System and method for identifying patient-specific location(s) for precision neuromodulation of one or more patient-specific brain network

Abstract

One or more patient-specific location(s) for a treating a medical condition can be selected for stimulation of a patient-specific brain network based on image(s) of at least a portion of the patient's brain (e.g., taken using resting state functional magnetic resonance imaging (rsfMRI), diffusion-weighted MRI (dMRI) of the brain of the patient or the like). The image(s) can be co-registered to an anatomical image (e.g., an anatomical MRI image) of the patient. For example, a first subset of the brain can be determined based on at least one brain region associated with a medical condition and the rsfMRI and a second subset of the brain can be determined based on a set of fiber tracts of the brain associated with the medical condition and on the dMRI. The one or more patient-specific location(s) can be chosen based at least in part on the first subset and the second subset.

Description

SYSTEM AND METHOD FOR IDENTIFYING PATIENT-SPECIFIC LOCATION(S) FOR PRECISION NEUROMODULATION OF ONE OR MORE PATIENT-SPECIFIC BRAIN NETWORKSTATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0001] This invention was made with government support under grants K01MH117346, R01MH136160, R21 MH126396, and R61MH132869 awarded by the National Institutes of Health. The government has certain rights in the invention.RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Application Serial No.63 / 789,561 , filed April 16, 2025, entitled “IDENTIFICATION OF PATIENT-SPECIFIC BRAIN TARGETS FOR DEVICE NEUROTHERAPEUTICS”. The entirety of this provisional application is hereby incorporated by reference for all purposes.TECHNICAL FIELD

[0002] This description relates generally to precision neuromodulation of a patient-specific brain network, and more specifically to systems and methods for selecting one or more patient-specific locations (e.g., target(s) and associated location(s)) for precision neuromodulation of the patient-specific brain network based on a comparison of different images of at least a portion of the patient’s brain.BACKGROUND

[0003] Patients suffering from brain conditions / diseases (referred to as “medical conditions”) have a renewed chance for successful treatment with new treatments like precision neuromodulation (e.g., including non-invasive treatments like transcranial magnetic stimulation (TMS) or transcranial ultrasound stimulation (TUS), invasive treatments like intracranial electrical stimulation with implanted electrodes (e.g., deep brain stimulation) or responsive neurostimulation (RNS), and the like). The medical conditions can include, but are not limited to, depression (including major depressive disorder), anxiety, movement disorders (including Parkinson’s disease), epilepsy, obsessive compulsive disorder (OCD), major depressive disorder (MDD), smoking addiction, migraines, other neurological / psychiatric conditions, and the like. The goal of precisionneuromodulation is to treat different medical conditions by engaging different brain region(s) and / or network(s) in various ways. Unfortunately, precision neuromodulation does not yet provide a reliable treatment efficacy. This is at least in part because the treatment efficacy strongly depends on the proper selection of patient-specific locations for stimulation, but the patient-specific locations are not properly selected for every patient. In fact, no rigorous scientific techniques exist that can guide the selection of such patient-specific locations for every patient suffering from different medical conditions.SUMMARY

[0004] Described herein is a rigorous scientific technique that can guide the selection of such patient-specific location(s) for every patient with different brain diseases / conditions (collectively referred to as “medical conditions”) based on images of at least a portion of the patient’s brain.

[0005] In some instances, a non-transitory machine-readable medium having machine-executable instructions for a treatment targeting system for patient-specific neurotherapeutics (also referred to as, “precision therapeutics”) that causes a processor core to execute operations is described. The operations include accessing a first dataset including resting state functional magnetic resonance imaging (rsfMRI) of a brain of a patient. The operations also include accessing a second dataset including diffusion-weighted MRI (dMRI) of the brain of the patient. The operations additionally include accessing a third dataset including anatomical MRI of the brain of the patient. The operations further include co-registering the first dataset and the second dataset to the third dataset to create a co-registered dataset. Additionally, the operations include using the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset, and to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset. In addition, the operations include targeting a patient-specific location to deliver a treatment of the medical condition based at least in part on the first subset of the brain and the second subset of the brain. The treatment is delivered to the location to treat the medical condition.

[0006] In other instances, a treatment targeting system for patient-specific neurotherapeutics is described. The system includes a memory for storing machine-readable instructions and a processor core for accessing the machine-readable instructions and executing the machine-readable instructions as operations. The operations include accessing a first dataset including resting state functional magnetic resonance imaging (rsfMRI) of a brain of a patient. The operations also include accessing a second dataset including diffusion-weighted MRI (dMRI) of the brain of the patient. The operations additionally include accessing a third dataset including anatomical MRI of the brain of the patient. The operations further include co-registering the first dataset and the second dataset to the third dataset to create a co-registered dataset. Additionally, the operations include using the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset, and to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset. In addition, the operations include targeting a patient-specific location to deliver a treatment of the medical condition based at least in part on the first subset of the brain and the second subset of the brain. The treatment is delivered to the location to treat the medical condition.

