager, for example, representing activity within the brain. The imaging interface 312 can include appropriate software components for communicating with an imaging system (not shown) or a repository for storing images (not shown) over a network via a network interface (not shown) or bus.

[0058] The first image is provided to a registration component 314, which segments the first image into multiple brain subregions. The identified subregions may include, for example, the forehead pole, the temporal pole, the superior forehead region, the medial limbus forehead region, the caudal frontal buckle, the rostral frontal buckle, the entorhinal region, the parahippocampal region, the pericarcicular region, the lingual region, the cuneus region, the isthmus region, the precuneus region, the paracentral lobule, and the fusiform region. In one example, the registration component 314 registers the first image with a standard gallery to provide the segmentation. In another embodiment, a convolutional neural network trained on multiple annotated image samples can be used to provide the segmented image. The registration component 314 registers the second image with the first image to determine the locations of connecting nodes in groups within the brain.

[0059] The registered second image is provided to the machine learning model 316. The machine learning model may utilize one or more pattern recognition algorithms to analyze the concatenated group of images provided as a classification and regression model indicating, for example, the likelihood that a patient will have a general neuropsychiatric function class problem within a specified time period, the likelihood that a patient will suffer from or have a specific neuropsychiatric function problem within a specific time period, the likelihood that a patient will respond to a general neuropsychiatric function treatment, or the likelihood that a patient will respond to a specific neuropsychiatric function treatment, and assign a clinical parameter to the user. It should be understood that the clinical parameter may be categorical or continuous. When multiple classification and regression models are used, the machine learning model may include an arbitration element to provide consistent results from the various algorithms. Depending on the output of the various models, the arbitration element may simply select one class from the model with the highest confidence, select multiple classes from all models that meet a threshold confidence, select one category through a voting process between the models, or assign a single numerical parameter based on the output of multiple models. Alternatively, the arbitration element itself may be implemented as a classification model that receives the output of other models as features and generates one or more output classes for the patient.

[0060] Machine learning models and any constituent models can be trained on training data representing various categories of interest. Training data can include, for example, registered connected body images or digits and / or categorical features extracted from the registered connected body images. For example, in a supervised learning model, a system can be trained using a set of examples with labels representing the desired output of the machine learning model. The training process for machine learning models varies depending on the implementation, but training typically involves statistically aggregating training data into one or more parameters related to the output class. For rule-based models such as decision trees, domain knowledge provided by one or more human experts can be used to substitute or supplement the training data to select rules for classifying users using the extracted features. Any of a variety of techniques can be used for models, including support vector machines, regression models, self-organizing mapping, k-nearest neighbor classification or regression, fuzzy logic systems, data fusion processes, enrichment and packaging methods, rule-based systems, or artificial neural networks.

[0061] For example, an SVM classifier can utilize multiple functions (called hyperplanes) to conceptually divide boundaries in an N-dimensional feature space, where each of the N dimensions represents one relevant feature of the feature vector. The boundaries define the range of features associated with each class. Thus, the output class and associated confidence value for a given input feature vector can be identified based on the location of the given input feature vector relative to the boundaries in the feature space. In one embodiment, an SVM can be implemented using a kernel method using a linear or nonlinear kernel.

[0062] An ANN classifier includes multiple nodes with multiple interconnects. Values ​​from a feature vector are fed to multiple input nodes. Each input node provides these input values ​​to one or more layers of intermediate nodes. A given intermediate node receives one or more output values ​​from the previous node. The received values ​​are weighted according to a set of weights established during classifier training. The intermediate nodes convert the received values ​​into a single output based on a transfer function at the node. For example, an intermediate node may sum the received values ​​and subject the sum to a binary step function. Nodes in the final layer provide confidence values ​​for the ANN's output classes, where each node has an associated value indicating the confidence of one of the classifier's associated output classes. Another example is using an autoencoder to detect outliers in health-related parameters as an anomaly detector, thereby identifying when various parameters are outside an individual's normal range.

