Brain map portal for alcohol use, obesity, temporal LOBE epilepsy, and alzheimer's

Abstract

The present invention provides systems and methods for treating a subject for a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes including comparing the imaged brain activity to a database of virtual representations of brain networks, wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm, wherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and making one or more determinations based on the results of the comparison, including where to place a treatment device.

Description

Docket: 92033-0048BRAIN MAP PORTAL FOR ALCOHOL USE, OBESITY, TEMPORAL LOBE EPILEPSY,AND ALZHEIMER’SCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This invention claims priority and the benefit of U. S. Provisional Patent Application Serial No. 63 / 738,222, filed December 23, 2024. and entitled “BRAIN MAP PORTAL FOR ALCOHOL USE, OBESITY, AND TEMPORAL LOBE EPILEPSY,” the entire contents of which are hereby incorporated by reference.FIELD OF INVENTION

[0002] The present invention generally relates to systems, methods and program products for producing, identifying and utilizing neuroimaging biomarkers that quantify brain network properties (connectomics) for the purpose of assessing alterations in network-based structure and function in neurological or psychiatric disorders, including obesity, alcohol use disorder, epilepsy, or Alzheimer’s.BACKGROUND OF THE INVENTION

[0003] Throughout this application, various publications are referenced, including referenced in parenthesis. The disclosures of all publications mentioned in this application in their entireties are hereby incorporated by reference into this application in order to provide additional description of the art to which this invention pertains and of the features in the art which can be employed with this invention.

[0004] Neural networks derived from imaging data (connectomes) have revolutionized the field of neuroscience over the past 25 years, providing computationally tractable constructs for describing the brain’s structure and function. Numerous data-driven approaches have been developed to define these networks, but all aim to achieve the same fundamental goal of dimensionality reduction.

[0005] Independent component analysis (ICA) - a special case of blind source separation (Correa et al., 2007) - is a signal processing method, often used for dimensionality reduction, that can detect networks as multivariate co-occurrence patterns across a volumetric time course. ICA has been shown to reliably isolate signals from classical “canonical” networks (e.g. sensorimotor, visual, auditory, central executive etc.) - useful for denoising / removing artifacts in fMRI - but is less effective at delineating behavior- or disease- specific connectomes14929-9580-4034v.1because these subtle patterns are obfuscated by the brain’s robust physiological “terrain’’ / intrinsic architecture (Beckmann et al., 2004; Smith et al., 2009).

[0006] Current state-of-the-art network modeling strategies that analyze primary neuroimaging data (using any network modeling strategy not just independent component analysis (“ICA ”) are often limited by a lack of generalizability. Inter-individual variability and practical cost-related limitations on sample size prevent these studies from identifying a robust mean coherent network structure that holds true when applied to individual patients in different clinical settings / at different imaging centers. This is, in part, also because the signals that are being targeted (particularly when modeling disease effects) are small relative to (and obfuscated by) the intrinsic architecture of (canonical networks that comprise) the human brain. Subtle context-specific patterns are obfuscated by the brain’s intrinsic architecture; canonical network signals must be removed in order to detect the more subtle patterns that govern individual behaviors or disease states.

[0007] Coordinate-based meta-analysis (CBMA) transcends the limitations of ICA by leveraging a sparse data format derived from published task-control or case-control contrasts, wherein the physiological terrain and background noise have been removed. However, such meta-connectomic methods are computationally expensive (often impractical or impossible without a computer cluster) and require expertly curated datasets to ensure compliance with statistical assumptions. Consequently, only a few M-ICA meta-analyses have been reported, all of which focused on replicating canonical architecture from trans-paradigm or trans-diagnostic datasets (Smith et al., 2009; Vanasse et al., 2018; 2022). Although canonical networks do collectively govern neurophysiology (and neuropathology, vis-a-vis the Network Degeneration Hypothesis; Seeley et al., 2009), their main limitation is that they lack a 1-to-1 mapping with any cognitive task or disease state (Vanasse et al., 2021). Canonical descriptions, though useful, are insufficient characterizations of brain states. Until now, meta-analytic ICA has never been applied to context-specific datasets to define networks governing individual behaviors or diseases.

[0008] Prior publications have described ICA (McKeown et al., 1998; Beckmann & Smith 2004) as well as its CBMA implementation for blind source separation (Smith et al., 2009; Vanasse et al., 2021). Briefly, the meta-analytic implementation of ICA extracts latent network structure by identifying coordinate co-occurrence patterns within and across published imaging experiments. To accomplish this, reported coordinates are modeled as spatial probability distributions, generating a 3D modeled alteration (MA) map for each experiment in a meta- 24929-9580-4034v.1analysis (Fox et al., 2014). These MA maps are treated as pseudo-time points (volumes) in a 4D dataset - the input to ICA. This series (X) is then decomposed into a pseudo-time series matrix (E. defining voxel-wise experiment dynamics) and spatial component map matrix S, containing one row per map,, such that X = ES. In doing so, the ICA algorithm attempts to explain as much variance in X as possible while optimizing spatial independence between d signal sources (potential networks).

[0009] Meta-analytic methods applied in CBMA subtract out the brain’s intrinsic architecture, reveal subtle changes in brain structure and function. However, most CBMA-based modeling methods only identify a single network structure and cannot delineate subnetworks. Most techniques for delineating sub-networks lack established guidelines for determining how many sub-networks truly exist in a particular brain state.

[0010] ICA is capable of delineating different network structures. However, state-of-the-art applications of ICA in primary data are relegated to de-noising data or extracting canonical network structures (i.e. the intrinsic architecture, mentioned above). Meta-analytic applications of ICA exist but are few and far between due to the fact that they require substantial computational resources; additionally, despite leveraging CBMA-subtraction methods, all meta-analytic applications of ICA have only detected canonical network structures. This is because the only existing meta-analytic ICA studies have examined trans-paradigm and transdiagnostic datasets (from which canonical network patterns reemerge, despite CBMA-subtraction). Context-specific applications of meta-analytic ICA do not exist and are non-obvious because (1) automated dimensionality estimation algorithms conventionally applied for ICA consistently estimate large numbers of component networks, far exceeding reasonable estimations for intrinsic dimensionality of specific brain states, (2) low-dimensionality estimation methods are non-obvious, (3) computational requirements have been prohibitive (for our low-dimensionality estimation approach, as well as meta-analytic ICA in general), prior to the recent advent of the BrainMap HPC Portal (patent pending), a high performance computing environment for developing neural network models as biomarkers, and most importantly, (4) no theoretical basis has been considered / asserted for developing clinically relevant neuroimaging biomarkers in this manner (via context-specific, low-dimensional applications of meta-analytic ICA).

[0011] As such, there is a need for systems and methods that can discover the neural network or networks that govern specific physiologic or pathologic brain states and are34929-9580-4034v.1applicable to individual patients in a clinical setting, and therefore can be used as a neuroimaging biomarker.BRIEF SUMMARY OF THE INVENTION

[0012] The systems and methods of the present invention can overcome challenges currently facing the field. The systems and methods can discover the neural network or networks that govern specific physiologic or pathologic brain states and are applicable to individual patients in a clinical setting, and ideally can be used as neuroimaging biomarkers.

[0013] The present disclosure includes a novel approach to neuroimaging biomarker discovery which overcomes the challenges currently facing the field, providing a high-powered computing system and employing methods to lead to the discovery and validation of networkbased (connectomic) brain models of normal brain function and of neurologic, psychiatric, developmental and system disorders. The technology encompasses analyzing neuroimaging studies using whole-brain contrasts (task vs control or case vs control) of spatially normalized statistical parametric images to yield location coordinates (x-y-z addresses) of effect loci in Cartesian space referenced to a neuroanatomical atlas. This format has exceptionally high information content and readily allows coordinate-based meta-analyses (CBMA) with either univariate (for effect spatial distributions) or multivariate (for network architecture) statistical methods.

[0014] More generally, the present invention relates to the use of statistical parametric images (SPI) of the human brain. SPI data typically has been 1) acquired over the whole brain; 2) analyzed in a voxel-wise manner as deviations from a null distribution; 3) and, reported as x-y-z locations of local maxima (foci) in a 3D space referenced to a neuroanatomical atlas within a Cartesian coordinate system, a standard first developed by Dr. Peter Fox. These measurements are often transformed into ‘standard space’ from ‘real space’ and were developed to allow scientists to more easily reproduce published experimental results. Coordinate-based meta-analysis (CBMA) has been performed on SPI data. This method improves upon previous methods via low-dimensional applications of meta-analytic ICA.

[0015] In embodiments, the techniques described herein relate to a method of treating a subject for a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes including: (a) imaging brain activity of said subject; (b) comparing the imaged brain activity to a database of virtual representations of brain networks associated with a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated 44929-9580-4034v.1changes, each network including a morphologic property and a physiologic property and, optionally, a temporal property, wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information including: (i) coordinate-based quantified brain activity data related to specific human subject tasks and stored in a task-activation database (TA DB); (ii) voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); and (iii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); and wherein the CBMA algorithm includes a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and (c) determining, based on the comparison, if the subject exhibits a brain network associated with a specific neurologic disorder, a specific psychiatric disorder, a specific development disorder, or specific age-associated changes, (d) wherein if the subject does exhibit said brain network associated with the specific neurologic disorder, the specific psychiatric disorder, the specific development disorder, or specific age-associated changes, administering a treatment to said subject for said neurologic disorder, psychiatric disorder, development disorder or age-associated changes, wherein the treatment includes a treatment device which has a spatial placement component, and the spatial placement component is determined by the locations in the imaged brain activity which correspond most strongly to said brain network.

[0016] In embodiments, the techniques described herein relate to a method, wherein the treatment includes transcranial magnetic brain stimulation, direct current brain stimulation, alternating cunent brain stimulation, deep brain stimulation, and / or focused ultrasound brain stimulation.

[0017] In embodiments, the techniques described herein relate to a method, wherein the imaging brain activity of said subject includes MRI, optionally fMRI.

[0018] In embodiments, the techniques described herein relate to a method, wherein the human brain network is associated with a neurologic disorder.

[0019] In embodiments, the techniques described herein relate to a method, wherein the human brain network is associated with a psychiatric disorder.

[0020] In embodiments, the techniques described herein relate to a method, wherein the neurologic disorder is epilepsy.54929-9580-4034v.1

[0021] In embodiments, the techniques described herein relate to a method, wherein the neurologic disorder is medial temporal lobe epilepsy.

[0022] In embodiments, the techniques described herein relate to a method, wherein the neurologic disorder is Alzheimer's.

[0023] In embodiments, the techniques described herein relate to a method, wherein the neurologic disorder is Tourette's syndrome.

[0024] In embodiments, the techniques described herein relate to a method, wherein the psychiatric disorder is an addiction disorder.

[0025] In embodiments, the techniques described herein relate to a method, wherein the psychiatric disorder is methamphetamine addiction.

[0026] In embodiments, the techniques described herein relate to a method, wherein the psychiatric disorder is alcohol use disorder.

[0027] In embodiments, the techniques described herein relate to a method, wherein the psychiatric disorder is obsessive compulsive disorder.

[0028] In embodiments, the techniques described herein relate to a method, wherein the psychiatric disorder is obesity.

[0029] In embodiments, the techniques described herein relate to a method, wherein the virtual representations of brain networks are low dimensional.

[0030] In embodiments, the techniques described herein relate to a method, wherein the virtual representations of brain networks are of a dimensionality less than 20, less than 10, less than 7, or less than 5.

[0031] In embodiments, the techniques described herein relate to a method, wherein virtual representations of brain networks do not contain sub-networks that are one or more of parcellation, noise, and / or canonical networks.

[0032] In embodiments, the techniques described herein relate to a method, wherein the virtual representations of brain networks are included of brain sub-networks that are not parcellated networks, noise networks, and canonical networks.

[0033] In embodiments, the techniques described herein relate to a method, wherein the virtual representations of brain networks are generated without separately identifying nodes and edges including the virtual representations of brain networks.64929-9580-4034v.1

[0034] In embodiments, the techniques described herein relate to a method, wherein the virtual representations of brain networks are derived from cross-sectional data, and are congruent with longitudinal network dynamics.

[0035] In embodiments, the techniques described herein relate to a method, wherein the Low-d M-ICA algorithm is incrementally applied for different dimensions.

[0036] In embodiments, the techniques described herein relate to a method, wherein the Low-d M-ICA algorithm contains stopping conditions including one or more of: three consecutive dimensions produce a sub-network parcellation, noise component, and / or a canonical network; and / or insufficient statistical power (d > n / 23), where d is dimension and n is the number of experiments in the meta-analysis.