[0007] In further instances, a method for targeting patient-specific neurotherapeutics to a patient-specific location is described. The method includes accessing, by a system comprising a processor, a first dataset including resting state functional magnetic resonance imaging (rsfMRI) of a brain of a patient. The method also includes accessing, by the system, a second dataset including diffusion-weighted MRI (dMRI) of the brain of the patient. The method additionally includes accessing, by the system, a third dataset including anatomical MRI of the brain of the patient. The method further includes co-registering, by the system, the first dataset and the second dataset to the third dataset to create a co-registered dataset.Additionally, the method includes using, by the system, the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset, and to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset. In addition, the method includes targeting, by the system, a patient-specific location in the brain to deliver a treatment of the medical condition based at least in part on the first subset of the brain and thesecond subset of the brain. The treatment is delivered to the location to treat the medical condition.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 illustrates an example computing environment implementing a treatment targeting system capable of selecting one or more patient-specific locations for treatment of a certain brain disease / condition based on one or more images of at least a portion of the patient’s brain.

[0009] FIG. 2 illustrates an example series of images associated with selecting a patient-specific location for transcranial magnetic stimulation (TMS) treatment of major depressive disorder (MDD) in one example.

[0010] FIG. 3 illustrates an example image showing the average target regions within the dorsolateral prefrontal cortex (DLFPC) and inferior parietal lobe (IPL) across 100 subjects.

[0011] FIG. 4 illustrates an example plot showing the impact of TMS as a relative changes in effectiveness of TMS (Hamilton Depression Rating Scale, HAMD) based on the stimulated target fraction (STF) of patient-specific locations.

[0012] FIG. 5 illustrates example images showing fields from applied TMS along with patient-specific locations and the associated STF for six subjects.

[0013] FIG. 6 illustrates a flow diagram of a method for selecting one or more patient-specific location for precision neuromodulation of a brain network to treat a medical condition.DETAILED DESCRIPTION

[0014] Using precision neuromodulation (also referred to herein as “neurotherapeutics”), targeted energy can be applied to one or more patient-specific locations (related to targets in the patient’s brain) to activate different brain region(s) and / or network(s) and modulate certain neural circuits to treat a variety of different brain conditions / diseases (referred to as “medical conditions”). Precision neuromodulation includes invasive treatments like transcranial magnetic stimulation (TMS) or transcranial ultrasound stimulation (TUS), invasive treatments like intracranial electrical stimulation with implanted electrodes (e.g., deep brain stimulation) or responsive neurostimulation (RNS), and the like. The medicalconditions can include, but are not limited to, depression (including major depressive disorder), anxiety, movement disorders (including Parkinson’s disease), epilepsy, obsessive compulsive disorder (OCD), major depressive disorder (MDD), smoking addiction, migraines, other neurological / psychiatric conditions, and the like. It should be understood that patient-specific location(s) associated with the one or more patient specific targets are referred to collectively herein as patient-specific locations.

[0015] In a normal brain, brain network(s) (located in one or more brain regions) can be activated to modulate certain neural circuits. For different medical conditions, one or more brain networks can be identified as causing or influencing the medical condition. Precision neuromodulation aims to treat the medical condition by stimulating one or more patient-specific location to artificially activate one or more brain network(s) to modulate associated neural circuits. For example, the dorsolateral prefrontal cortex (DLPFC) has been identified as a cause of MDD and TMS can be applied to the DLPFC to treat MDD. The treatment efficacy of precision neuromodulation strongly depends on the proper selection of the one or more patient-specific locations and remains low because the one or more patient-specific locations are not properly selected for every patient using traditional techniques. For example, the one or more brain networks can be within relatively large regions of the brain, and brain variations between patients impacts which portions of those regions are most strongly associated with the condition(s). Described herein are systems and methods that can guide the selection of such patient-specific locations for patients suffering from different medical conditions based on a comparison of different images of at least a portion of the patient’s brain, thereby enhancing the reliability of targeting patient-specific brain networks and choosing one or more patient-specific locations.

[0016] FIG. 1 illustrates an example computing environment 100 implementing a treatment targeting system 102 capable of selecting a patient-specific location for treatment of a medical condition via precision neuromodulation. The treatment targeting system 102 selects the patient-specific location based on analysis of patient-specific imaging data 104 (e.g., shown at two possible locations in FIG. 1). Although the imaging data can include many different types of functional images combined with anatomical images, for ease of description and illustration rsfMRI, dMRI, and anatomical MRI datasets of the brain of the patient are described herein.Additionally, TMS is a specific type of precision neuromodulation that is described herein for ease of description and illustration.

[0017] The computing environment 100 includes a processor core 110, a memory 112, a user input / output (I / O) interface 114, and a computing environment communication interface 116, which are operably connected for computer communication. The processor core 110 performs general computing to execute instructions stored in the memory 112, including instructions associated with treatment targeting system 102. The instructions cause the processor core 110 to execute operations. The memory 112 also stores instructions associated with an operating system that controls and / or allocates resources of computing environment 100, including resources associated with treatment targeting system 102. Memory 112 represents a non-transitory machine-readable memory (or other medium), such as random access memory (RAM), a solid state drive, a hard disk drive or a combination thereof.

[0018] Processor core 110 accesses memory 112 and executes the machine-readable instructions as operations. Processor core 110 can be a variety of various processors including multiple single- and multi-core processors, co-processors, and other multiple single and multicore processor and co-processor architectures.