[0063] Many ANN classifiers are fully connected and feedforward. However, convolutional neural networks include convolutional layers, where nodes from previous layers are connected to only a subset of nodes in the convolutional layer. A recurrent neural network is a neural network where the connections between nodes form a directed graph along a timeline. Unlike feedforward networks, recurrent neural networks can combine feedback from states with earlier inputs, such that the output of a recurrent neural network for a given input can be a function of not only the input but also one or more previous inputs. As an example, a long short-term memory (LSTM) network is a modified version of a recurrent neural network that is better able to remember past data in memory.

[0064] A rule-based classifier applies a set of logical rules to extracted features to select an output class. Typically, rules are applied sequentially, with the logical outcome of each step influencing the analysis of subsequent steps. Specific rules and their order can be identified from any or all of the training data, analogical reasoning from previous situations, or prior domain knowledge. One example of a rule-based classifier is a decision tree algorithm, in which feature values ​​in a feature set are compared to corresponding thresholds in a hierarchical tree structure to select a class for a feature vector. A random forest classifier is a modification of the decision tree algorithm using a bootstrap aggregation or "bagging" method. In this method, multiple decision trees are trained on random samples of the training set and average (e.g., mean, median, or mode) results across the multiple trees are returned. For classification tasks, the results from each tree are classified, allowing for the use of mode results. Clinical parameters generated by the machine learning model 316, regardless of the particular model, may be provided to the user via a user interface on the display 320 or stored in a non-transitory computer-readable medium 310, such as an electronic medical record associated with the patient.

[0065] In view of the above-described structural and functional features, the example method may be better understood with reference to Figures 4-9. For ease of explanation, the example method of Figures 4-9 is illustrated and described as being executed serially; however, it should be understood and appreciated that the example is not limited by the illustrated order, as in other examples, some operations may occur in a different order, multiple times, and / or simultaneously with the order illustrated herein. Furthermore, not all of the illustrated operations need to be performed to implement a method in accordance with the present invention.

[0066] 4 illustrates a first method 400 for targeting neuromodulation in a patient's brain for use in treating, diagnosing, or managing neuropsychiatric functions. At 402, a first image showing brain structure is acquired from a first imaging system. At 404, a second image showing brain connectivity is acquired from one of the first and second imaging systems. In one embodiment, the second image is one of a set of images representing brain connectivity acquired by applying neuromodulation to a location within a region of interest associated with the image and measuring the activity level of at least one subregion of a plurality of brain subregions in response to the applied neuromodulation. Alternatively, the second image can be acquired by diffusion tensor imaging.

[0067] At 406, a segmented first image is generated by segmenting the first image into a plurality of brain subregions, such that at least a subset of a plurality of voxels comprising the first image is each associated with one of the plurality of subregions. It should be understood that for a given neuropsychiatric function, not all portions of the first image may be of interest, and therefore only voxels representing the plurality of subregions may be included in the segmentation. At 408, locations within the brain region of interest are selected as targets for neuromodulation based on the segmented first and second images.

[0068] 5 illustrates a second method 500 for targeting neuromodulation in a patient's brain for use in treating, diagnosing, or managing neuropsychiatric functions. At 502, a first image showing brain structure is acquired from a first imaging system. At 504, a second image showing brain connectivity is acquired from one of the first and second imaging systems. In one embodiment, the second image is one of a set of images representing brain connectivity acquired by applying neuromodulation to a location within a region of interest associated with the image and measuring the activity level of at least one subregion of a plurality of brain subregions in response to the applied neuromodulation. Alternatively, the second image can be acquired by diffusion tensor imaging.

[0069] At 506, a segmented first image is generated by segmenting the first image into a plurality of brain subregions, such that at least a subset of a plurality of voxels comprising the first image is each associated with one of the plurality of subregions. It should be understood that for a given neuropsychiatric function, not all portions of the first image may be of interest, and therefore only voxels representing the plurality of subregions may be included in the segmentation. At 508, locations within the brain region of interest are selected as targets for neuromodulation based on the segmented first image and the second image.