[0037] In embodiments, the techniques described herein relate to a method, wherein the information associated with the neurologic disorder or the psychiatric disorder is modeled as a plurality of Gaussian probability distributions with full-width half-maximum values that are irrespective of the amount of information.

[0038] In embodiments, the techniques described herein relate to a method for generating a virtual representation of a human brain network associated with a selected neurologic disorder or psychiatric disorder, the human brain network including a morphologic property’ and a physiologic property and, optionally, a temporal property, the method including: (a) selecting, by one or more computers of a high-performance computing portal, a neurologic disorder or psychiatric disorder; (b) accessing, by the one or more computers, a plurality of databases storing group-averaged coordinate-based spatially-normalized data populated from a plurality of human brain subject, each stored on one or more computer-readable media, including: (i) a task-activation database (TA DB) including coordinate-based quantified brain activity data related to specific human subject tasks; (ii) a voxel-based morphometry- database (VBM DB) including voxel-based morphometric data; and (iii) a voxel-based physiology database (VBP DB) including voxel-based physiological data; (c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected neurologic disorder or psychiatric disorder; (d) applying, by the one or more computers, a multivariate coordinate-based meta-analysis (CBMA) algorithm, including a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network including a morphologic property and a physiologic property and,74929-9580-4034v.1optionally, a temporal property associated with the selected neurologic disorder or psychiatric disorder, and generating a virtual representation of the human brain network.

[0039] In embodiments, the techniques described herein relate to a method, wherein the virtual representation of the human brain network is used as priors in machine learning, artificial intelligence, or statistical models.

[0040] In embodiments, the techniques described herein relate to a method, wherein the human brain network is associated with a neurologic disorder.

[0041] In embodiments, the techniques described herein relate to a method, wherein the human brain network is associated with a psychiatric disorder.

[0042] In embodiments, the techniques described herein relate to a method, wherein the presenting step (E) includes transmitting the virtual representation of said human brain network to a user device of the user.

[0043] In embodiments, the techniques described herein relate to a method, wherein the presenting step (E) includes causing the virtual representation of said human brain network to be displayed on a user device of the user.

[0044] In embodiments, the techniques described herein relate to a method for generating a virtual representation of a human brain network associated with a selected developmental disorder, systemic disorder, or age-associated change, the human brain network including a morphologic property and a physiologic property and, optionally, a temporal property, the method including: (a) obtaining, by one or more computers of a high-performance computing portal, the selected developmental disorder or systemic disorder; (b) accessing, by the one or more computers, a plurality of databases storing group-averaged coordinate-based spatially-normalized data populated from a plurality of human brain subject, each stored on one or more computer-readable media, including: (i) a task-activation database (TA DB) including coordinate-based quantified brain activity data related to specific human subject tasks; (ii) a voxel-based morphometry database (VBM DB) including voxel-based morphometric data; and (iii) a voxel-based physiology database (VBP DB) including voxel-based physiological data; (c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected developmental disorder, systemic disorder, or age-associated changes; (d) applying, by the one or more computers, a multivariate coordinate-based meta-analysis (CBMA) algorithm, including a Low-dimensionality Meta-analytic Independent Component Analysis 84929-9580-4034v.1(Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network including a morphologic property and a physiologic property and, optionally, a temporal property associated with the selected developmental disorder, systemic disorder, or age-associated changes, and generating a virtual representation of the human brain network; and (e) presenting, by the one or more computers, to a user the virtual representation of said human brain network.

[0045] In embodiments, the techniques described herein relate to a system including: one or more computers of a high-performance computing portal including memory wherein the memory stores computer readable instructions including program code that, when executed, cause the one or more computers to perform the steps of: (a) obtaining, by the one or more computers of the high-performance computing portal, a selected neurologic disorder or psychiatric disorder; (b) accessing, by the one or more computers, a plurality of databases storing group-averaged coordinate-based spatially-normalized data populated from a plurality of human brain subjects, each stored on one or more computer-readable media, including: (i) a task-activation database (TA DB) including coordinate-based quantified brain activity data related to specific human subject tasks; (ii) a voxel-based morphometry database (VBM DB) including voxel-based morphometric data; and (iii) a voxel-based physiology database (VBP DB) including voxel-based physiological data; (c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected neurologic disorder or psychiatric disorder; (d) applying, by the one or more computers, a multivariate coordinate-based metaanalysis (CBMA) algorithm, including a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network including a morphologic property and a physiologic property and, optionally, a temporal property associated with the selected neurologic disorder or psychiatric disorder, and generating a virtual representation of the human brain network; and (e) presenting, by the one or more computers, to a user the virtual representation of said human brain network.

[0046] In embodiments, the techniques described herein relate to a system, wherein the human brain network is associated with a neurologic disorder.

[0047] In embodiments, the techniques described herein relate to a system, wherein the human brain network is associated with a psychiatric disorder.94929-9580-4034v.1

[0048] In embodiments, the techniques described herein relate to a system, wherein the presenting step (E) includes transmitting the virtual representation of said human brain network to a user device of the user.

[0049] In embodiments, the techniques described herein relate to a method for delivery of transcranial magnetic stimulation (TMS) to a subject with a neurologic or psychiatric disorder, including: (a) obtaining a virtual representation of a brain network associated with the neurologic disorder or a psychiatric disorder, wherein the virtual representation was generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information including: (i) coordinate-based quantified brain activity data related to specific human subject tasks stored in a task-activation database (TA DB); (ii) voxel-based morphometric data stored in a voxelbased morphometry database (VBM DB); and (iii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); and wherein the CBMA algorithm includes a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; (b) identifying, based at least in part on a connectivity-based parcellation derived from the virtual representation of a brain network, a first upstream target brain region; (c) computing, in at least one computing system, a region-seeded connectivity map of the first upstream target brain region of the subject based at least in part on the virtual representation of a brain network; (d) identifying at least one downstream projection region of the subject associated with the first upstream target brain region of the subject; (e) computing, in the at least one computing system, a region-seeded connectivity map of the at least one downstream projection region of the subject; (f) identifying a first upstream projection region of the subject that demonstrates functional connectivity to the at least one downstream projection region of the subject; (g) determining positioning for a TMS coil with respect to the first upstream projection region of the subject in accordance with a cortical column cosine aiming principle; and (h) causing delivery of the TMS to the subject via the TMS coil according to the determined positioning.BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Embodiments of the present invention will be described with references to the below referenced accompanying figures.

[0051] Fig. 1 depicts a database content and meta-analysis environment in accordance with embodiments of the present invention.104929-9580-4034v.1

[0052] Fig. 2 depicts a high performance computing portal in accordance with embodiments of the present invention.

[0053] Fig. 3A depicts a method in accordance with embodiments of the present invention.

[0054] Fig. 3B depicts a web application in accordance with embodiments of the present invention.

[0055] Fig. 3C depicts low dimensionality estimation guidelines, in accordance with embodiments of the present invention.

[0056] Fig. 4 depicts dimensionality estimation for emotive behavior, in accordance with embodiments of the present invention.

[0057] Figs. 5A-5E depict network interpretation of results for emotive behavior obtained in accordance with embodiments of the present invention.

[0058] Fig. 6 depicts dimensionality estimation for mesial temporal lobe epilepsy pathology in accordance with embodiments of the present invention.

[0059] Fig. 7 depicts behavioral associations of MTLE network regions generated in accordance with embodiments of the present invention compared with behavioral associations of regions that were identified via simple spatial convergence across studies in a conventional activation likelihood estimation (ALE) meta-analysis of the same set of MTLE experiments.

[0060] Figs. 8A-8B depict meta-analytic connectivity models constructed for both MTLE networks identified by M-ICA.

[0061] Figs. 9A-9C depict Mango toolbox analyses of two MTLE networks and a noise component in accordance with embodiments of the present invention.

[0062] Fig. 10 depicts a comparison of two MTLE Meta-ICA networks generated in accordance with embodiments of the present invention compared with networks generated by ALE.

[0063] Fig. 11 depicts a method in accordance with embodiments of the present invention.

[0064] Fig. 12A depicts changes in phosphatidylethanol from the end of IR-TMS treatment to 1 month in accordance with embodiments of the present invention.

[0065] Fig. 12B depicts changes in phosphatidylethanol from the end of IR-TMS treatment to 6 months in accordance with embodiments of the present invention.114929-9580-4034v.1

[0066] Figs. 13A-13B depict node-and-edge network representations of Alzheimer’s disease (AD)-related atrophy (VBM) and hypometabolic (VBP) changes generated in accordance with embodiments of the present invention.

[0067] Fig. 14 depicts A meta-analytic statistical parametric map of gray-matter atrophy in obesity (vs. lean controls) in accordance with embodiments of the present invention.

[0068] Fig. 15 depicts Atrophy Functional Network modeling of network degeneration in obesity in accordance with embodiments of the present invention.

[0069] Fig. 16 depicts a personalized TMS treatment plan in accordance with embodiments of the present invention.

[0070] Fig. 17 depicts a dimensionality estimation for obesity', in accordance with embodiments of the present invention.

[0071] Fig. 18 depicts flow diagram of the study selection process for addiction VBM studies in accordance with embodiments of the present invention.

[0072] Fig. 19 depicts independent SUD and BA ALE maps and overlap region in accordance with embodiments of the present invention.

[0073] Fig. 20 depicts pooled SUD + BA ALE map and interactive regions in accordance with embodiments of the present invention.

[0074] Figs. 21 A-C depicts results of whole image paradigm class analysis, in accordance with embodiments of the present invention.

[0075] Figs. 22A-B depicts results of SUD cluster specific paradigm class analysis in accordance with embodiments of the present invention.

[0076] Figs. 23A-F depicts results of pooled SUD + BA cluster specific paradigm class analysis.

[0077] Fig. 24 depicts a dimensionality estimation for Alzheimer’s disease I mild cognitive decline, in accordance with embodiments of the present invention.

[0078] Fig. 25 depicts a dimensionality estimation for alcohol use disorder, in accordance with embodiments of the present invention.

[0079] Fig. 26 depicts a dimensionality estimation for substance use disorder / behavioral addiction, in accordance with embodiments of the present invention.124929-9580-4034v.1

[0080] Figs. 27A-B depict a rich co-alteration model of MTLE.

[0081] Figs. 28A-H show complex patterns in MTLE pathology' reveal a unique disease network (A) that exhibits modular architecture (B. C, F) congruent with physiologic behaviors (D. G) often affected during the ictal or peri-ictal period (E. H).

[0082] Figs. 29A-C show that the MDN Thalamus and Medial Frontal Gyms are the most influential extra-hippocampal nodes in the MTLE Co- Alteration Network.DETAILED DESCRIPTION OF THE INVENTIONGeneral

[0083] Where a numerical range is provided herein, it is understood that all numerical subsets of that range, and all the individual integers contained therein, are provided as embodiments of the invention. For example, 30 to 70 includes the subset of 30 to 35, the subset of 40-60, etc. as well as every' individual integer value, e.g. 30, 31, 32, 33, and so on.

[0084] “And / or” as used herein, for example with option A and / or option B, encompasses the separate embodiments of (i) option A, (ii) option B, and (iii) option A plus option B.

[0085] All combinations of the various elements described herein are within the scope of the invention unless otherwise indicated herein or otherwise clearly contradicted by context.

[0086] One skilled in the art will readily appreciate that the specific methods and results discussed herein are merely illustrative of embodiments of the invention as described more fully in the claims that follow thereafter.Overview of system and methods

[0087] Fig. 11 depicts a method in accordance with embodiments of the present invention. As shown in Fig. 11, methods of the present invention include imaging brain activity of a subject; comparing the imaged brain activity to a database of virtual representations of brain networks, wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm, wherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and making one or more determinations based on the results of the comparison. Low-d M-ICA may alternatively be referred to as M-ICA.

[0088] Fig. 1 depicts a database content and meta-analysis environment in accordance with embodiments of the present invention. As shown in Fig. 1, the BrainMap Project curates134929-9580-4034v.1published neuroimaging literature, catalogues coordinate-data and meta-data, and distributes software, in support of performing rigorous coordinate-based meta-analyses (CBMA) via activation likelihood estimation (ALE). The three Sectors of the BrainMap Database were sequentially developed to house task-activation (Functional Sector), structural (VBM Sector), and resting-state functional (VBP Sector) neuroimaging studies compliant with the statistical assumptions and requirements of CBMA. BrainMap Software include Scribe, Sleuth, and GingerALE, for coding new studies, accessing database content, and meta-analyzing these data, respectively. The BrainMap Database and Software are all made available for use on the BrainMap Community Portal.