[0019] User I / O interface 114 provides software and hardware to facilitate data input and output between computing environment 100 and a user. This can include input devices such as a keyboard, mouse, touchpad, touchscreen, microphone, etc., as well as output devices such as display(s) (e.g., light-emitting diode (LED) display panel(s), liquid crystal display (LCD) panel(s), plasma display panel(s), and / or touch screen display(s), etc.), speaker(s), etc. User I / O interface 114 provides graphical input controls for a user interface, which can include software and hardware-based controls, interfaces, touch screens, or touch pads or plug and play devices for a user to provide user input.

[0020] The memory 112 includes the treatment targeting system 102, which includes an imaging analysis model 118 and a target selection module 120 that operate in concert and / or stages to select a target location for treatment of the medical condition based on the patient-specific imaging data 104. In various examples, the treatment targeting system 102, based on a medical condition (e.g., associated with a specific brain network, etc.) of a patient to be treated via precision neuromodulation, analyzes multiple modalities of imaging (e.g., rsfMRI, dMRI, and / oranatomical MRI such as TlwMRI and / or T2wMRI stored in the patient-specific imaging data 104, etc.) of the brain of the patient and selects a target location for the treatment based on: functional parcels of the brain associated with the condition and / or brain network, fiber tracts of the brain associated with the condition and / or brain network, structural features of the brain associated with a treatment type of the treatment, etc.

[0021] The imaging analysis module 118 accesses MRI datasets in the patient-specific imaging data 104, including: a first dataset that includes rsfMRI of the brain of the patient, a second dataset that includes dMRI of the brain of the patient, and a third dataset that includes anatomical MRI (e.g., TlwMRI and / or T2wMRI, etc.) of the brain of the patient. The imaging analysis module 118 additionally co-registers the first dataset and the second dataset to the third dataset.

[0022] The imaging analysis module 118 analyzes the rsfMRI dataset to determine a first subset of the brain that includes brain region(s) associated with the medical condition to be treated, for example, parcels of the brain that are part of a functional brain network associated with the condition. In various examples, the imaging analysis module 118 defines a patient-specific brain parcellation that defines parcels of the brain that activate synchronously or simultaneously with each other in the rsfMRI data, indicating functional brain network connections between those parcels. Based on the patient-specific brain parcellation and a target functional brain network (e.g., the FPN, the salience network, the default mode network, etc.) associated with the medical condition to be treated, in various examples the imaging analysis module 118 determines the first subset by determining a subset of the brain parcels of the patient-specific brain parcellation that are within the target functional brain network.

[0023] The imaging analysis module 118 analyzes the dMRI dataset to determine a second subset of the brain that includes brain region(s) associated with the medical condition to be treated, for example, fiber tracts of the functional brain network associated with the condition and / or cortical regions connected to those fiber tracts. In various examples, the imaging analysis module 118 determines the second subset by extracting a subset of the tractography of the dMRI dataset to estimate fiber tracts associated with the brain network associated with the condition, for example, based on the underlying fiber density.

[0024] In some examples, the imaging analysis module 118 analyzes the anatomical MRI dataset to determine a third subset of the brain that includes portions of the brain that are relevant to the type of treatment (e.g., TMS is more effective on gyri than on sulci, etc.).

[0025] The target selection module 120 selects the target location based on one or more of: the first subset (e.g., from the rsfMRI dataset, etc.), the second subset (e.g., from the dMRI dataset, etc.), the third subset (e.g., from the anatomical MRI dataset, etc.), the treatment type, and / or the condition to be treated. In various examples, the target selection module 120 selects the target location based on the intersection of the first subset and the second subset. In some such examples, the target selection module selects the target location as the intersection of the first subset, the second subset, and the third subset.

[0026] Additionally, in various examples, the target selection module 120 further refines the target location based on the condition and / or the treatment type. For example, when the brain network is the FPN, the condition is MDD, and the treatment type is TMS, the intersection of the first subset, the second subset, and the third subset includes patient-specific regions of the DLPFC and the inferior parietal lobe (IPL). TMS applied to the DLPFC has been identified as effective in treating MDD, and in some such examples, the target selection module 120 selects the patient-specific region of the DLPFC as the target location.

[0027] The target brain network and the associated brain regions that include parcels and fiber tract(s) can vary based on the medical conditions to be treated. In a first example, the target brain network is the FPN, the brain regions are the DLPFC and the IPL, and fiber tracts include superior longitudinal fasciculus (SLF)-II tracts. In a second example, the target brain network is the salience network, the brain regions include the ventromedial prefrontal cortex (vmPFC), anterior insula (Al), and anterior cingulate cortex (ACC), and the fiber tracts include the cingulum bundle (CB) (connected to the ACC) and the uncinate fasciculus (UF) (connected to the Al and vmPFC). In a third example, the target brain network is the default mode network, the brain regions include the medial prefrontal cortex (mPFC), posterior cingulate cortex (PCC), and the angular gyrus, and the fiber tracts include the cingulum bundle (CB) (connecting the mPFC and PCC) and the superior longitudinal fasciculus (SLF)-I fiber tracts (connecting the mPFC and the angular gyrus). Other examples are also useable in connection with various examples.