[0070] At 510, neuropsychiatric function cues are presented to the patient, and at 512, the patient's feedback in response to the cues is measured. Feedback may include a clinician's observation of the patient's appearance and behavior, the patient's self-report of disease symptoms, and measured brain electrical activity and biological characteristic parameters. At 514, the effectiveness of neuromodulation is determined based on the measured feedback. At 516, it is determined whether the effectiveness of neuromodulation meets a threshold. If yes (Y), then at 518, identifying a neuromodulation location is valid, and the method ends. If not, then the selected location is deemed invalid, and the method returns to 508 to select a new location within the region of interest as a target for neuromodulation. It should be understood that steps 510, 512, 514, and 516 can be performed during or immediately after treatment, shortly after treatment (e.g., hours to days), or longer after treatment (e.g., weeks to months).

[0071] 6 illustrates a method 600 for identifying risk of neuropsychiatric function based on imaging of a patient's brain. At 602, a first image showing brain structure is acquired from a first imaging system. At 604, a second image showing brain connectivity is acquired from one of the first and second imaging systems. In one embodiment, the second image is one of a set of images representing brain connectivity acquired by applying neuromodulation to a location within a region of interest associated with the image and measuring the activity level of at least one subregion of a plurality of brain subregions in response to the applied neuromodulation. Alternatively, the second image can be acquired by diffusion tensor imaging.

[0072] At 606, a segmented first image is generated by segmenting the first image into a plurality of brain subregions, such that at least a subset of a plurality of voxels comprising the first image is associated with one of the plurality of subregions. It should be understood that, since not all portions of the first image are of interest for a given neuropsychiatric function, only voxels representing the plurality of subregions may be included in the segmentation. At 608, the segmented first and second images are represented as a machine learning model trained on imaging data of a plurality of patients with known outcomes. At 610, a clinical parameter indicative of a patient's risk of neuropsychiatric dysfunction is generated from the representation of the segmented first and second images.

[0073] In one embodiment, a set of numerical features is extracted from the segmented first and second images, and the set of numerical features is provided to the machine learning model. In another embodiment, the segmented first and second images are directly fed to the machine learning model. In another embodiment, the second image is registered with the first image to provide registered connectivity groups representing the positions of nodes within the connectivity groups for the plurality of subregions, and the registered connectivity groups or a set of digital features representing the registered connectivity groups can be provided to the machine learning model. The clinical parameters are any continuous or categorical parameters indicative of a patient's risk associated with an impaired neuropsychiatric function, including the likelihood of exhibiting one of the impaired neuropsychiatric functions and one of the multiple impaired neuropsychiatric functions in an impaired neuropsychiatric function category including the impaired neuropsychiatric function, the likelihood that the patient has one of the impaired neuropsychiatric functions and one of the multiple impaired neurocardiological functions in an impaired neuropsychiatric function category including the impaired neuropsychiatric function, and the likelihood that the patient will respond to treatment for the impaired neuropsychiatric function. The clinical parameters may then be stored on a non-transitory computer-readable medium or displayed to a user on an associated output device.

[0074] 7 illustrates a method 700 for targeting neuromodulation in a patient's brain for use in one of improving, diagnosing, treating, and managing neuropsychiatric function. At 702, a first image showing brain structure is acquired from a first imaging system. For example, the first imaging system may be a computed tomography (CT) system or a magnetic resonance imaging (MRI) system. At 704, a second image showing brain connectivity is acquired from one of the first imaging system and a second imaging system. For example, the second imaging system may be an MRI system, and the image may be generated by diffusion tensor imaging.

[0075] At 706, for each of a plurality of voxels within the region of interest, a first utility value associated with directly modulating tissue within the region of interest is identified from the first image. In one example, this is accomplished by identifying various tissue types in the first image and assigning values ​​based on the tissue type. In another example, a portion of the first image representing the region of interest is placed into a histological atlas that assigns values ​​to the first utility value based on each of a plurality of locations within the histological atlas. For example, these values ​​can be determined based on the concentration of inhibitory or excitatory neurons expected at a given location based on data captured within the region of interest, such as histological data.

[0076] At 708, for each voxel of the plurality of voxels, a second utility value associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest is identified from the second image. In one example, for each of the plurality of voxels, a set of locations outside the region of interest that are indirectly modulated when modulating the given voxel is determined from the second image. Each set of locations has an associated indirect utility value, typically representing the adjustment of the value of the tissue. Corresponding strengths of connections between the voxel and each group of locations can also be determined from the second image, with the strengths between each group of locations and the given voxel being represented as connection weights. It should be understood that the weights can be determined as linear or nonlinear functions of connection strengths. The second utility value for each voxel can be the sum of the products of the indirect utility values ​​for that set of locations and the associated connection weights.