[0089] Fig. 2 depicts a high performance computing portal in accordance with embodiments of the present invention.

[0090] In embodiments, manual portions of the pipeline include the initial database search and M-ICA output interpretations. In embodiments, these portions can be automated. In embodiments, these steps can be performed through GUI applications. In embodiments, the GUI applications are made available on the BrainMap Community Portal. In embodiments, the GUI applications are locally through downloadable Java applications distributed on the BrainMap website (brainmap.org).

[0091] Fig. 3A depicts a method in accordance with embodiments of the present invention. As shown in Fig. 2A, the M-ICA pipeline extracts latent network structure from patterns in published coordinate data by applying independent component analysis to modeled neuroimaging experiments in the BrainMap Database.

[0092] Fig. 3B depicts a web application in accordance with embodiments of the present invention. As shown in Fig. 2B, the M-ICA web application provides user-friendly interface on the BrainMap Community Portal that combines the M-ICA pipeline with cloud computing resources at the Texas Advanced Computing Center.

[0093] Fig. 3C depicts low dimensionality estimation guidelines in accordance with embodiments of the present invention. As shown in Fig. 2C, the present invention provides low dimensionality estimation guidelines for defining the network composition of specific physiologic and pathologic brain states. M-ICA is applied to a coordinate-based dataset to extract an incremental number of components, beginning at a dimensionality7(d) of 1. Distinct components are progressively identified at each value of d until three consecutive “non-new netw orks" are identified. Non-new networks may be 1) parcellations of a network identified at144929-9580-4034v.1a prior value of d, 2) noise, appearing as punctate gaussians in space and signifying incoherent non-network structure, or 3) canonical network structures which are expected to emerge when all context-specific network structures have been detected.BrainMap HPC Portal

[0094] In embodiments of the systems described herein, an example is referred to as BrainMap HPC Portal (“BrainMap High Performance Computing Portal’'). BrainMap HPC Portal is a cloud-based environment designed to promote the development of neuroimaging biomarkers that quantify brain network properties (i.e., connectomics) for the purpose of assessing alterations in network-based structure and function in a wide variety of health-related conditions including neurologic and psychiatric disorders, and normal changes across the lifespan including development and aging.

[0095] BrainMap HPC Portal can achieve its goals via meta-analytic modeling of data gleaned from the coordinate-reporting neuroimaging literature, a large and well-standardized corpus. In embodiments, data for meta-analytic modeling are provided by three BrainMap Databases (DBs): Task-activation database (TA DB), voxel-based morphometry database (VBM DB), and voxel-based physiology' database (VBP DB). The BrainMap DBs are steadily expanding, being supplemented by BrainMap team efforts and community’ contributions. All data are curated by the BrainMap team, including selection, coordination and arrangement. The BrainMap DBs are copyrighted by the Board of Regents, University of Texas System.

[0096] In embodiments, each Database may be configured to store data from publication based upon a structured standardized coding scheme. Data may be sourced from published human neuroimaging experimental results using the coding scheme.

[0097] Meta-analytic software tools are customized instantiations of open-source software applications or software tools developed by the BrainMap team. For most applications, assistance with implementation was provided by the developers.

[0098] The BrainMap HPC Portal is a customized instance of the Texas Advanced Computing Center (“TACC”) “Core Experience Portal” (CEP) framework. The CEP framework is a cloneable “plain vanilla” portal developed by TACC as a starting point for portal-project development. The CEP codebase is open-source, but in this instance has been extensively customized for BrainMap HPC. Unique to the BrainMap Community’ Portal are several enhancements made over the Core Experience Portal. This includes (1) The integration of the taxonomy viewer into the front page, (2) integration of a NIFTI file format viewer into154929-9580-4034v.1the data management system for easier data visualization, (3) approximately a dozen workflows for (" Coordinate Based Meta- Analysis) CBMA and file visualization, and (4) brandings, logos, citations, links, and fair use text specific to the BrainMap project.

[0099] The backend of the portal relies heavily on Tapis to provide all HPC integrations and data management capabilities. Tapis is an open source framework which was developed at TACC. The Tapis API also provides a fully functional command line interface to the assets in the BrainMap Community Portal (as a programmatic alternative to the point-and-click web interface).

[0100] The systems provided herein in embodiments provide a complete, fully contained environment for multivariate meta-analytic modeling of human neural systems physiology and pathophysiology. For example, the BrainMap Portal environment contains: 1) extensive (and extensible) data resources in the form of BrainMap records, each representing peer-reviewed publications reporting data adhering to the strict quality standards of coordinate-based metaanalysis; 2) extensive (and extensible) analytic resources in the form of mass-univariate and multivariate statistical-analysis pipelines implemented in formats optimized for CBMA; and 3) high-performance and high-throughput computing (HPC, HTC).

[0101] In embodiments, The BrainMap suite of established Java CBMA applications have been implemented with the portal. Multiple CBMA model-construction applications have been implemented as containerized python pipelines. Primary data (fMRI, sMRI) model-validations pipelines are optionally included.

[0102] In embodiments, systems and methods include a high performance computing portal, as shown in Fig. 2.

[0103] In embodiments, the portal Brainmap HPC architecture interfaces and integrates Topware, Midware, and Deepware. In embodiments, Topware provides a user interface to portal resources. In embodiments, Midware bridges the User Interface with HPC Deepware. In embodiments, Deepware supports resources accessed by but not edited or directly called by users. This architecture provides comprehensive, sophisticated, network-model development and validation capabilities within user-friendly, secure, cloud-computing environment.

[0104] In embodiments, the high performance computing portal is three-tiered. In embodiments, in Topware (1st tier), connectome researchers can access BrainMap HPC via a w eb-based interface. A point-and-click menu provides access to a suite of tools for accessing BrainMap data by SQL query (e.g., Sleuth - software designed to search the BrainMap 164929-9580-4034v.1databases, create workspace data sets, plots and export the subjects and locations for metaanalysis) and constructing command-line jobs to perform a variety of mass univariate and multivariate CBMA workflows (FSL MELODIC, MACM - meta-analytic connective modeling, CBP - coactivation-based parcellation, GTM - graph theory modelling). In embodiments, tools for image-data visualization (Mango, Papaya, FSL View), metadata-informed interpretation, and network-model visualization can be launched to view and analyze job outcomes in place (using Model Interface). Functionality is also provided to facilitate uploading User Files, such as structural MRI (sMRI) and functional MRI (fMRI). In embodiments, a purpose of primary data upload is out-of-sample validation of meta-analytic network models, which is done by assessing model goodness-of-fit, model-based discrimination of patients from controls, and other metrics. In embodiments, download of network models, validation data and other user-created intellectual property is also supported. In embodiments, the web-based interface may be an instance of TACC's open-source " Core Experience Portal" code base, which has been extensively customized for BrainMap HPC.

[0105] In embodiments, midware (2nd tier) may use Tapis (a framework providing a hosted, unified web-based API for securely managing computational workload) to provide its HPC integration and data-management capabilities. In embodiments, Tapis provides user authentication, job management, user notification, and load balancing. In embodiments. Tapis provides routines for association with the BrainMap web portal ("tenancy"), data-use request and data-access management and monitoring functions for the BrainMap SQL DBs, and application definitions for CBMA and primary data pipelines. Data provenance and job histories are tracked in Tapis log files. In embodiments, the Tapis API also provides a fully functional command-line interface (CLI) to all the assets of the BrainMap HPC portal, as an alternative to the point-and-click interface for more experienced users. The BrainMap HPC CLI enables advanced users to script and automate tasks, run many jobs simultaneously, and augment the portal offerings by adding new applications and workflows.

[0106] In embodiments, Deepware (3rdtier) may be made of assets accessed by users only via the Tapis midware level and may consist of hardware and software. These include TACC high performance computers (where the CBMA applications run) and virtual machines supporting various functions (e.g., the landing page and taxonomy server, the three BrainMap SQL DBs, and precomputed images and co-occurrence matrices supporting MACM, CBP and other applications). In embodiments, the MACM. CBP and other applications may consist of174929-9580-4034v.1modules configured to perform certain algorithms or calculations running on the high performance computers.

[0107] In embodiments, developers and system administrators (e.g., those at TACC) may have access to all levels of the architecture. In embodiments, other developers (e.g., Brainmap developers) manage BrainMap software (Sleuth, Mango, Papaya) and data structures (BrainMap SQL databases, co-occurrence matrices). Community developers can add new analysis applications, in cooperation with TACC and BrainMap developers. Community developers can build new analysis pipelines independently, which can be made openly accessed by TACC developers, for example by using Developer Tools.Imaging brain activity

[0108] In embodiments, the imaging brain activity of said subject may include MRI. In embodiments, the imaging brain activity of said subject may include rs-fMRI.Neurologic disorders and psychiatric disorders

[0109] In embodiments, the human brain network may be associated with a neurologic disorder, and / or with a psychiatric disorder. In embodiments, the human brain network may be associated with one or more neurologic disorders, and / or psychiatric disorders.

[0110] In embodiments, the neurologic disorder is epilepsy. In embodiments, the neurologic disorder is medial -temporal lobe epilepsy.

[0111] In embodiments, the neurologic disorder is Alzheimer’s.

[0112] In embodiments, the psychiatric disorder is addiction. In embodiments, the psychiatric disorder is alcohol use disorder. In embodiments, the psychiatric disorder is obesity'. Obesity' can be understood as a psychiatric disorder as it is addiction to food.

[0113] In embodiments, the human brain network may be associated with a developmental disorder, with a systemic disorder, and / or with age-associated changesDatabases

[0114] In embodiments, systems and methods in accordance with the present invention include a plurality of databases. In embodiments, the plurality- of databases includes group-averaged coordinate-based spatially-normalized data. In embodiments, the data is populated from a plurality of human brain subjects.

[0115] In embodiments, the plurality of databases includes one or more of: a task-activation database (TA DB) which includes coordinate-based quantified brain activity data related to 184929-9580-4034v.1specific human subject tasks; a voxel-based morphometry database (VBM DB) comprising voxel-based morphometric data; and a voxel-based physiolog)’ database (VBP DB) comprising voxel-based physiological data, to name a few.

[0116] Examples of a task-activation database for human brains, voxel-based morphometry database and voxel-based physiology database can be found e.g. at brainmap.org. BrainMap is a database of published task and structural neuroimaging experiments with coordinate-based results (x,y,z) in Talairach or MNI space. BrainMapWeb at apps.rii.uthscsa.edu / bmapWeb / is a web application for searching and retrieving data from the task database. A task-activation database is accessible at portal.brainmap.org.

[0117] Task activation can be. for example cognitive / emotion or sensory perception. Exemplary task activations are listed in Table 1.Table 1.Cognition & Emotion PerceptionAttention HearingLanguage TasteMemory InteroceptionMusic SmellNegative Emotion TouchPositive Emotion VisionReasoningSocial

[0118] Voxel- based morphometry of the human brain / CNS is established (see, e.g., Whitwell, J Neurosci. 2009 Aug 5; 29(31): 9661-9664) and databases can be created therefrom. Examples include using T1 -weighted volumetric MRI scans and performing statistical tests across all voxels in the image to identify volume differences between groups.

[0119] Voxel-based physiology of the human brain / CNS. determining and quantifying neurologically relevant physiological parameters, is also established. Physiology parameters include cerebral blood volume (CBV), cerebral blood flow (CBF), cerebral metabolic rate of oxygen (CMR02), and oxygen extraction fraction (OEF).

[0120] In embodiments, the Databases have been mirrored within Tapis architecture using the Oracle® Goldengate function.Multivariate coordinate-based meta-analysis (CBMA) algorithm

[0121] In embodiments, systems and methods of the present invention include a multivariate coordinate-based meta-analysis (CBMA) algorithm.194929-9580-4034v.1

[0122] The multivariate CBMA aspect is significant. Multivariate CBMA as used in the methods and systems computes interactions among voxels (or regional groups of voxels), treating each voxel (or region) as a variable (hence, multivariate). In embodiments, each voxel may correspond to a three-dimensional area (e.g., an area bounded by a cube or rectangular prism) of imaged brain tissue. Intervoxel co-occurrence patterns (covariances) serve to identify brain networks. That brain functions arise from multi-regional, inter-connected, neural networks has been presumed for more than a century. However, robust methods for mapping human neural networks in vivo (“connectomics”) have been lacking. Multivariate CBMA overcomes this difficulty by extracting networks as emergent properties (latent variables) from BrainMap data.