[0028] Additionally, other networks that connect known brain regions via known fiber tracts can be used in various examples, including brain networks not yet identified and / or for conditions not yet known to be treatable via precision neuromodulation, including via precision neuromodulation treatment types not yet employed, once sufficient information regarding the brain network, condition, and / or treatment type is available. In various examples, for a given brain network to be stimulated to treat a given medical condition via a given treatment type of precision neurotherapeutics, patient-specific parcels in brain regions of the given brain network are selected by the target selection module 120 based on a first subset determined by the imaging analysis module 118 from the rsfMRI dataset, a second subset determined by the imaging analysis module 118 from the dMRI dataset, and in some examples the overlapping region(s) from the rsfMRI and the dMRI datasets are further refined based on a third subset by the imaging analysis module 118 from the anatomical MRI dataset, and / or based on the condition or characteristics of the given treatment type (e.g., effectiveness in connection with different portions of brain anatomy, etc.) to select a target location for treatment of the given medical condition.

[0029] Targeting of TMS uses one of two types of methods: standard and advanced (brain image-based). The standard method uses a one-size-fits-all approach to place the TMS coil at the scalp 5-cm anterior to the primary hand-motor hotspot. Examples combine structural connectivity based on diffusion-weighted MRI tractography with functional connectivity from rsfMRI to enhance the reliability of the targeting results. There is a recent advanced targeting method based on MRI data that was approved by the FDA in the Stanford Accelerated Intelligent Neuromodulation Technique (SAINT) approach. The SAINT approach uses patientspecific resting-state functional MRI (rsfMRI)-based functional connectivity maps to define a functional subunit (e.g., a cluster of voxels) as the brain target. However, targeting approaches based on resting state functional connectivity are highly unreliable and have great uncertainty. Additionally, the SAINT targeting method is specific to defining the dorsolateral prefrontal cortex for the treatment of major depressive disorder using TMS and is not generalizable to other brain regions, disorders, or treatment modalities. Using SAINT, information from a resting-state functional MRI (rsfMRI) dataset, a diffusion-weighted MRI (dMRI) dataset, and / or an anatomical MRI (e.g., T1 -weighted (T1w) MRI and / or T2-weighted (T2w) MRI, etc.)dataset can be integrated to define a target location specific to the brain of the patient, such as a superficial patch of the cortical brain surface (e.g., within the DLPFC, etc.) as the target for treatment via TMS, a deeper location in or near a subcortical structure (e.g., subthalamic nucleus (STN), internal globus pallidus (GPi), etc.) for treatment via DBS or TUS, etc.

[0030] FIGS. 2-5 show experimental results of an experimental use of the example computing environment 100 of FIG. 1 in selecting patient-specific locations for TMS treatment of MDD.

[0031] The rsfMRI dataset is applied to estimate patient-specific brain parcellations based on the correlation between brain regions. The cortical region related to the targeted brain network is extracted by examples to localize the functional connectivity-guided target region. One example use case involves the cortical regions of the FPN defined based on subject-specific brain parcellation. The FPN includes the DLPFC and the angular gyrus.

[0032] 'The DLPFC is the brain region for FDA-cleared TMS treatment protocols for major depressive disorder. However, brain regions such as the DLPFC are usually very large and heterogeneous, and accurate identification of the patientspecific coordinates of the subregions is relevant for the treatment indication of interest in order to maximize efficacy and safety.

[0033] The dMRI dataset is applied in to estimate the whole brain tractography which characterizes the structural connections between brain regions. Various examples extract target fiber cluster(s) related to the targeted brain network(s) to further identify the target region based on structural connectivity. For the example FPN use case, the target fiber cluster(s) include the cortical region of the superior longitudinal fasciculus (SLF)-II fiber bundles that connect the DLPFC and the angular gyrus of the parietal lobe. The overlapping region between the cortical area of the target fiber cluster(s) (e.g., of the SLF-II fiber bundles, etc.) is applied in conjunction with the rsfMRI dataset to further refine the target region. The combination of rsfMRI and dMRI data provides a target location based on both functional connectivity and structural connectivity.

[0034] The anatomical MRI is applied to further refine the brain region defined by functional connectivity and structural connectivity-guided targets. In various examples, the refinement based on anatomical MRI includes either or both of two refinements. The first refinement involves removing outlier regions based on ananatomical prior. As one example, the region between the functional and structural connectivity-guided FPN in the prefrontal cortex (PFC) may extend to regions outside the standard DLPFC. Using an anatomical prior (based on e.g., the Brodmann atlas, etc.) these outlier regions are removed. The second refinement involves removing regions in the target location that are poorly stimulated (or not stimulated) via the planned treatment technique. Continuing with the example use case of using TMS to stimulate the FPN, sulci that are poorly stimulated by TMS are removed from the target location in some examples. TMS induces a stronger electric field at the gyrus surface, while regions in the depth of the sulcus are difficult to target via TMS. In some examples, deeper portions of a target region within sulci are removed, providing a more realistic and effective target for TMS that can be effectively stimulated given the biophysical properties of TMS-induced electric fields. The combination of rsfMRI, dMRI, and anatomical MRI used provides a target location (e.g., cortical target patch in TMS use cases, etc.) selected for a specific patient based on functional connectivity, structural connectivity, and anatomical knowledge.