[0077] At 710, an overall utility value for each of the plurality of voxels is determined based on at least the first utility value and the second utility value. Each overall utility value can be determined as a linear or nonlinear function of the first utility value and the second utility value. In one example, the overall utility value is the sum of the first utility value and the second utility value. At 712, an optimal location for neuromodulation is identified based on the overall utility value for each of the plurality of voxels. For example, a suitable optimization algorithm, such as gradient descent or simulated annealing, can be used to find the optimal location. It should be understood that neuromodulation has an influence volume having a center point and a shape, and the optimal location and shape of the influence volume can be determined. In one example, the region of interest includes at least a portion of the adenoseptal nucleus and the ventral internal capsule, and the optimization can be limited so that the influence volume has a center point. The center point is located between 7 mm and 12 mm from the midline on both the left and right sides of the brain, between 1 mm and 6 mm anterior to the anterior cortex (AC), and between 2 mm superior and 2 mm inferior to the AC.

[0078] In one embodiment, functional imaging, such as functional MRI or positron emission tomography (PET), can be used to assess the effectiveness of neuromodulation and adjust targets accordingly for both current and future patients. For example, a third image showing brain activity can be acquired from an appropriate functional imaging system at a first time, and neuromodulation can be applied to the identified optimal location at a later time. A fourth image showing brain activity can be acquired from the functional imaging system at a later time. The third and fourth images can then be compared to identify the effectiveness of neuromodulation. Based on the changes in brain activity resulting from the neuromodulation, the first utility values ​​of voxels within the influence volume and the connection weight values ​​associated with voxels within the influence volume can be adjusted based on the comparison. Furthermore, target-related parameters, such as a function for calculating connection weights from connection strengths determined by fiber bundle imaging and a histology-based atlas value for the first utility value, can be adjusted for future patients.

[0079] In some embodiments, referring to Figures 8 and 9, a method for improving a patient's addictive behavior or addiction to an addictive chemical is provided. The method may include delivering FUS or DBS signals to a treatment site including the septal nucleus 802, the ventral internal capsule 804, or both. The treatment site is located 7 millimeters (mm) to 12 mm from the midline 901 on both the left and right sides of the brain, 1 mm to 6 mm anterior to the anterior association point (AC) 902, and 2 mm superior to the AC and 2 mm inferior to the AC. The midline is an imaginary line through the center of the brain. The AC 902 and the posterior association point (PC) 904 are laterally oriented connective white matter beams connecting the two cerebral hemispheres along the midline. The central association point (MCP) 906 is the midpoint between the AC and the PC. The AC, PC, and MCP are reference points or landmarks used in imaging methods such as MRI, PET, and CT. While the method described above is described for the AC, the treatment site can be determined based on the PC or MCP. 10-12, for FUS, the treatment site 1002 is the center of an influence volume 1004 that directly receives the FUS signal. While FIG. 10 shows an influence volume of 5 mm x 5 mm x 7 mm, the influence volume may vary depending on the brain region being treated (e.g., the afferent nucleus). For DBS, as shown in FIG. 12, the treatment site 1202 is the region of the brain into which the centers of the electrical contacts of the conductive wire are inserted to directly receive the electrical signal.

[0080] With respect to addiction, the neural target to which stimulation is delivered may be one component of the patient's incentive circuit, such as the adenoseptal nucleus and ventral internal capsule. Stimulation may be applied unilaterally or bilaterally to the neural target sites. Table VIII provides an exemplary list of neural target sites, exemplary forms of neuromodulation, and exemplary neuromodulation parameters that may be utilized at these neural target sites as part of a patient's treatment.

[0081] (Table VIII) TIFF2025528428000009.tif141156

[0082] In some embodiments relating to FUS, the dose includes a system frequency of 0.1 to 3 MHz, an intensity or power of 10 W to 200 W, a pulse duration of 10 msec to 1000 msec (on) and 10 msec to 990 msec (off), a total duration of 30 seconds to 30 minutes, and a number of ultrasound transducers / elements of 1 to 1024. It should be understood that various ultrasound systems have different operating parameters. Table IX includes parameters relevant to ultrasound systems that may be used with the systems and methods herein.