[0123] Multivariate CBMA algorithms emerged in the mid-2000 ’s, being invented by BrainMap investigators (reviewed in Fox et al, 2014). These highly sophisticated analytics now include: independent components analysis (1CA; Smith. Fox et al, 2009; Vanasse et al.2018; 2021; Figure 20); meta-analytic connectivity modeling (MACM; Robinson et al., 2010, 2012); graph theory modeling (GTM; Crossley et al., 2013, 2014, 2018; Cauda et al. 2018; Gray et al, 2020); connectivity-based parcellation (CBP; Bzdok et al, 2012; Barron et al, 2014); author-topic modeling (ATM: Yeo et al, 2015; Ngo et al, 2019); and, structural equation modeling (SEM; Chiang, 2020, 2021). Collectively, these reports have repeatedly validated multivariate CBMA by comparison with other methods for connectivity analysis, including resting-state fMRI, diffusion fractography, structural covariance, post-mortem cytoarchitecture, and electrophysiology.

[0124] The systems and methods are important in view of the Network Degeneration Hypothesis (NDH). The NDH postulates that in some dementing disorders, pathophysiology propagates along functional connectivity networks (Seely et al, 2009). NDH was confirmed and vastly extended in scope by BrainMap investigators. Meta-ICA demonstrated a close correspondence between healthy, functional connectivity networks and the structural pathology¬ networks observed in greater than 40 neurologic and psychiatric brain disorders (Vanasse et al, 2018, 2021). Note that this was not low dimensional ICA as disclosed in the present application. Neurological disorders exhibiting network pathology included both dementiainducing disorders, and many others (Table 2). All psychiatric disorders with a sufficient volume of published studies for valid meta-ICA exhibited network degeneration (Table 2). This is unprecedented evidence that structural pathology propagating along functional connectivity networks is a general property of neuropsychiatric disorders.204929-9580-4034v.1

[0125] Informed by their own ground-breaking work, BrainMap investigators have focused their efforts on meta-connectomic modeling of brain disorders. A primary objective is to use meta-connectomics to create network-biomarkers adaptable to per-subject application to rs-fMRI, as a neuroimaging adaptation of personalized medicine. This approach has worked for image-guided targeting of neuromodulation therapy for PTSD (Fox, Lancaster & Salinas, 2022; Fox, Salinas et al.. 2023). A diagnostic biomarker for Multiple Sclerosis (Chiang, et al., 2019) achieved sufficient confirmation in a small, out-of-sample, primary-data validation (Chiang et al. 2021) to support patent applications (Chiang et al. 2020).

[0126] Metaconnectomics (by multivariate CBMA) can identify and validate connectomic biomarkers for ahost of brain disorders (Table 2). Barriers have included prior data-distribution models. Data-transfer volumes required for mass metaconnectomics have been prohibitive also. HPC access is limited and requires optimization to this application. Algorithms are challenging to implement.Table 2 - non-limiting embodiments of neurologic disorders and psychiatric disorders:Neurologic Disorders Psychiatric DisordersAlzheimer’s Disease SchizophreniaMild Cognitive Impairment Bipolar DisorderFrontotemporal Dementia Major Depressive DisorderHuntington’s Disease Posttraumatic Stress DisorderPick’s Disease Generalized Anxiety DisorderCortico-basilar degeneration Obsessive Compulsive DisorderAmyotrophic Lateral Sclerosis Panic DisorderMultiple Sclerosis Substance Use DisordersTemporal Lobe Epilepsy Behavioral AddictionsJuvenil Myoclonic Epilepsy Asperger’s SyndromeChronic Pain Disorder Unspecified PsychosisTraumatic Brain Injury Anorexia NervosaLow-d M-ICA

[0127] In embodiments, systems and methods of the present invention include a multivariate coordinate-based meta-analysis (CBMA) algorithm, which includes a Low-214929-9580-4034v.1dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm. The Low-d M-ICA may also be referred to in this M-ICA.

[0128] In embodiments, Low-d M-ICA combines 1) neuroimaging data (e.g. three curated, filterable BrainMap databases). 2) meta-analytic (CBMA-based) ICA applied to contextspecific data, and 3) a novel, iterative dimensionality estimation algorithm (implementing our guidelines) for delineating subnetworks.

[0129] A key challenge when applying ICA is determining d, the number of network components (signal sources) that should be extracted for a particular dataset. The ICA machine learning algorithm conventionally estimates dimensionality (d) using standard probabilistic model selection methods (e.g. Akaike or Bayesian Information Criteria, Laplace approximation, etc.). These metrics typically suggest d > 20 for most datasets since the penalties associated with overparameterization (loss of parsimony) are often outweighed by the algorithmic objective of explaining variance. Current automated approaches are thus ill-suited for low dimensional applications in M-ICA. since they result in estimates of d that exceed the meaningful network dimensionality of most context-specific brain states. In lieu of an automated approach, the number of components extracted by ICA must be explicitly specified and a new approach to dimensionality estimation is required.

[0130] Our low-< estimation strategy (Fig. 3C) relies on a stepwise approach to blindsource separation, applying M-ICA at incremental values of J (1, 2, 3...) to decompose a given dataset into independent signal sources (i.e. networks). We containerized the meta-analytic ICA workflow for use in a High-Performance Computing environment to ensure the practicality7of this approach (Fig. 3B). At d = 1, a spatial map (i.e. signal source) explaining the most variance will be extracted. As d increases, additional parameters are included to explain variance contributed by the next-most dominant signal sources. Fortunately, although ICA always attempts to explain maximal variance, regions included in each spatial map must be internally consistent with experiment dynamics, reflecting coordinate co-occurrence patterns observed in the literature (Beckmann 2004). As a result, networks contributing to distinct sets of experiments will emerge at separate values of d but may be split (parcellated) into separate components once most variance has been explained.

[0131] Each spatial map produced by M-ICA has four possible interpretations: 1) novel network (Towne et aL, 2022); 2) canonical network (Smith et al., 2009); 3) parcellation of a network (Smith et al., 2009); or 4) noise (Smith et al., 2009). The first spatial maps that appear224929-9580-4034v.1will generally be novel networks, assuming the dataset has been restricted to a specific context. As more of the latter patterns (i.e. non-novel network) appear, dominant novel networks are less likely to be identified. The stopping condition is defined as three non-novel networks identified at consecutive values of d. Optimal “intrinsic” dimensionality is then defined as the last value of d where a novel network was detected..

[0132] Canonical networks may emerge if the literature selection is too broad (e.g. transparadigm or trans-diagnostic input). They may be introduced by a particular paradigm in functional imaging experiments (e.g. visual networks in functional experiments or default mode network in resting-state experiments). Alternatively, they may appear later on after most variance has been explained (i.e. after novel networks have been identified). Once the majority of variance in the dataset has been explained, novel or canonical networks identified at prior values of d will parcel into sub-networks. Noise components, appearing as random gaussians in space, will also emerge at this point or earlier if the input dataset contains coordinate content lacking coherent network structure.

[0133] One final constraint that must be considered is statistical power. The most stringent existing guidelines for conventional CBMAs suggest that meta-analytic maps should each represent a minimum of 23 experiments, to ensure that convergent results are not skewed by an individual experiment (Eickhoff et al., 2016; Frahm et al., 2023). Therefore, we recommend an additional stopping criterion, that d not exceed ir / 23, where n is the number of experiments in the meta- analysis.Neuroimaging data

[0134] Low-d M-ICA, is a coordinate-based meta-analytic pipeline for defining networks and efficiently estimating intrinsic dimensionality of physiologic and pathologic brain states. Low-d M-ICA couples state-of-the-art software with High Performance Computing to analyze neuroimaging data. In embodiments, the neuroimaging data is one or more of task and structural neuroimaging. In embodiments, the neuroimaging data has coordinate-based results (x,y,z). In embodiments, the neuroimaging data is in the Talairach or MNI space. In embodiments, the neuroimaging data is the BrainMap Database.Virtual representations and biomarkers

[0135] Low-d M-ICA enables the rapid and reproducible modeling of hundreds of different behaviors and diseases (neurological and psychiatric). Further, Low-d M-ICA leverages principles of dynamic systems to maximize the translational potential of the network models234929-9580-4034v.1that it identifies. Neuroimaging biomarkers (network structures) derived using Low-d M-ICA can be applied in resting-state functional magnetic resonance imaging (rs-fMRI) per-subject and are thus optimally positioned for development as clinical tools.

[0136] The clinical utility of biomarkers derived using this tool have many uses. Applications for biomarkers produced by Low-d M-ICA include (1) diagnosis; (2) functional imaging metrics for monitoring disease course; (3) preoperative neurosurgical planning; (4) targeted delivery of neuromodulation therapy; (5) quantifying effects of behavior- or diseasemodifying agents; and (6) stratifying patients in clinical trials, to name a few.Theoretical underpinning

[0137] The advancements made by Low-d M-ICA are non-obvious, as they are driven by the first explicit neuroimaging application of ergodicity. By considering neural networks as ergodic dynamic systems, Low-d M-ICA was developed as an approach for accessing network structures and sub-network structures (i.e. identifying biomarkers) that are distinct from canonical network architecture and simultaneously applicable in a clinical setting. Low-d M-ICA will promote establishment of the “ergodicity-inference approach” as the new state-of-the-art for developing neuroimaging biomarkers that can translated to patient care. Within this framework, a novel method for estimating the intrinsic dimensionality of different brain states is presented.

[0138] Ergodicity indicates that group observations of a dynamic system at single time points will characterize the dynamics of a single observation of the system over time. In the brain, this would mandate that network properties derived from cross-sectional data will be observed longitudinally in individuals. The implication for neuroimaging is that meta-analytic sampling can access network architecture (data structures) useful for detecting networks per-subject. Ergodicity has been implicitly shown in healthy subjects by graph theory (Crossley et al., 2013) and other analytics (Smith et al., 2009). Equivalent functional architecture was identified by connectomic meta-analysis of task-based studies and connectomic analysis of temporally concatenated rs-fMRI. Diseases follow collectively similar yet individually distinct patterns (Vanasse et al, 2021), motivating ergodic hypotheses. The Invention, Low-d M-ICA, is the first explicit application of this principle.Smoothing kernel

[0139] In embodiments, M-ICA applies a uniform, standardized Full-Width HalfMaximum (e.g. FWHM = 12mm) gaussian when modeling coordinates. This is unlike any244929-9580-4034v.1other coordinate based meta-analytic (e.g. ALE-based) method, as conventional approaches model spatial uncertainty by applying a FWHM proportional to the sample size of the study that reported the coordinate. For example, in ALE-CBMA, larger studies (higher n) are modeled with tighter Gaussians (smaller FWHM), under the assumption that spatial uncertainty of an effect (published coordinate) decreases with more subjects.

[0140] In M-ICA, using a uniform smoothing kernel is conceptually equivalent to assuming that every coordinate comes from a study with the same sample size (e g. for a 12 mm FWHM, this would assume n = 4). Uniform FWHM is non-intuitive, yet it stabilizes blindsource separation when ICA is applied for CBMA, enabling M-ICA to generate candidate network structures that can be applied clinically.Uses of systems and methods

[0141] The systems and methods provided herein in non-limiting embodiments can be employed to:

[0142] create a computer environment designed for discovery and validation of networkbased (connectomic), brain models of normal brain function and of neurologic, psychiatric, developmental, and systemic disorders;

[0143] create a computer environment to create connectomic models of brain disorders for use as neuroimaging biomarkers, by which is meant quantitative indices of brain physiology or pathophysiology7to be used for diagnosis, prognosis, disease-progression monitoring, or disease risk prediction either clinically or in the context of clinical trials;

[0144] create a computer environment to create connectomic models of brain disorders for use in treatment planning neuromodulation therapy, including: transcranial magnetic brain stimulation; direct current brain stimulation; alternating current brain stimulation; deep brain stimulation; focused ultrasound brain stimulation; or other neuromodulation techniques not yet described or discovered;

[0145] create a computer environment to create connectomic models of brain disorders for use in preoperative planning of neurosurgical interventions;

[0146] create a computer environment in manner that integrates coordinate-based, group-averaged data gleaned from the neuroimaging literature with multivariate, meta-analytic pipelines for model discovery, or integrates primary neuroimaging data from patient and control samples with multivariate, primary-data pipelines for model validation, or both;254929-9580-4034v.1

[0147] create a computer environment in manner that integrates connectomic modeldiscovery’ functions with connectomic model-validation functions within a high-performance computing environment;

[0148] create a computer environment as a cloud-based, online platform accessible by remote users;

[0149] create a computer environment in a manner that uses resting-state functional MRI (rs-fMRI) as primary data for disease-network quantification for use as biomarkers;

[0150] create a computer environment to provide coordinate-based, group-averaged data from the neuroimaging literature in a curated database, to allow filtered queries by users to enable the design of disorder-specific biomarkers; and / or

[0151] create a computer environment in which the coordinate-based, group-averaged data includes task-activation fMRI data, or voxel-based morphometric data (VBM), or voxel-based physiological data (VBP), or any’ combinations thereof.