[0035] FIG. 2 illustrates a series of images 210-290 associated with selection of a patient-specific location for TMS treatment of MDD in one example. Based on patient-specific MRI datasets (e.g., a rsfMRI dataset, a dMRI dataset, and an anatomical MRI dataset such as TlwMRI and / or T2wMRI, etc.), a specific location is selected to apply TMS to treat MDD in the patient.

[0036] Image 210 shows an example of rsfMRI. In image 220, the rsfMRI dataset is analyzed to define a subject-specific brain parcellation. Regions with the same color indicated strongly synchronized activity that forms a functional brain network. In image 230, a subset of the parcellations in 220 that form the FPN are indicated.

[0037] Image 240 shows dMRI acquired from the same subject to compute the whole-brain tractography, which characterizes the structural / axonal connectivity between brain regions. In image 250, a subset of the tractography is extracted to estimate the SLF-II fiber tracts connecting the DLPFC and the IPL. In image 260, the brain regions connected to the SLF-II tracts are computed and highlighted by the underlying fiber density.

[0038] TMS induces effective stimulation in the cortical surfaces (e.g., cortical gyrus) with much weaker stimulation in the sulcus. In image 270, anatomical MRI isused to estimate the sulcus map to refine the target. Image 280 shows the overlap region between the rsfMRI-based FPN regions of image 230 and the tractography-based cortical regions of image 260. The regions in image 290 are refined based on the sulcus maps to define brain targets in the DLPFC and IPL.

[0039] While FIG. 2 shows one example of combining data from rsfMRI, dMRI, and anatomical MRI to select a target location for precision neuromodulation, in various examples actions discussed in connection with FIG. 2 are performed via different algorithms, with different parameters and / or images, and / or in a different order (and / or simultaneously). Additionally, although images 210-230 are discussed before images 240-260, which are discussed before images 270-290, in various examples the rsfMRI, dMRI, and anatomical MRI datasets can be analyzed in any order and / or simultaneously.

[0040] FIG. 2 shows one example to illustrate a potential method for selecting a target location for one specific brain network (e.g., the frontoparietal network). In various examples, techniques discussed herein can be applied to other brain networks by choosing suitable diffusion MRI tractography fiber bundles (e.g., via a dMRI dataset) and brain parcellations (e.g., via a rsfMRI dataset).

[0041] Various examples use rsfMRI to define brain network parcellations in connection with selection of a patient-specific location, in contrast to using rsfMRI to compute the functional connectivity and select a single voxel based on the functional connectivity map as the brain target. Additionally, various examples combine diffusion MRI tractography (and in some examples, additional data from anatomical MRI) with brain parcellations identified via rsfMRI to further refine patient-specific locations.

[0042] Additionally, while FIG. 2 discusses selecting a patient-specific location for the application of TMS, various examples can select target locations for other device neurotherapeutic techniques applied to the same region or network (e.g., the FPN) and / or other specific brain regions or networks.

[0043] The reliability and reproducibility of various examples were tested using a dataset of 100 subjects. Referring to FIG. 3, illustrated is imaging showing the average patient-specific locations within the DLFPC and IPL across the subjects. The target regions, especially the DLPFC region on the left side of the image, are very dispersed, indicating significant inter-subject variability of the patient-specificlocations. FIG. 3 shows one advantage of the selection of patient-specific locations, since a fixed target location is not optimal for all subjects.

[0044] A retrospective analysis was performed to evaluate the relevance of selecting patient-specific locations for TMS treatment for major depressive disorders. The treatment was performed using only functional MRI-based voxel selection for targeting. The stimulated target fraction (STF) for each subject was computed to quantify if the actual target is consistent with the patient-specific location, such that a higher STF indicates the treatment was applied closer to the patient-specific location according to various examples.

[0045] FIG. 4 shows a chart of the impact of TMS based on the stimulated target fraction (STF) of patient-specific locations. As can be seen in FIG. 4, a higher STF is correlated to better treatment response. Referring to FIG. 5, illustrated are images showing fields from applied TMS along with patient-specific locations and the associated STF for six subjects. The three subjects on the right side of FIG. 5 (with higher STF) responded to treatment, while the three subjects on the left side of FIG.5 (with lower STF) did not respond to treatment.

[0046] In view of the foregoing structural and functional features described above, an example method will be better appreciated with reference to FIG. 6.While, for purposes of simplicity of explanation, the example method of FIG. 6 are shown and described as executing serially, it is to be understood and appreciated that the present examples are not limited by the illustrated order, as some actions could in other examples occur in different orders, multiple times and / or concurrently from that shown and described herein. Moreover, it is not necessary that all described actions be performed to implement a method.

[0047] Referring to FIG. 6, illustrated is a flow diagram of a method 600 for selecting a patient-specific location for precision neuromodulation of a brain network. In other examples, the blocks of the example method 600 are a set of machine-readable instructions on a non-transitory machine-readable medium or are a set of operations performed by a processor executing machine-readable instructions as the operations. The method uses specific kinds of MRI imaging but it will be understood that this is only for ease of explanation. Other forms of imaging can be used as will be understood by reading this example.