[0083] (Table IX) TIFF2025528428000010.tif209152 *Values ​​for three types of converters

[0084] Figure 13 is a schematic block diagram illustrating an exemplary system of hardware components capable of implementing the exemplary systems and methods shown in Figures 1-12, such as the system for neuromodulation targeting shown in Figure 1. System 1300 can include various systems and subsystems. System 1300 can be a personal computer, a laptop computer, a workstation, a computer system, an electrical appliance, an ASIC (Application Specific Integrated Circuit), a server, a server blade center, a server field, etc.

[0085] The system 1300 may include a system bus 1302, a processing unit 1304, a system memory 1306, storage devices 1308 and 1310, a communication interface 1312 (e.g., a network interface), a communication link 1314, a display 1316 (e.g., a video screen), and input devices 1318 (e.g., a keyboard, touch screen, and / or a mouse). The system bus 1302 is in communication with the processing unit 1304 and the system memory 1306. Additional storage devices 1308 and 1310, such as hard disk drives, servers, proprietary databases, or other non-volatile memory, may also be in communication with the system bus 1302. The system bus 1302 interconnects the processing unit 1304, the storage devices 206-1310, the communication interface 1312, the display 1316, and the input device 218. In some examples, the system bus 1302 further interconnects additional ports (not shown), such as universal serial bus (USB) ports.

[0086] System 1300 can be implemented within a computing cloud, where features of system 1300 (e.g., processing unit 1304, communication interface 1312, and storage devices 1308, 1310) can represent a single hardware implementation or multiple hardware implementations, and applications can run on multiple (i.e., distributed) implementations of hardware (e.g., computers, routers, memory, processors, or combinations thereof). Alternatively, system 1300 can be implemented on a single dedicated server.

[0087] The processing unit 1304 may be a computing device and may include an application specific integrated circuit (ASIC). The processing unit 1304 performs the operations of the examples of the present invention by executing a set of instructions. The processing unit may include a processing core.

[0088] Additional storage devices 1306, 1308, and 1310 may store text or compiled data, programs, instructions, database searches, and any other information necessary for the operation of the computer. Additional storage devices 1306, 1308, and 1310 may be implemented as computer-readable media (integral or removable), such as memory cards, magnetic disk drives, optical disks (CDs), or servers accessible over a network. In some examples, additional storage devices 1306, 1308, and 1310 may include text, images, video, and / or audio, portions of which may be retrieved in a human-understandable format.

[0089] Additionally or alternatively, the system 1300 can access external data or query sources via a communication interface 1312 , which is in communication with the system bus 1302 and a communication link 1314 .

[0090] In operation, system 1300 can be used to implement one or more portions of a system in accordance with the present invention. According to some examples, computer-executable logic for implementing the quality assurance system can reside in system memory 1306 and one or more of storage devices 1308 and 1310. Processing unit 1304 executes computer-executable instructions from system memory 1306 and one of storage devices 1308, 1310. It should be understood that computer-readable media can include multiple computer-readable media operably connected to the processing unit.

[0091] In some aspects, a method for improving impaired neuropsychiatric function in a patient is provided. The method includes obtaining a parameter indicative of the patient's brain connectivity, for example, measuring stress, anxiety, or craving. The parameter can be obtained by obtaining an image indicative of the patient's brain connectivity via an imaging system and extracting the parameter indicative of the patient's brain connectivity from the image. Also, physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof, associated with the patient's impaired neuropsychiatric function are measured. Treatment can be provided or adjusted based on the parameter indicative of the patient's brain connectivity and the measurement of the patient's physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof. The measurement of the physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof, can be responsive to presentation of a cue associated with a neuropsychiatric disorder indicative of impaired neuropsychiatric function.