[0152] In embodiments, the neurologic disorder comprises an epilepsy and the method is employed to effect presurgical mapping in epilepsy. In embodiments, the method further comprises performing an epilepsy surgery’ based thereon.

[0153] In embodiments, TMS or IR-TMS is used as atreatment. TMS treatments have been described in US 11,458,326 B2, incorporated by reference. Brain networks generated in accordance with embodiments of the present invention can be used to provide patient specific TMS treatments.Computer implemented systems and methods

[0154] V arious inventive concepts may be embodied as a non-transitory computer readable storage medium (or multiple non-transitory computer readable storage media) (e.g., a computer memory’ of any suitable type including transitory' or non-transitory' digital storage units, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement one or more of the various embodiments described above. When implemented in software (e.g., as an app), the software code may be executed on any suitable processor or collection of processors, whether provided in a single computer or distributed among multiple computers.264929-9580-4034v.1

[0155] Further, it should be appreciated that a computer may be embodied in any of a number of forms, such as a rack-mounted computer, a desktop computer, a laptop computer, or a tablet computer, as non-limiting examples. Additionally, a computer may be embedded in a device not generally regarded as a computer but with suitable processing capabilities, including a Personal Digital Assistant (PDA), a smartphone or any other suitable portable or fixed electronic device.

[0156] Such computers, such as the computers of the HPC, may be capable of performing the methods and algorithms in this disclosure. In embodiments, such methods and algorithms may be incapable of being performed by hand - e.g., with pen and paper.

[0157] Also, a computer may have one or more communication devices, which may be used to interconnect the computer to one or more other devices and / or systems, such as, for example, one or more networks in any suitable form, including a local area network or a wide area network, such as an enterprise network, and intelligent network (IN) or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks or wired networks.

[0158] Also, a computer may have one or more input devices and / or one or more output devices. These devices can be used, among other things, to present a user interface. Examples of output devices that may be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that may be used for a user interface include keyboards, and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, a computer may receive input information through speech recognition or in other audible formats.

[0159] The non-transitory computer readable medium or media may be transportable, such that the program or programs stored thereon may be loaded onto one or more different computers or other processors to implement various one or more of the embodiments described above. In embodiments, computer readable media may be non- transitory media.

[0160] The terms ‘"program;’ “app,” and “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that may be employed to program a computer or other processor to implement various embodiments as described above. Additionally, it should be appreciated that, according to one aspect, one or more computer programs that when executed perform methods of this application need not274929-9580-4034v.1reside on a single computer or processor but may be distributed in a modular fashion among a number of different computers or processors to implement various embodiments of this application.

[0161] Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or distributed as desired in various embodiments.

[0162] Databases may include computer readable memory (also referred to as ‘memory'). For example, data storage space 3memlN may be and / or include computer readable memory, used to store data as described in the disclosure. Memory may be embodied by suitable hardware, including but not limited to the following: hard disk drives, serial advanced technology attachment (SATA) hard drives, SATA solid state drives (SSDs), non-volatile memory express (NVMe) SSDs. tape drives.

[0163] Also, data in databases may be stored in computer-readable media in any suitable form. For simplicity of illustration, databases may be shown to have fields that are related through location in the data structure. Such relationships may likewise be achieved by assigning storage for the fields with locations in a computer-readable medium that convey relationship between the fields. How ever, any suitable mechanism may be used to establish a relationship between information in fields of a data structure, including through the use of pointers, tags or other mechanisms that establish relationship between data elements.EXAMPLES

[0164] Examples are provided below7to facilitate a more complete understanding of the invention. The following examples illustrate the exemplary7modes of making and practicing the invention. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only.Example - Testing Low-d M-ICA via study replication

[0165] We first tested the statistical algorithm implemented in our Low-d M-ICA Pipeline Application by replicating the findings of Smith et al. (2009) and Vanasse et al. (2021) - the first neuroimaging meta-analyses performed using independent component analysis. Although these prior studies do not broach the topics of context-specific networks and low dimensionality284929-9580-4034v.1estimation (essential elements of our invention), they applied the same fundamental algorithm (MELODIC-ICA from the FMRIB Software Library) at high dimensions, analyzing transparadigm and trans-diagnostic datasets to identify many (>20) general network structures. Successful replication of the results from Smith et al. (2009) and Vanasse et al. (2021) at d = 20 demonstrated the high-performance computing capabilities of the Low-d M-ICA Pipeline Application. Note that the consistency of the results was enhanced by our use of a standardized FWHM to modeling each experiment input to the meta-analysis.Example - Testing Low-d M-ICA on emotion behavior

[0166] After operationalizing the computational resources and codifying our theoretical approach to dimensionality estimation, we first used the pipeline to model the networks governing emotive behavior, applying M-ICA to 1798 functional imaging experiments performed in healthy subjects (input data summarized in Table 3).

[0167] Fig. 4 depicts dimensionality estimation for emotive behavior, in accordance with embodiments of the present invention. As shown in Fig. 4, identifying an optimal d of four emotion-specific network patterns from 1798 task-activation experiments was performed in healthy subjects. Network 2 was parcellated at d = 5 and two canonical networks were detected at subsequent dimensions, satisfying the stopping condition.

[0168] In this experiment, input data in Table 3 was used.Table 3.EmotionSign Subtype Papers Subjects Experiments Conditions Locations + Happiness 88 1480 200 310 1472 + Humor 6 98 14 20 116 + Reward / Gain 199 3877 791 708 6100 + Total (Pos) 379 7094 1224 1323 9422 Anger 49 969 119 169 781 Anxiety 38 848 100 124 752 Disgust 51 825 126 162 1099 Embarrassment 13 301 31 40 238 Fear 92 1617 235 288 1634 Guilt 9 176 30 39 179 Punishment 15 420 58 54 219 Sadness 66 1079 165 219 1160 Total (Neg) 368 6981 1055 1252 7860 n.a. Intensity 9 152 33 36 203294929-9580-4034v.1Total (All) 613 11534 2124 2195 16306n.a.Redundancy Corrected _Ja586 1888 15614 total

[0169] An optimal dimensionality of four components was determined (Fig 4), including action / execution, reward / fronto-parietal / cingulate, attend on / cogniti on, and limbic networks (see network interpretation in Figs. 5A-E). These findings corroborate a prior four-network delineation of the functional architecture underlying emotion processing (Marshall et al., 2019; Brain; Hunt et al., 2023 - BMC Neuroscience).

[0170] Figs. 5A-5E depict network interpretation of results for emotive behavior obtained in accordance with embodiments of the present invention. Fig. 5A depicts Emotion-ICA vs. Canonical Networks. Fig. 5A shows that M-ICA networks are distinct from canonical networks (-combinations of pieces of multiple canonicals). Fig. 5B depicts Emotion ALE vs. Canonical Networks. As shown in Fig. 5B, emotion sub-types load on similar sets of canonical networks. The mapping is not 1-to-l, but similar portions of the same canonical networks. Fig. 5C depicts Emotion ALE vs. Emotion-ICA. As shown in Fig. 5C, M-ICA networks load differentially on the spatial patterns of emotion sub-category ALE maps; none correlate exclusively with a single map. Thus, conventional meta-analytics do not delineate the network patterns revealed by M-ICA. Fig. 5D depicts behavior analysis of four networks generated in accordance with embodiments of the present invention. Fig. 5E depicts emotion-specific sub-domains of four networks generated in accordance with embodiments of the present invention.Example - Low-d M-ICA derived biomarker for mesial temporal lobe epilepsy Background

[0171] MTLE (“mesial temporal lobe epilepsy"’) is classically described as a neural network disorder involving abnormal (synchronous) neuronal firing that propagates in a stereotypic manner along the seizure network.

[0172] Currently, clinical imaging methods focus primarily on using MRI to visually identify structural lesions and radiotracer-based approaches (e.g. PET & SPECT) to reveal functional abnormalities. In conjunction with a myriad of other qualitative metrics, these assessments guide clinical management.

[0173] However, non-invasive functional imaging with resting-state fMRI (measuring the blood oxygen level dependent (BOLD) signal as a proxy for neural activity ) is not incorporated 304929-9580-4034v.1as a diagnostic, despite long-standing evidence suggesting its potential utility in guiding preoperative mapping (Barron et al., 2015) and recent validation of BOLD derived metrics for the more invasive methods (PET / SPECT) currently in used in standard assessment. Unfortunately, clinical applications of rs-fMRI in epilepsy diagnosis and management are limited by a lack of quantitative tool for interpreting the data.

[0174] Research methods (e.g. network modeling) have been developed to characterize changed structure and distribution in functional network dynamics from imaging. However, while functional network modeling has shown promise for developing data-driven biomarkers applicable to rs-fMRI (in epilepsy, Barron et al., 2013, 2015; as well as other neurologic disorders, Chiang et al. 2019), many such biomarkers fail to generalize across imaging centers or, in the case of certain structural imaging metrics, are simply not applicable at the per-subject level (i.e. cannot be applied to individual patients - e.g. voxel-based morphometry / VBM analysis of grey matter atrophy on T1 -weighted MRL Ashbumer & Friston 2000). Consequently, no quantitative neuroimaging biomarkers have successfully been translated to support patient care.

[0175] When non-invasive tools are inconclusive / offer insufficient guidance for the assessment of these patients (with MTLE or other epilepsies), highly invasive monitoring techniques are often required (e.g. electrodes surgically implanted in the brain). These limitations in patient management have prompted further research to elucidate the pathophysiology of MTLE and develop new network modeling techniques for developing quantitative interpretive tools that can be applied to individual patients in support of diagnosis, surgical planning, and stratification of potential MTLE sub-groups / sub-types.

[0176] More recently, MTLE effects have been shown to exhibit pathologic changes in both the structure and function of the brain that appear to be more widespread and complex than previously thought - evidenced by a meta-analysis of VBM (updated from Barron et al.2013) and VBP (novel meta-analysis; Towne et al. 2022). Informed by the inventors’ Ergodic Inference Approach to Translational Neuroimaging Biomarker Development, the Low-d M-ICA tool was used to identify two novel sub-networks in MTLE as potential seizure propagation routes and putative biomarkers for validation in clinical (per-subject) rs-fMRI data.314929-9580-4034v.1Methods

[0177] The MTLE Sub-Network Models comprise an image analysis algorithm for diagnosing and stratifying mesial temporal lobe epilepsy (MTLE) patients, with additional value for informing pre-surgical planning in MTLE by predicting seizure-onset laterality. Resting-state functional magnetic resonance imaging (rs-fMRI) serves as the input data, and the outputs include metrics of functional connectivity (e.g. model fit statistics and path correlation coefficients).

[0178] The input for both MTLE Sub-Network Models requires acquisition of three wholebrain MRI sequences: rs-fMRI to demonstrate the blood-oxygen-level dependent (BOLD) signal, a gradient-echo fieldmap for distortion correction of rs-fMRI, and a Tl-weighted image for image registration during image preprocessing. The preprocessing pipeline includes motion correction, BO unwarping, slice timing correction, and spatial smoothing (Woolrich et al.2001). Additional noise-reduction steps include independent component analysis-based cleanup (e.g. ICA- AROMA) and nuisance regression (Pruim et al. 2015). After image preprocessing, the regions-of-interest (ROI) specified by fMACM nodes (derived using Low-d M-ICA) are used to extract the mean timeseries data. The edges specified by the fMACM model are used to define the functional connectivity for path coefficient computation.

[0179] Connections between the brain regions identified in MTLE-SN1 and MTLE-SN2 were confirmed meta-analytically using functional meta-analytic connectivity modeling (fMACM). which is a method for determining the spatial convergence of results from published neuroimaging papers (i.e. the model describes consistent findings in the literature). The fMACM model leverages the large volume of existing literature and is not biased toward any particular dataset. Therefore, it is robust to disease variation and clinically useful. Specifically, construction of the fMACM model was undertaken by using coordinate-based publications in the BrainMap neuroimaging database.