[0048] At block 610, the method 600 includes accessing (e.g., via the computing environment 100, the treatment targeting system 102, the imaginganalysis module 118, etc.) a first dataset (e.g., within the patient-specific imaging data 104, etc.) that includes previously obtained rsfMRI of a brain of a patient who has a medical condition to be treated via the precision neuromodulation. At block 620, the method 600 includes accessing (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) a second dataset (e.g., within the patient-specific imaging data 104, etc.) that includes previously obtained dMRI of the brain of the patient. The images need only record different structural and / or functional properties. Additionally, at block 630, the method 600 includes accessing (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) a third dataset (e.g., within the patient-specific imaging data 104, etc.) that includes previously obtained anatomical MRI (e.g., T1w MRI and / or T2w MRI, etc.) of the brain of the patient. This image need only be an anatomical image. At block 640, the method 600 includes co-registering (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) the first (e.g., rsfMRI) dataset and the second (e.g., dMRI) dataset to the third (e.g., anatomical MRI) dataset.

[0049] At block 650, the method 600 includes determining (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) a first subset of the brain based on brain region(s) associated with the medical condition and the first (e.g., rsfMRI) dataset. In various examples, based on the first (e.g., rsfMRI) dataset, a patient-specific brain parcellation is determined, indicating brain parcels that activate synchronously / simultaneously, indicating they are connected within a functional brain network. Based on the patient-specific brain parcellation and a known brain network associated with the condition to be treated, patient-specific brain parcels within functional regions of the known brain network are determined (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) as the first subset.

[0050] At block 660, the method 600 includes determining (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) a second subset of the brain based on fiber tract(s) associated with the medical condition and the second (e.g., dMRI) dataset. In various examples, based on the second (e.g., dMRI) dataset, patient-specifictractography is determined, indicating white matter tracts that connect functional regions of the brain of the patient. Based on the patient-specific tractography and a known brain network associated with the condition to be treated, functional regions connected to patient-specific fiber tracts of the known brain network are determined (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) as the second subset.

[0051] At block 670, the method 600 includes determining (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) a third subset of the brain based on anatomical features of the brain related to the treatment type (e.g., portions of the brain where the treatment is more or less effective, such as gyri / sulci, etc.) and the third (e.g., anatomical MRI) dataset. In various examples, based on the third (e.g., anatomical MRI) dataset, patient-specific brain regions where treatment is determined, indicating white matter tracts that connect functional regions of the brain of the patient. Based on the patient-specific anatomical features of the brain relative to the treatment type, anatomical features where the treatment type has improved effectiveness are determined (e.g., via the computing environment 100, the treatment targeting system 102, the imaging analysis module 118, etc.) as the third subset.

[0052] At block 680, method 600 includes selecting (e.g., via the computing environment 100, the treatment targeting system 102, the target selection module 120, etc.) a target location for the treatment based on the first subset (e.g., determined from the rsfMRI dataset), the second subset (e.g., determined from the dMRI dataset), and / or the third subset (e.g., determined from the anatomical MRI dataset). In various examples, the third subset is determined based on the intersection between the first subset, the second subset, and / or the third subset. In some examples, the intersection between the first subset, the second subset, and / or the third subset is further refined, such as based on the condition to be treated (e.g., TMS applied to the DLPFC can be effective in treating MDD, thus the intersection can be refined to a portion of the intersection within the DLPFC, etc.).

[0053] At block 690, method 600 includes applying the treatment to the target location selected in block 680. The treatment can be applied with parameters (e.g., frequency, duration, intensity, etc.) appropriate for the treatment type and condition to be treated. However, the patient-specific location selected at block 680 provides for greater treatment reliability and effectiveness.

[0054] What have been described above are examples. It is, of course, not possible to describe every conceivable combination of components or methodologies, but one of ordinary skill in the art will recognize that many further combinations and permutations are possible to target different regions, for the benefit of different brain disorders, using a variety of interventional neurotechnologies. Accordingly, the disclosure is intended to embrace all such alterations, modifications, and variations that fall within the scope of this application, including the appended claims. As used herein, the term "includes" means includes but not limited to, the term "including" means including but not limited to. The term "based on" means based at least in part on. Also as used herein, the term "set" means one or more elements (e.g., where the elements can be anything, such as datasets, nodes, relationships, etc.), and a “subset” of a set A refers to any set B where every element of set B is an element of set A (note that every set A is a subset of itself, as every element of set A is an element of set A). Additionally, where the disclosure or claims recite "a," "an," "a first," or "another" element, or the equivalent thereof, it should be interpreted to include one or more than one such element, neither requiring nor excluding two or more such elements.

[0055] In this description, unless otherwise stated, "about," "approximately" or "substantially" preceding a parameter means being within + / - 10 percent of that parameter. Modifications are possible in the described embodiments, and other embodiments are possible, within the scope of the claims.

Claims

CLAIMSWhat is claimed is:

1. A non-transitory machine-readable medium having machine-executable instructions for a treatment targeting system for patient-specific neurotherapeutics that cause a processor core to execute operations, the operations comprising:accessing a first dataset comprising resting state functional magnetic resonance imaging (rsfMRI) images of a brain of a patient;accessing a second dataset comprising diffusion-weighted magnetic resonance imaging (dMRI) images of the brain of the patient;accessing a third dataset comprising anatomical magnetic resonance imaging (MRI) images of the brain of the patient;co-registering the first dataset and the second dataset to the third dataset to create a co-registered dataset;using the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset;using the co-registered dataset to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset; andtargeting a location in the brain of a patient-specific location to deliver a treatment of the medical condition based at least in part on the first subset of the brain and the second subset of the brain,wherein the treatment is delivered to the location to treat the medical condition.