[0092] In some aspects, a method for screening an individual with impaired neuropsychiatric function to determine whether the individual is suitable for treatment is provided. The method includes obtaining a parameter representative of the patient's brain connectivity. The parameter can be obtained by obtaining an image representative of the patient's brain connectivity via an imaging system and extracting the parameter representative of the patient's brain connectivity from the image. The method also includes measuring physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof, associated with the patient's impaired neuropsychiatric function, and determining whether the individual is a suitable candidate for treatment based on the measurement of the parameter representative of the patient's brain connectivity and the physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof. The measurement of the physiological, cognitive, psychosocial, behavioral parameters, or a combination thereof, can be responsive to presentation of a neuropsychiatric disorder-related clue indicating an impairment of the neuropsychiatric function. Specific details are provided in the above description to provide a thorough understanding of the embodiments. However, it should be understood that embodiments can be practiced without these specific details. For example, physical components can be depicted in block diagrams to avoid cluttering the embodiments with unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.

[0093] 13 The above techniques, blocks, steps, and apparatus embodiments may be implemented in various forms. For example, these techniques, blocks, steps, and apparatus may be implemented in hardware, software, or a combination thereof. For hardware embodiments, the processing unit may be implemented within one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processors (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described above, and / or combinations thereof. It should be noted that the embodiments may be described as a process, which may be depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. A flowchart may describe operations as a sequential process, but many operations may be performed in parallel or simultaneously. The order of operations may also be rearranged. A process terminates when its operations are completed, but may have additional steps not included in the diagram. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination corresponds to the function returning to the calling function or the main function. Additionally, embodiments may be implemented in hardware, software, scripting languages, firmware, middleware, microcode, hardware description languages, and / or any combination thereof. When implemented in software, firmware, middleware, scripting languages, and / or microcode, the program code or code segments to perform the necessary tasks may be stored on a machine-readable medium such as a storage medium. A code segment or machine-executable instructions may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a script, a class, or any combination of instructions, data structures, and / or programming languages.A code segment can be coupled to another code segment or a hardware circuit by passing and / or receiving information, data, arguments, parameters, and / or memory contents. Information, arguments, parameters, data, etc. can be passed, forwarded, or transmitted by any suitable means, including memory sharing, message passing, handover, network transmission, etc.

[0094] For firmware and / or software embodiments, methods may be implemented with modules (e.g., procedures, functions, etc.) that perform the functions described herein. Any machine-readable medium tangibly embodying instructions may be used to implement the methods described herein. For example, software code may be stored in a memory. The memory may be implemented within the processor or external to the processor. As used herein, the term "memory" refers to any type of long-term, short-term, volatile, non-volatile, or other storage medium, and is not limited to any particular type of memory, or any particular number of memories, or media types on which the memory is stored.

[0095] And, as disclosed herein, the term "storage medium" can refer to one or more memories for storing data, including read-only memory (ROM), random-access memory (RAM), magnetic RAM, core memory, magnetic disk storage media, optical storage media, flash memory devices, and / or other machine-readable media for storing information. The terms "computer-readable medium" and "machine-readable medium" include, but are not limited to, portable or fixed storage devices, optical storage devices, wireless channels, and / or various other storage media that contain or carry storable commands and / or data. It should be understood that a "computer-readable medium" or a "machine-readable medium" can include multiple media operatively connected to a processing unit.

[0096] The foregoing is illustrative. It is, of course, not possible to describe every conceivable combination of components or methods, but one skilled in the art will recognize that many other combinations and arrangements are possible. Accordingly, the present invention is intended to embrace all such changes, modifications, and variations that fall within the scope of this application, which includes the appended claims. As used herein, the term "comprises" means including, but not limited to, and the term "comprises" means including, but not limited to. The terms "based" and "based on" mean based at least in part on. Furthermore, when the invention or claims refer to "one," "one," "first," or "other" element or the like, this should be construed as including one or more such elements, and not necessarily excluding two or more such elements.

Claims

1. 1. A method for targeting neuromodulation in the brain of a patient for use in one of improving, diagnosing, and managing neuropsychiatric function, comprising: acquiring a first image from a first imaging system showing the brain structure; acquiring a second image indicative of the brain connectivity from one of the first imaging system and a second imaging system; identifying, for each voxel of a plurality of voxels within a region of interest, a first utility value from the first image associated with directly modulating tissue within the region of interest; determining, for each voxel of the plurality of voxels, a second utility value from the second image associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest; determining an overall utility value for each voxel of the plurality of voxels based on at least the first utility value and the second utility value; and identifying an optimal location for neuromodulation based on the overall utility value of each voxel of the plurality of voxels.