[0180] The functional network structure is modeled as an AR(1) autoregressive process using structural equation modeling (SEM). This unified SEM approach is taken to improve the temporal representation of rs-fMRI, which includes lag variables to correct for the autocorrelations in rs-fMRI timeseries data. The observed variables are defined by the mean timeseries data (i.e. fMACM nodes). The paths are defined by the functional connectivity between the nodes (i.e. fMACM edges). The path analysis computes standardized semi-partial regression coefficients for each fMACM edge based on maximum likelihood estimation324929-9580-4034v.1(Arbuckle 2017). Additional paths may be incorporated in an exploratory manner. Overall model fit statistics are computed to demonstrate the degree of fMACM model fit to rs-fMRI. The root mean square error of approximation (RMSEA) is utilized as the primary fit criterion given its relative insensitivity to the effects of sample size; an RMSEA of < 0.08 indicates a reasonably good fit to the data (Browne and Cudeck 1993).

[0181] The model consists of two components: (1) “nodes” defined by regions of localized grey matter atrophy or functional changes in MTLE; and (2) “edges” defined by healthy functional connectivity involving the specified nodes (i.e. regions of pathologic change in MTLE). The nodes were determined by using voxel-based morphometry results from 1,599 MTLE patients and 1,441 healthy controls (HC). The edges were computed by using functional imaging results sampled from the BrainMap Task- Activation database, representing 110,352 HC.

[0182] For each network obtained by M-ICA, functional connections between distinct neuroanatomical regions were defined / identified via fMACM. Similar to M-ICA, fMACM leverages co-occurrence patterns for network modeling. Unlike M-ICA, fMACM relies on serial application of mass-univariate statistics, rather than mass-multivariate methods, to identify connections between pre-specified nodes (regions of the brain). Nodes were defined by the clusters in each network obtained via M-ICA.

[0183] The premise for fMACM is that functionally connected brain regions will coactivate in physiologic task performance and. consequently, be co-reported in task-activation studies. Thus, given a set of task-activation experiments reporting at least one coordinate within a given region, spatial convergence of coordinates (from the same set of experiments) in other regions of the brain implies a functional connection with those regions. In the present application of fMACM to an M-ICA network, each cluster is used as a seed to search the BrainMap Database and obtain all experiments reporting coordinates within that seed-region; then, a conventional ALE meta-analysis is performed and, from the resultant ALE map, statistics are computed for all other cluster-regions in the M-ICA network. Significant spatial convergence in a cluster-region for a given seed-region indicates a functional connection between the seed-region pair. Further discussion of fMACM methods are available elsewhere (Kotkowski et al., 2018).

[0184] The algorithmic applications of MTLE-SN1 and MTLE-SN2 provides path-level and model-level output for diagnosis, patient stratification by differential loadings on each sub-334929-9580-4034v.1network, and prediction of seizure-onset laterality. Specific path coefficients are used for each diagnostic purpose. The validation analysis with age and sex-matched HC is underway to quantify the diagnostic accuracy via logistic regression modeling and construction of a receiver operating characteristic curve (AUC will be quantified with and without lag variables). A similar approach is taken to quantify predictive accuracy for seizure-onset laterality7. The model fit statistic of RMSEA tracks differential loadings on each sub-network model for stratifying patients.

[0185] Seed-to-region analysis was performed.

[0186] The resultant networks (spatial maps) obtained from the melodic_IC.nii.gz file were loaded into Mango for cluster identification. From each cluster, the coordinate with the maximum (pseudo-)Z score was identified and a spherical region of interest (ROI) was constructed at this location. The ROI was then used to seed / search the Task- Activation Sector of the BrainMap Database using Sleuth software, to obtain all experiments (Normal Mapping, Activations Only, Diagnosis-Normals) reporting foci within the specified ROI. A conventional ALE meta-analysis was then performed using GingerALE to obtain a map of convergent findings that co-occur with coordinates reported within the ROI.

[0187] ROIs defined by maximum Z score within clusters identified for the MTLE networks. Input data quantities for seed-to-region analyses are provided below, (i.e. amount of data returned by each BrainMap Database Task-Activation Sector search, per cluster-ROI / node included in each M-ICA network’s fMACM analysis). These results are shown in Tables 8 and 9.Results

[0188] Two novel MTLE-specific connectomes were identified, illustrating the ability of M-ICA to identify distinct co-occurrence patterns that exist across structural and functional neuroimaging literature, even for smaller datasets (Fig. 6). The regions within these networks were associated with distinct subsets of behaviors (Fig. 7, Figs. 9A-9C, and Fig. 10) that are altered in ictal and peri-ictal symptoms of MTLE. This suggests that each network is clinically plausible and may mediate different aspects of MTLE semiology7. Note that differences between these two network components were not due to structural-functional effect modification, as VBM and VBP experimental contributions were not significantly skewed I / 2homogeneity: p > 0.05; Tables 5-7).344929-9580-4034v.1

[0189] We then compared the M-ICA results for MTLE with a conventional massunivariate CBMA voxel-wise operator: activation likelihood estimation (ALE). Unlike M-ICA, ALE simply computes voxel-wise spatial convergence across studies using a local operator which cannot delineate multiple network patterns. To demonstrate this methodological distinction, we performed an ALE meta-analysis on the same set of 74 experiments, identifying the most robust regions of MTLE pathology. Notably, the M-ICA networks formed a superset of the regions identified by ALE. This demonstrates that M-ICA is not only capable of defining disease specific networks but can also reveal novel network regions undetected by conventional CBMA methods.

[0190] Additional spatial comparisons underscored the novelty of the MTLE networks identified by M-ICA. Neither network resembled any canonical network (R < 3; Tables 5-7). Notably, the only regions shared by both networks were collocated with the most significant regions identified by ALE: the hippocampus and medial dorsal nucleus of the thalamus. Beyond these two regions, both networks were spatially distinct (R=0.00158) - unsurprising, considering that ICA optimizes spatial independence. However, given that the networks were separately identified (not parcellated from d = 1) and extended from the mesial temporal seizure-onset zone and thalamus (regions known to be connected), these patterns of co-occurrent pathology likely represent distinct seizure propagation routes (though additional testing would be required). Connectivity among the regions within M-ICA networks was further confirmed by constructing functional meta-analytic connectivity models (fMACM) for each MTLE component (Figs. 8A-8B & Tables 5-7), extending our group’s prior work demonstrating how MTLE pathology propagation mirrors healthy functional connections (Barron et al., 2012) and providing additional support for the Network Degeneration Hypothesis (Seeley et al., 2009).

[0191] Taken together, the results in emotion and MTLE demonstrate that the CBMA multivariate ICA algorithm implemented in M-ICA can detect latent meso- and macroscale connectomic properties within tabular results of mass-univariate neuroimaging publications. Statistically, this is an example of ergodicity, by which sufficiently large cross-sectional analyses can reveal dynamic processes exhibited across time in individual systems (Isvoranu et al., 2022). Arguably, the most impactful (> 5,000 citations) demonstration of meta-analytic ergodicity is trans-paradigm ICA meta-analysis of task-activation literature, discussed previously, that exposed the canonical network architecture observed dynamically in primary resting-state IMRI timeseries (Smith et al., 2009). Ergodicity underlies the successful354929-9580-4034v.1application of meta-analytic models to resting- state fMRI data for functional network modeling and, ultimately, for biomarker development.

[0192] In conclusion, M-ICA represents a significant advancement in neuroimaging network analysis by enabling the rapid, reproducible extraction of context-specific neural networks from published coordinate data. Unlike traditional ICA approaches often used to identify canonical brain networks in primary’ imaging data, M-ICA relies on a sparse data substrate, to remove dominant physiological architectures and facilitate the discovery of novel, behavior- and disease-specific networks. By combining tailored dimensionality estimation with a cloud-based computational framework, M-ICA provides a powerful tool for investigating the latent connectomic architecture underlying diverse physiologic and pathologic brain states. This open-source, scalable approach to data-driven network modeling thus sets the stage for improved mechanistic understanding and biomarker development in neuroscience research.

[0193] Fig. 6 depicts dimensionality estimation for mesial temporal lobe epilepsy pathology in accordance with embodiments of the present invention. As shown in Fig. 7, the optimal d of two network components is estimated. A noise component was identified at d = 3 and further network detection was prohibited by power requirements.

[0194] Fig. 7 depicts behavioral associations of MTLE network regions generated in accordance with embodiments of the present invention compared with behavioral associations of regions that were identified via simple spatial convergence across studies in a conventional activation likelihood estimation (ALE) meta-analysis of the same set of MTLE experiments.

[0195] Figs. 8A-8B depict meta-analytic connectivity models constructed for both MTLE networks identified by M-ICA. As shown in Figs. 8A-8B. it was confirmed that the regions identified as pathology coordinate co-occurrence patterns share plausible physiologic connections, in accordance with the Network Degeneration Hypothesis.

[0196] In this experiment, input data in Table 4 w as used.Table 4.MTLEModality Metric Papers Patients Controls Experiments Locations TotalVBMz,, 23 751 1054 42 385(Atrophy)VBP ALFF 8 325 205 16 101fALFF 1 26 26 2 10 ReHo 5 106 131 11 65364929-9580-4034v.1ACF 1 19 19 1 16 SUVR 1 26 26 2 11 Total (VBP) 13 325 334 32 203 Total (All) 36 1076 1388 74 588

[0197] Figs. 9A-9C depict Mango toolbox analyses of two MTLE networks and a noise component in accordance with embodiments of the present invention.

[0198] Fig. 10 depicts a comparison of two MTLE Meta-ICA networks generated in accordance with embodiments of the present invention compared with networks generated by ALE.

[0199] Tables 5-7 depict statistical analyses of MTLE pathology.Table 5.Observed Percent Contribution to Each NetworkVBM VBPNetwork 1 74.3% 25.7%Network 2 67.5% 32.5%

[0200] test of homogeneity. We fail to reject the null hypothesis (p > 0.05) that the distribution of VBM and VBP experiments is homogenous between the networks.Table 6.ExperimentsVBM Totals (normalized from (% MA maps * VBP ExpectedExpected MELODIC) normalized totals)IC1 20.01 12.9 32.9075IC2 24.99 16.1 41.0925Totals (MA maps) 45 29 74% Total 45 / 74=60.81% 29 / 74=39.19%Table 7.Experiments(weighted VBM Totals (normalized from VBP Observedcontribution of MA Observed MELODIC) maps per IC)IC1 24.46 8.45 32.9075IC2 27.74 13.35 41.0925Totals (MA maps) 45 29 74% Total 45 / 74=60.81% 29 / 74=39.19%