2. The non-transitory machine-readable medium of claim 1 , wherein the at least one brain region associated with the medical condition comprises a first brain region associated with the medical condition and a second brain region associated with the medical condition.

3. The non-transitory machine-readable medium of claim 2, wherein the first subset of the brain comprises a first parcel of the brain within the first brain regionand a second parcel of the brain within the second brain region, and wherein the first dataset comprises rsfMRI data indicating simultaneous activation of the first parcel of the brain and the second parcel of the brain.

4. The non-transitory machine-readable medium of claim 2, wherein the set of fiber tracts comprises fiber tracts that connect the first parcel of the brain to the second parcel of the brain.

5. The non-transitory machine-readable medium of claim 4, wherein the medical condition is major depressive disorder, wherein the first brain region is a dorsolateral prefrontal cortex (DLPFC) and the second brain region is an inferior parietal lobule (IPL) within the frontoparietal network (FPN), and wherein the fiber tracts that connect the first parcel of the brain to the second parcel of the brain comprise superior longitudinal fasciculus (SLF)-II tracts.

6. The non-transitory machine-readable medium of claim 4, wherein the medical condition is associated with the salience network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region, wherein the first brain region is a ventromedial prefrontal cortex (vmPFC), the second brain region is an anterior insula (Al), and the third brain region is an anterior cingulate cortex (ACC), and wherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or uncinate fasciculus (LIF) tracts.

7. The non-transitory machine-readable medium of claim 4, wherein the medical condition is associated with the default mode network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region, wherein the first brain region is a medial prefrontal cortex (mPFC), the second brain region is a posterior cingulate cortex (PCC), and the third brain region is an angular gyrus, and wherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or superior longitudinal fasciculus (SLF)-I tracts.

8. The non-transitory machine-readable medium of claim 1 , wherein the operations further comprise determining a third subset of the brain based on a treatment type of the treatment and on the third dataset, and wherein selecting the target location for the treatment of the medical condition is further based on the third subset of the brain.

9. The non-transitory machine-readable medium of claim 8, wherein the treatment type of the treatment is a transcranial magnetic stimulation (TMS), and wherein the third subset comprises one or more gyri of the brain.

10. The non-transitory machine-readable medium of claim 8, wherein the treatment type of the treatment is a transcranial ultrasound stimulation (TUS).

11. The non-transitory machine-readable medium of claim 1 , wherein a treatment type of the treatment is a deep brain stimulation (DBS) via an implantable pulse generator (IPG), and wherein the target location is a target location for implantation of the IPG.

12. The non-transitory machine-readable medium of claim 1 , wherein the operations further comprise applying the treatment to the location to treat the medical condition.

13. The non-transitory machine-readable medium of claim 1 , wherein the anatomical MRI comprises at least one of T1 -weighted (T1w) MRI or T2-weighted (T2w) MRI.

14. A treatment targeting system for patient-specific neurotherapeutics, the system comprising:a memory for storing machine-readable instructions; anda processor core for accessing the machine-readable instructions and executing the machine-readable instructions as operations, the operations comprising:accessing a first dataset comprising resting state functional magnetic resonance imaging (rsfMRI) of a brain of a patient;accessing a second dataset comprising diffusion-weighted magnetic resonance imaging (dMRI) of the brain of the patient;accessing a third dataset comprising anatomical magnetic resonance imaging (MRI) of the brain of the patient;co-registering the first dataset and the second dataset to the third dataset to create a co-registered dataset;using the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset;using the co-registered dataset to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset; andtargeting a location in the brain of a patient-specific location to deliver a treatment of the medical condition based at least in part on the first subset of the brain and the second subset of the brain,wherein the treatment is delivered to the location to treat the medical condition.

15. The treatment targeting system of claim 14, wherein the at least one brain region associated with the medical condition comprises a first brain region associated with the medical condition and a second brain region associated with the medical condition.

16. The treatment targeting system of claim 15, wherein the first subset of the brain comprises a first parcel of the brain within the first brain region and a second parcel of the brain within the second brain region, and wherein the first dataset comprises rsfMRI data indicating simultaneous activation of the first parcel of the brain and the second parcel of the brain.

17. The treatment targeting system of claim 15, wherein the set of fiber tracts comprises fiber tracts that connect the first parcel of the brain to the second parcel of the brain.

18. The treatment targeting system of claim 17, wherein the medical condition is major depressive disorder, wherein the first brain region is a dorsolateral prefrontal cortex (DLPFC) and the second brain region is an inferior parietal lobule (IPL), and wherein the fiber tracts that connect the first parcel of the brain to the second parcel of the brain comprise superior longitudinal fasciculus (SLF)-II tracts.