2. Determining the second utility value for each voxel of the plurality of voxels includes: for a given voxel of the plurality of voxels, identifying a set of locations outside the region of interest in the second image that are indirectly adjusted when adjusting the given voxel, each location in the set of locations having an associated indirect utility value; Identifying, from the second image, strengths of connections between the given voxel and each location in the set of locations as connection weights, respectively; 2. The method of claim 1, further comprising: determining the second utility value for the given voxel as a sum of products of the indirect utility values ​​and the connection weights for each location in the set of locations.

3. acquiring a third image indicative of brain activity from one of the first imaging system, the second imaging system, and a third imaging system at a first time; applying neuromodulation to the optimal location at a second time after the first time, the predetermined location being within a volume of influence associated with the optimal location; acquiring a fourth image indicative of the brain activity from one of the first imaging system, the second imaging system, and the third imaging system at a third time after the second time; 3. The method of claim 2, further comprising modifying the connection weights of each location in the set of locations of the given voxel based on the third image and the fourth image.

4. 2. The method of claim 1, wherein determining the first utility value for each voxel of the plurality of voxels comprises registering a portion of the first image showing the region of interest with a histology-based atlas, the histology-based atlas having an assigned value of the first utility value for each location of a plurality of locations in the histology-based atlas.

5. The method of claim 4 , wherein the assignment values ​​in the histology-based atlas indicate expected concentrations of excitatory neurons in the region of interest.

6. The method of claim 4 , wherein the assignment values ​​in the histology-based atlas indicate expected concentrations of inhibitory neurons in the region of interest.

7. acquiring a third image indicative of brain activity from one of the first imaging system, the second imaging system, and a third imaging system at a first time; applying neuromodulation to the optimal location at a second time after the first time, the applied neuromodulation having a volume of influence associated with the optimal location; acquiring a fourth image indicative of the brain activity from one of the first imaging system, the second imaging system, and the third imaging system at a third time after the second time; 5. The method of claim 4, further comprising: modifying at least one of the assignments of the first utility values ​​of the plurality of locations in the histology-based atlas based on the third image and the fourth image.

8. acquiring a third image indicative of brain activity from one of the first imaging system, the second imaging system, and a third imaging system at a first time; applying neuromodulation to the optimal location at a second time after the first time, the applied neuromodulation having a volume of influence associated with the optimal location; acquiring a fourth image indicative of the brain activity from one of the first imaging system, the second imaging system, and the third imaging system at a third time after the second time; The method of claim 1 , further comprising: modifying the first utility value for at least one voxel in the volume of influence based on the third image and the fourth image.

9. 2. The method of claim 1, wherein identifying the optimal location of neuromodulation based on the overall utility value for each voxel of the plurality of voxels comprises identifying the optimal location and optimal shape of an influence volume associated with the optimal location of neuromodulation based on the overall utility value for each voxel of the plurality of voxels.

10. applying neuromodulation to said optimal location; measuring feedback including one of a physiological parameter, a cognitive parameter, a psychosocial parameter, and a behavioral parameter of the patient; determining an efficacy of the neuromodulation based on the measured feedback; and 10. The method of claim 1, further comprising: selecting a new location within the region of interest from the global utility value for each voxel of the plurality of voxels if the efficacy of the neuromodulation does not meet a threshold.

11. applying neuromodulation to the optimal location having a volume of influence; measuring feedback including one of a physiological parameter, a cognitive parameter, a psychosocial parameter, and a behavioral parameter of the patient; determining an efficacy of the neuromodulation based on the measured feedback; and The method of claim 1 , further comprising: modifying the first utility value of at least one voxel within the volume of influence based on the determined effectiveness of the neuromodulation.