[0201] Tables 8 and 9 depict results for the MTLE network 1 and 2 identified in accordance with embodiments of the present invention.374929-9580-4034v.1Table 8Cluste Talairach Talairach BrainMap Task- Activation Sector Content, Per r Daemon Label Coordinates Cluster SeedIndexX Y Z Paper Subject Experiment Condition Foci s s1 Righ Sub-lobar Extra- 2 2 164 3344 198 436 301 t nuclear 1 962 Left Lingual Gyrus -4 -2 120 2455 138 308 1968 2 4Righ Cerebellar Tonsil 3 67 1587 76 176 120 t 0 5 3 26 64 Left Cerebellar Tonsil 80 1422 90 219 1703 5 3 5 0 4 65 Left Precentral Gyrus 2 126 2184 169 351 2595 1 6 3 4 06 Left Supramarginal 2 108 2581 128 268 188Gyrus 5 5 8 32 2Left Middle Occipital 162 3505 215 477 349 Gyrus (BAI 8) 3 8 1 88 0 08 Left Precuneus -2 2 88 2015 96 197 124(BA31 ) 6 4 869 Righ Precentral Gyrus 5 3 103 2139 142 274 220 t 0 1 0 0210 Righ Middle Frontal 4 3 -4 118 2477 134 315 188 t Gyrus 2 2 7 11 Left Superior Frontal 6 8 32 690 36 85 520Gyrus 1 2212 Left Middle 6 153 2963 192 425 272Temporal Gyrus 4 6 86 413 Righ Middle Occipital 4 171 3671 207 471 300 t Gyrus 0 7 1 96 014 Left Inferior 183 3592 221 491 348Temporal Gyrus 5 5 1 4 (BA20) 0 6 215 Left Paraliippocampu 79 1623 85 188 142 s 3 1 1 54 8 6384929-9580-4034v.116 Left Middle Frontal 3 -6 127 2395 147 319 184 Gyrus 4 4 9217 Left Cerebellar 97 2101 113 262 174Declive 2 5 1 96 2 218 Left Middle Frontal 5 6 120 2478 132 335 181Gyms 2 8419 Left Middle 4 42 971 50 133 785Temporal Gyms 5 24 420 Left Superior 1 152 2955 190 410 289Temporal Gyms 4 3 6 3 (BA41) 4 421 Left Inferior 27 659 32 76 530Temporal Gyms 6 1 2(BA20) 0 6 0Table 9Cluste Talairach Talairach BrainMap Task- Activation Sector Content, Per r Daemon Label Coordinates Cluster SeedI ndexX Y Z Paper Subject Experiment Condition Foci s sI Left Hippocampus 85 1619 99 225 1463 1 1 0 0 4 82 Left Caudate Body -6 1 6 110 2260 122 281 1694 5 3 Left Parahippoearnpu -2 111 2657 123 276 185 s 2 3 32 64 Left Paracentral -2 4 132 3517 146 331 214Lobule (BA31) 1 4 945 Righ Caudate Body 8 1 6 158 3564 180 427 254 t 4 0 6 Left Cingulate Gyrus -4 4 4 464 9361 614 1346 974(BA24) 6 9 7 Righ Cingulate Gyms 6 2 44 996 46 99 764 t 1 648 Left Superior 8 -8 98 1971 108 252 192Temporal Gyms 5 549 Righ Thalamus 1 1 266 5128 327 704 665 t 0 1 2 82394929-9580-4034v.110 Left Superior 1 36 759 40 98 618 Temporal Gyrus 3 4 36 011 Righ Sub-Lobar 1 _2 1 205 4103 247 578 503 t Extra-Nuclear 0 0 8 12 Left Middle Frontal 2 2 152 2697 183 418 279Gyrus 5 6 4 5013 Left Medial Dorsal -4 6 242 5937 280 605 468Nucleus 1 1 (Thalamus) 614 Righ Parahippocampu 2 -8 85 2045 107 219 135 t s 0 2 3215 Left Superior Frontal 6 6 86 1981 99 229 156Gyrus (BA6 ) 1 2 3416 Left Sub-Lobar 2 8 168 2984 202 462 313Extra-Nuclear 3 3417 Righ Cuneus (BA30) 6 6 88 1690 96 212 135 t 6 3618 Righ Anterior 4 3 2 137 3157 146 314 221 t Cingulate Gyrus 6 2 4 (BA32)19 Righ Paracentral 4 4 65 1328 71 159 103 t Lobule (BA31) 3 4 5020 Righ Medial Frontal 1 1 18 399 18 39 198 t Gyros (BA25) 2 8 1821 Left Insula 1 125 2536 153 316 2193 1 0 1 8 022 Left Superior Frontal 2 5 94 2231 99 222 163Gyrus 1 6 2 1423 Left Medial Frontal -4 6 0 42 794 47 107 612Gyrus 224 Left Medial Frontal 2 19 450 20 56 313Gyros 1 4 10 825 Left Superior -6 4 114 2309 163 307 251Temporal Gyros 5 0826 Righ Middle Frontal 2 0 5 165 4362 203 433 307 t Gyrus (BA6) 4 8 0 27 Left Cingulate Gyrus _2 2 2 190 4042 215 446 317(BA32) 6 6 7404929-9580-4034v.128 Rig b Cingulate Gyrus 8 4 73 1545 79 185 129 t (BA24) 1 2 5 029 Left Medial Frontal -2 3 3 104 2402 112 272 140Gyms (BA9) 8 2 7 30 Righ Middle Frontal 2 2 6 97 2998 111 241 184 t Gyrus 0 0 4 31 Left Thalamus 2 164 2998 190 438 3221 2 1 2 632 Righ Parahippocampu 2 -2 133 3251 152 308 273 t s 0 3 6433 Left Frontal Sub- 4 4 133 3165 148 336 217Gyral 4 4 50

[0202] Methods in accordance with the present invention have also been in Tow ne et al.2025, incorporated herein by reference.

[0203] The results were also confirmed via graph theory modeling. This model characterizes higher level properties of MTLE pathology (i.e. Network topology). Nodes and edges were defined using an adapted meta-analytic approach, first described by Cauda et al., 2018 (doi: 10.1093 / brain / awy252). Peak coordinates in an ALE image (loose statistical thresholding) w ere identified and selected as nodes. Co-alteration patterns w ere then identified across experiments, computing edges as patel's k. A total of 118 nodes and 93 edges were identified. Network modularity was computed to identify two separate MTLE module networks; each model was comprised of separate regions associated with distinct behavioral associations, concordant with the findings from the M-ICA-derived sub-networks. A variety of other graph theory parameters were then computed, confirming the topological importance of the core nodes identified in the ALE / MACM model and uniquely revealing the medial frontal gyrus as a key node in MTLE network topology.

[0204] Figs. 27A-B depict a rich co-alteration model of MTLE. Coordinates for the M-GTM model are shown in Tables 14 and 15.

[0205] Figs. 28A-H show complex patterns in MTLE pathology’ reveal a unique disease network (A) that exhibits modular architecture (B, C, F) congruent with physiologic behaviors (D, G) often affected during the ictal or peri-ictal period (E, H). (A) Meta-analytic graph theory modeling (Meta-GTM) was used to construct anode-edge model, deriving 118 nodes and 93 edges from coordinates of MTLE pathology’ reported across 74 imaging experiments (45 structural, 29 resting-state functional). A total of 30 nodes had degree > 1 in the disease model.414929-9580-4034v.1(B) Visualization in Cytoscape (v3.8.2) qualitatively confirmed the hippocampus and medial dorsal nucleus (MDN) thalamus as crucial disease nodes. Clusters of MTLE network pathology were detected using a spectral partition-based approach (Newman 2006), defining four modules. Modules 3 and 4 were comprised of only two nodes. Modules comprised of >2 nodes are depicted above on template brain images. (C) Module 1 (comprised of 15 nodes, 40 edges) spanned anterior neocortical and paracentral regions as well as deep nuclear, mesial temporal, and other limbic structures. (D) To assess Module 1 as a plausible MTLE network, behaviors physiologically associated with the 15 nodes were identified via Behavioral Analysis in Mango (v4.1), computing spatial comparisons to coordinate data from 21,435 task-activation experiments, 107,167 healthy subjects, stored in the BrainMap database (http: / / rii.uthscsa.edu / mango / mango.html). Associations were congruent with network dysregulations observed in semiology (amnestic and social-emotional deficits). (E) Similarly, Disease Analysis was used to assess regional similarity to know n patterns of structural disease pathology' (4,398 experiments, 115,627 subjects), further corroborating network plausibility'. (F) Module 2 (comprised of 11 nodes, 32 edges) incorporated cerebellar, parietooccipital, precentral, and prefrontal cortices. (G) Behavioral Analysis was applied Module 2’s nodes, identifying language behavioral domains affected in MTLE. (H) Disease Analysis of the Module 2 network demonstrated high specificity for MTLE pathology.

[0206] Figs. 29A-C show that the MDN Thalamus and Medial Frontal Gyms are the most influential extra-hippocampal nodes in the MTLE Co-Alterati on Network. (A) For the 30 nodes with degree > 1, MATLAB (R2020b) and Cytoscape (v3.8.2) were used to compute topological metrics for centrality, clustering and topological coefficients, neighborhood connectivity, efficiency (average shortest path length), radiality, eccentricity, and nodal stress. (B) The overall influence of each individual node on MTLE network topology' was computed by normalizing each metric and computing an average across all parameters. The Medial Frontal Gyrus, MDN Thalamus, and Hippocampus are the three most influential nodes in the MTLE Co- Alteration Network. Cerebellar tonsil, precentral, and superior temporal gy ri also showed high network centrality within their respective modules. Inferior parietal lobule and sub-gyral hippocampus w ere influential (but not central) due to high neighboring connectivity, clustering, and topological coefficients. M3 was most efficient, with one edge to hippocampus. (C) Intermodular connections are mediated by the Medial Frontal Gyrus, MDN Thalamus, and Hippocampus. The medial frontal gyrus in Module 2 shares edges with the superior temporal, precentral, and postcentral gy ri in Module 1.424929-9580-4034v.1Example - functional connectivity biomarker for alcohol use disorderBackgroundAlcohol use disorder

[0207] Excessive alcohol consumption is responsible for more than 3 million deaths worldwide each year (WHO, 2024) and is the third leading preventable cause of death in the US (Mokdad et al.. 2004). Unhealthy levels of alcohol consumption increase the risk for health problems (Rehm et al., 2003a, 2003b) and are associated with significant behavioral and economic consequences: costing the US $249 billion in 2010 (the most recent year of data available; CDC, 2022). Thus, there is a great need for effective treatments and objective alcohol use monitoring.

[0208] Behavioral and medication-based therapies for AUD exist. But for many treatmentseeking individuals, these therapies fail to affect drinking or fail to prevent relapse. The most widely used medication for AUD is naltrexone, but its efficacy is low, helping only a fraction of those in need. Novel therapeutic approaches, especially non-drug or otherwise non-invasive options, are needed.Transcranial magnetic stimulation

[0209] Transcranial magnetic stimulation (TMS) is a noninvasive, electromagnetic neuromodulatory therapy. TMS is FDA-approved as being safe and effective for major depressive disorder (MDD) and obsessive-compulsive disorder (OCD) but is not yet approved for AUD. TMS treatment trials in AUD have been promising but not compelling. No prior trial, however, has been performed with the sophisticated and effective TMS methods now available. TMS efficacy is maximized by connectomic conceptualization of the disorder-related neural disfunction, targeting brain networks (connectomes) known to be altered in the condition being treated.

[0210] Several trials have examined the safety and efficacy of TMS as an AUD treatment but had mixed results. Schluter et al (2019) delivered 10 Hz rTMS to the F3 location of the international 10-20 EEG system (approximately the left DLPFC) over ten treatment sessions (i.e. 30,000 TMS pulses) in 80 patients with AUD, however, they did not find any reductions in alcohol consumption or cravings. Perini et al. (2020) reported a placebo-controlled rTMS study, in which they delivered 10 Hz rTMS to the insula using a deep TMS coil — which has less stimulation focality than normal TMS coils — in patients with AUD over three weeks (22,500 TMS pulses). However, they did not find any differences in alcohol cravings or434929-9580-4034v.1consumption between the rTMS group (N=29) compared to placebo group (N=27). A third, recent study (Belgers et al., 2022) delivered 10 Hz rTMS to the F4 location (i.e. approximately right DLPFC) over ten treatment sessions (i.e. 30,000 TMS pulses). This group did find significant differences in alcohol cravings and consumption between the rTMS (N=14) and placebo (N=14) groups, however, the sample size of this study was very low. Across these studies, imprecise targeting approaches and limited number of delivered TMS pulses may have contributed to the large variations and muted response rates. We propose to enhance these early TMS trial approaches by using state-of-the art TMS delivery systems, patient-specific targeting strategies, and an accelerated TMS delivery schedule — which allows us to deliver many more TMS sessions in one day, thereby increasing the total number of pulses delivered throughout a study.

[0211] TMS is a neuromodulation therapy approved by the U. S. Food and Drug Administration (FDA) as a stand-alone treatment for Major Depressive Disorder (MDD) and obsessive-compulsive disorder (OCD), but not yet for AUD (Fitzgerald, 2021). TMS applies a strong (~2 Tesla), rapidly changing electro-magnetic field at the scalp to induce electrical current flow in underlying brain tissues. TMS has well-established applications in basic neuroscience and psychiatric therapies and can alter cerebral neurophysiology in a long-term manner. Although TMS is FDA-approved for several indications, TMS devices and procedures in clinical use are far from optimal. Psychiatric TMS treatments are most commonly performed using a hand-held TMS coil to locate primary motor cortex, followed by a scalp-surface measurement (5-6 cm anterior) to identify the treatment target site: the DLPFC. Coil orientation is established using the nose as the aiming point in a manner crudely optimized for the motor cortex (by orienting the E-field quasi-normal to the central sulcus; Fox et al., 2004) rather than optimized for the target site's gyral anatomy. Moreover, differences in patient head sizes can lead to targeting more anterior or posterior portions of the DLPFC (Fox et al.. 2012). With this in mind, ongoing work in MDD has demonstrated that TMS efficacy is maximized by 1) targeting brain networks known to be altered in the condition being treated; 2) treatment personalization through image guidance; 3) repeatable TMS coil delivery systems (neuronavigation); and 4) “accelerated’' treatment, with multiple TMS treatment sessions per day (Fitzgerald, 2021).