19. The treatment targeting system of claim 17,wherein the medical condition is associated with the salience network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region,wherein the first brain region is a ventromedial prefrontal cortex (vmPFC), the second brain region is an anterior insula (Al), and the third brain region is an anterior cingulate cortex (ACC), andwherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or uncinate fasciculus (UF) tracts.

20. The treatment targeting system of claim 17,wherein the medical condition is associated with the default mode network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region, wherein the first brain region is a medial prefrontal cortex (mPFC), the second brain region is a posterior cingulate cortex (PCC), and the third brain region is an angular gyrus, andwherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or superior longitudinal fasciculus (SLF)-I tracts.

21. The treatment targeting system of claim 14, wherein the operations further comprise determining a third subset of the brain based on a treatment type of the treatment and on the third dataset, and wherein selecting the target location for the treatment of the medical condition is further based on the third subset of the brain.

22. The treatment targeting system of claim 21 , wherein the treatment type of the treatment is a transcranial magnetic stimulation (TMS), and wherein the third subset comprises one or more gyri of the brain.

23. The treatment targeting system of claim 14, wherein the treatment type of the treatment is a transcranial ultrasound stimulation (TUS).

24. The treatment targeting system of claim 14, wherein the treatment type of the treatment is a deep brain stimulation (DBS) via an implantable pulse generator (IPG), and wherein the target location is a target location for implantation of the IPG.

25. The treatment targeting system of claim 14, wherein the operations further comprise applying the treatment to the target location to treat the medical condition.

26. The treatment targeting system of claim 14, wherein the anatomical MRI comprises at least one of T1 -weighted (T1w) MRI orT2-weighted (T2w) MRI.

27. A method for targeting patient-specific neurotherapeutics to a patient-specific location, the method comprising:accessing, by a system comprising a processor, a first dataset comprising resting state functional magnetic resonance imaging (rsfMRI) of a brain of a patient;accessing, by the system, a second dataset comprising diffusion-weighted magnetic resonance imaging (dMRI) of the brain of the patient;accessing, by the system, a third dataset comprising anatomical magnetic resonance imaging (MRI) of the brain of the patient;co-registering, by the system, the first dataset and the second dataset to the third dataset to create a co-registered dataset;using, by the system, the co-registered dataset to determine a first subset of the brain based on at least one brain region associated with a medical condition of the patient and on the first dataset;using, by the system, the co-registered dataset to determine a second subset of the brain based on a set of fiber tracts of the brain associated with the medical condition and on the second dataset;targeting, by the system, a location in the brain of a patient-specific location to deliver a treatment of the medical condition based at least in part on the first subset of the brain and the second subset of the brain,wherein the treatment is delivered to the location to treat the medical condition.

28. The method of claim 27, wherein the at least one brain region associated with the medical condition comprises a first brain region associated with the medical condition and a second brain region associated with the medical condition.

29. The method of claim 28, wherein the first subset of the brain comprises a first parcel of the brain within the first brain region and a second parcel of the brain within the second brain region, and wherein the first dataset comprises rsfMRI data indicating simultaneous activation of the first parcel of the brain and the second parcel of the brain.

30. The method of claim 28, wherein the set of fiber tracts comprises fiber tracts that connect the first parcel of the brain to the second parcel of the brain.

31. The method of claim 30, wherein the medical condition is major depressive disorder, wherein the first brain region is a dorsolateral prefrontal cortex (DLPFC) and the second brain region is an inferior parietal lobule (IPL), and wherein the fiber tracts that connect the first parcel of the brain to the second parcel of the brain comprise superior longitudinal fasciculus (SLF)-II tracts.

32. The method of claim 30,wherein the medical condition is associated with the salience network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region,wherein the first brain region is a ventromedial prefrontal cortex (vmPFC), the second brain region is an anterior insula (Al), and the third brain region is an anterior cingulate cortex (ACC), andwherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or uncinate fasciculus (UF) tracts.

33. The method of claim 30,wherein the medical condition is associated with the default mode network, the at least one brain region associated with the medical condition further comprises a third brain region associated with the medical condition and the first subset of the brain comprises a third parcel of the brain within the third brain region, wherein the first brain region is a medial prefrontal cortex (mPFC), the second brain region is a posterior cingulate cortex (PCC), and the third brain region is an angular gyrus, andwherein the fiber tracts comprise one or more of cingulum bundle (CB) tracts or superior longitudinal fasciculus (SLF)-I tracts.

34. The method of claim 27, wherein the operations further comprise determining a third subset of the brain based on a treatment type of the treatment and on the third dataset, and wherein selecting the target location for the treatment of the medical condition is further based on the third subset of the brain.

35. The method of claim 34, wherein the treatment type of the treatment is a transcranial magnetic stimulation (TMS), and wherein the third subset comprises one or more gyri of the brain.

36. The method of claim 27, wherein the treatment type of the treatment is a transcranial ultrasound stimulation (TUS).

37. The method of claim 27, wherein the treatment type of the treatment is a deep brain stimulation (DBS) via an implantable pulse generator (IPG), and wherein the target location is a target location for implantation of the IPG.

38. The method of claim 27, wherein the operations further comprise applying the treatment to the target location to treat the medical condition.

39. The method of claim 27, wherein the anatomical MRI comprises at least one of T1 -weighted (T1w) MRI or T2-weighted (T2w) MRI.

Images