12. the region of interest includes at least a portion of the adenoseptal nucleus and the ventral internal capsule; 2. The method of claim 1, wherein identifying the optimal locations for neuromodulation based on the global utility value of each voxel of the plurality of voxels comprises identifying the optimal locations such that a volume of influence associated with the optimal locations has a center point, the center point being located between 7 mm and 12 mm from a midline on each side of the brain, between 1 mm and 6 mm anterior to the anterior association cortex (AC), and between 2 mm above and 2 mm below the AC.

13. 1. A system comprising: an imaging interface, a targeting component, and a neuromodulation system; the imaging interface receives a first image from a first imaging system showing brain structure and a second image from one of the first imaging system and a second imaging system showing brain connectivity; The targeting component comprises: identifying, for each voxel of a plurality of voxels within the region of interest, a first utility value from the first image associated with directly modulating tissue within the region of interest; identifying, for each voxel of the plurality of voxels, a second utility value from the second image associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest; determining an overall utility value for each voxel of the plurality of voxels based on at least the first utility value and the second utility value; identifying an optimal location for neuromodulation based on an overall utility value for each voxel of the plurality of voxels; The neuromodulation system delivers neuromodulation to the optimal location.

14. 14. The system of claim 13, wherein the neuromodulation system is a deep brain stimulation system, and the targeting component further identifies at least one electrode to activate based on the global utility value for each voxel of the plurality of voxels.

15. 14. The system of claim 13, wherein the neuromodulation system is a focused neuromodulation system, and the targeting component further specifies a number and direction of focal points for excitation based on the overall utility value of each voxel of the plurality of voxels.

16. a feedback component; the feedback component measures one of a physiological parameter, a cognitive parameter, a psychosocial parameter, and a behavioral parameter of the patient and determines the effectiveness of the neuromodulation based on the measured feedback; 14. The system of claim 13, wherein the targeting component selects a new location within the region of interest from the overall utility value of each voxel of the plurality of voxels if the effectiveness of the neuromodulation does not meet a threshold.

17. a feedback component; the feedback component measures one of a physiological parameter, a cognitive parameter, a psychosocial parameter, and a behavioral parameter of the patient and determines the effectiveness of the neuromodulation based on the measured feedback; The system of claim 13 , wherein the targeting component modifies the first utility value of at least one voxel within the volume of influence based on the identified effectiveness of the neuromodulation.

18. 14. The system of claim 13, wherein the targeting component selects target locations for the adenoseptal nucleus and the ventral internal capsule such that an influence volume associated with the optimal location has a center point located between 7 mm and 12 mm from the midline on both the left and right sides of the brain, between 1 mm and 6 mm anterior to the anterior association (AC), and between 2 mm superior and 2 mm inferior to the AC.

19. 1. A method for improving neuropsychiatric function in a patient, comprising: selecting an influence volume having a center point within a target region including the adenoseptal nucleus and the ventral internal capsule; and transmitting neuromodulation to the selected volume of influence; The central point is located between 7 mm and 12 mm from the midline on both the left and right sides of the brain, between 1 mm and 6 mm anterior to the anterior association (AC), and between 2 mm superior and 2 mm inferior to the AC.

20. selecting the influence volume acquiring a first image from a first imaging system indicative of a brain structure of the patient; acquiring a second image indicative of the brain connectivity from one of the first imaging system and a second imaging system; identifying, for each voxel of a plurality of voxels within a region of interest, a first utility value from the first image associated with directly modulating tissue within the region of interest; determining, for each voxel of the plurality of voxels, a second utility value from the second image associated with indirectly modulating tissue outside the region of interest by modulating tissue within the region of interest; determining an overall utility value for each voxel of the plurality of voxels based on at least the first utility value and the second utility value; and determining the volume of influence based on an overall utility value for each voxel of the plurality of voxels.

21. 20. The method of claim 19, wherein transmitting neuromodulation to the selected volume of influence comprises delivering focused ultrasound to the selected volume of influence.

22. measuring feedback including one of a physiological parameter, a cognitive parameter, a psychosocial parameter, and a behavioral parameter of the patient; determining an efficacy of the neuromodulation based on the measured feedback; and 20. The method of claim 19, further comprising: if the neuromodulation efficacy does not meet a threshold, selecting a volume of influence having a center point within a target region.

Images