[0212] The vast majority of TMS clinical trials have not employed image-guided neuronavigation systems. This is despite the fact that the original motivation to target DLPFC was based on physiological abnormalities observed in MDD with PET (George et al., 1997)444929-9580-4034v.1and despite the fact that numerous trials have cited the need to accommodate individual differences in structural and functional neuroanatomy, even recommending per-patient imaging as the means to this end (e.g. O’Reardon et al., 2007). In a meta-analytic review of TMS trials in psychiatric disorders, Slotema et al. (2010) argued that a large portion of interparticipant and inter-trial variability7is attributable to inadequate neuroanatomical specification of treatment targets and delivery imprecision. This growing awareness that TMS effects are multi-regional and network-based has led to recommendations that neuronavigation be implemented based on connectomic principles, i.e. based on the FC network properties of the brain regions stimulated (Fox et al., 2012, 1997). Drs. Fox & Salinas have developed and applied neuroimaging methods in conjunction with acute and chronic TMS delivery for more than 15 years and have leveraged this 1st-hand knowledge to develop IR-TMS neuronavigation methods, including FC-based targeting. Dr. Fox and Dr. Salinas both hold patents regarding these technological advancements (Fox et al., U. S. Pat. No: 11,458,326 B2, 2022). These innovations have enabled us to accurately predict the functional brain responses of IR-TMS locally (Krieg et al., 2013) and throughout the targeted networks (i.e. connectomically; Salinas et al., 2016). These advancements give us the ability to assess how TMS treatments may affect local and remote brain activity, when delivered in a patient-specific, repeatable manner. CBMA Application to Psychopathology

[0213] Application of CBMA to psychopathology is grounded in Voxel-Based Morphometry (VBM; Ashbumer and Friston, 2000). VBMs are case-control contrasts of group-averaged structural MRIs, which extract disease-related, spatially reliable gray-matter alterations. The same statistical framework can be applied to functional brain images such as 18 FDG-PET, SPECT, and BOLD-based physiological images processed for Regional Homogeneity (ReHo; Zang et al., 2004) and Amplitude of Low- Frequency Fluctuation (ALFF; Hoptman et al., 2010). Functional case-control contrast of this literature is here termed Voxel-Based Physiology (VBP).Network Degeneration in Mental Disorders

[0214] Network architecture is a fundamental, multi-scale motif of central nervous system(CNS) organization. Network architecture is intrinsic to the CNS’ core function: information processing. In the CNS, information is encoded and transferred by neural firing dynamics. Functional neuroimaging methods (functional MRI [fMRI], PET, and SPECT) detect these electrophysiological events via neurovascular coupling. Inter-regional temporal covariances in hemodynamic / metabolic imaging modalities underlie all neuroimaging 454929-9580-4034v.1Functional Connectivity(FC) analytics. Microscale(cellular) networks exist beneath the spatial resolution of non-invasive neuroimaging(~0.5mm). Mesoscale and macroscale functional connectomics, however, are imageable. Task- activation and resting-state fMRI (rs-fMRI) times series imaging are widely used for FC modeling, both in healthy controls and in disorders. The dense, cellular interconnectivity required for CNS information processing places neural elements at risk of network-based disease processes of various t pes. Seeley et al. (2009) noted that dementing disorders exhibit network architecture closely matching rs-fMRI functional connectivity, presumably by transsynaptic transmission of disease-related proteins (tau, beta amyloid, alpha-synuclein, etc.). Neurological disorders with confirmed network propagation include proteinopathies, neurotropic viruses, epilepsy, and disconnection-induced diaschisis. For psychiatric disorders, the underlying mechanisms causing network-based brain alterations (atrophy and hypophysiology) are not yet well characterized, but likely reflect damage related to metabolic cost associated with excessive neuronal firing and nodal stress (Cauda et al., 2018; Crossley et al., 2016; Vanasse et al., 2021). Vanasse addressed this by a massive (9,000 cohorts), meta-analysis comparing the BrainMap TA and VBM DBs to demonstrate that disease entropy and behavioral entropy were highly correlated (p=0.0006, R2 = 0.532), inferring that nodal stress underlies network degeneration.

[0215] Meta-analytic independent components analysis (ICA) of the VBM literature was first reported by the Fox laboratory, combining psychiatric and neurological disorders (Vanasse et al., 2018). The psychopathology literature is now sufficiently large for analysis alone. Figure B.5.3 Panel A shows an ICA reduction of the BrainMap VBM / VBP psychopathology archive as of 12 / 2023). This analysis included >50,596 participants with 13 psychiatric disorders (ICDF codes). Eighteen spatial dimensions (IC) were computed (Figure B.5.3 Panel A). Disorder loadings on IC dimensions are illustrated (Figure B.5.3 Panel B). AUD (F10.1) loaded on seven ICs: 1, 2, 9, 10, 11, 15, 18 (Figure B.5.3 Panel B). Of these seven ICs, four contained cortical regions accessible to TMS. Although these results are highly promising, for biomarker uses, per-disorder, node-and-edge models are superior to transdiagnostic models.The Brain Map Biomarker Pipeline

[0216] Node-and-edge models are the most widely applied, versatile and quantitative family of network algorithms. Generally defined, nodes are variables; edges are covariances among variables. For neuroimaging, nodes are brain regions; edges are pairwise (node-to-node) covariances. Depending upon use case, edges can be binary (0,1), integers, or real numbers. Real numbers provide the highest level of quantitation and are used here.464929-9580-4034v.1

[0217] Meta-analytic co-alteration modeling (MACM) is a generic term for multivariate network modeling of the coordinate-based literature. The first MACM method was DAG modeling, a node-and-edge method, above. The most widely used MACM method is ALE-based pairwise covariance (Robinson et al., 2010). Subsequently developed node-and-edge MACM methods include: topological Graph Theory Modeling (GTM; Crossley et al., 2016); Bayesian Graphical Modeling (BGM; Cauda et al., 2018); and Structural Equation Modeling (SEM; Chiang et al., 2021) - All implemented in the BCPTM biomarker pipeline, below.

[0218] A key characteristic of our node- and-edge, biomarker-development strategy is using disorder-driven foci as nodes. The motivation for using VBM / disorder-driven nodes is the compelling demonstration by Smith et al. (2011) that node selection is the single most salient determinant of accurate network discovery in neuroimaging. If nodes are improperly placed, models are insensitive. That is. using atlas- based, general-purpose nodes or using hubs from '‘canonicar networks derived from rs-fMRI in healthy controls as nodes, substantially degrades network-model sensitivity (Beckmann and Smith, 2004; Smith et al., 2009). To optimize node selection, ALE meta- analysis is applied to disorder-specific VBM / VBP data. Model edges are also data driven, either from VBM / VBP covariances (e.g. Cauda et al., 2018) or from task-activation covariances in controls (e.g. Chiang et al., 2021), depending on the application.

[0219] The BCPTM biomarker- discovery pipeline is illustrated (Figure B.6.1.; Fox 2023, patent pending). As shown, ALE is applied to VBM / VBP case-control data for node discovery. MACM is used for edge discovery7. The meta-analytic node-and-edge model is then applied to primary7(per-participant) rs-fMRI for model validation, optimization and use. Applications of BCPTM-derived FC biomarkers include: neuromodulation targeting TMS targeting (Fox et al., U. S. Pat. No: 11,458,326 B2, 2022); disease diagnosis and treatment-outcome evaluation (Chiang et al., 2021;pat. pending).! n preparation for the present proposal, the BCP pipeline was applied to addiction disorders. VBM / VBP results in addiction disorders, both substanceuse disorder (SUD) and behavioral addictions (BA) were identified through the Brain Map® database (67 cohorts), supplemented by literature retrieval (38 cohorts), a cumulative corpus of 105 peer-reviewed experiment reports representing 4,148 patients and 4139 controls (Banajaa et al., in review). Of these, the most reports were of AUD: 26 reports representing 921 patients and 975 controls. Stimulants, nicotine, opiates, cannabinoids, internet gaming disorder, and pathological gambling were also included. Based on the spatial convergence of SUD and BA, the disorders were pooled for biomarker development (Figure B.6.2). The high474929-9580-4034v.1similarity of the ALE map (Panel A) and the ICA (Panel B; D=1) is evidence that the convergent ALE effects form a tightly connected network. Multiple cortical regions are evident, with DLPFC being the strongest. DLPFC connectivity (Panel Concludes Nucleus Accumbens (NAc), a reward-system hub and anterior insula, a Salience Network hub. For TMS targeting, FC will be optimized per patient between each of three cortical targets and NAc. The complete ALE / MACM model (Panel D) will provide secondary study outcome variables, quantifying alterations in model fit statistics and edge weightsMaterials and methods

[0220] IR TMS was delivered in 3 separate treatment arms: 1) Left, dorsolateral prefrontal cortex (DLPFC); 2) dorsomedial prefrontal cortex (DMPFC); 3) orbitofrontal cortex (OFC). IR-TMS will be delivered in an accelerated format (defined as > 1 session / day) consisting of four treatments per day for 10 treatment days (40 total) over a two-week treatment period (Mon-Fri). We chose these regions because they are implicated in AUD by meta-analytic network modeling of the case-control neuroimaging literature, strengthened by preclinical literature related to substance use disorders and recovery (e.g. Mobini et al., 2002; Wallis and Miller, 2003; West et al., 2013).

[0221] Stimulation of each of these implicated networks is expected to disrupt craving-associated neural activity patterns and thereby decrease drinking. An accelerated IR-TMS format is used to lower dropout rate by minimizing treatment duration and travel demands. IR-TMS outcomes will be compared with those of daily naltrexone administration, which is FDA-approved to diminish craving and drinking in AUD.

[0222] To demonstrate the effectiveness of IR-TMS and verify clinical outcomes, we used a direct biomarker of alcohol use, phosphatidylethanol (PEth). Both our biological and neuroimaging biomarkers are “patient specific.’7Changes in PEth which reflect effective treatment will be determined based on the patient’s own baseline levels. Similarly, the neuroimaging functional connectivity biomarker is optimized per patient using their own fMRI-based functional connectivity (FC) map.

[0223] We use Peth to help validate a novel and innovative new biomarker of AUD, using Coordinate-Based Meta-Analysis (CBMA) Mapping and modeling the mental operations of the human brain is a fundamental neuroscience objective that has been thoroughly revolutionized by neuroimaging. In this transformative, still-ongoing enterprise, the use of atlas -referenced, 3-D neuroanatomical coordinates has played a pivotal role. Meta-analyses —484929-9580-4034v.1by differentiating stable, reproducible effects from noise — are the bedrock of this collective truth.

[0224] We assess past 90-day drinking using a TLFB at baseline to establish recent and typical drinking patterns. We will use this as the baseline upon which we will assess changes in drinking over the course of treatment and in subsequent follow-up visits.

[0225] We assess AUD severity and recovery using two diagnostic tools: DSM-5 substance use checklist and TLFB assessment.

[0226] The Diagnostic Statistical Manual-5 Substance Use Checklist is an 11-item questionnaire that measures the degree (mild, moderate, severe) to which an individual meets criteria for a substance use disorder for the following substances: alcohol, cannabis, hallucinogens (LSD and others), inhalants, opioids, sedatives, hypnotics, anxiolytics, stimulants (amphetamines, cocaine, and others), tobacco, other substance use. It is integrated with current DSM-5 diagnostic criteria for AUD.

[0227] Timeline Follow Back (TLFB) documentation of alcohol use. TLFB is a widely utilized tool for measuring alcohol consumption patterns. By incorporating a calendar-based format, the TLFB minimizes recall bias and memory distortion, which increases the accuracy of self-reported alcohol consumption (Sobell et al., 2001; Sobell and Sobell, 1992). The TLFB utilizes various memory cues, such as birthdays, and holidays to assist participant recall contributing to the establishment of a reliable timeline of alcohol consumption episodes. The Alcohol TLFB has demonstrated excellent reliability and validity' metrics and has been successfully used across diverse populations, including clinical and community samples (Carey et al., 2004; Cook et al.. 2019; Hareskov Jensen et al.. 2023; Roy et al., 2008; Sobell et al., 2001; Wray et al., 2016). Another benefit of the Alcohol TLFB is that it produces detailed data on a variety' of frequency and quantity' metrics, such as the total number of drinking days, the average number of drinks per day, and the number of days when binge drinking occurs, and the highest amount of alcohol consumed in one sitting. Furthermore, it works well for recording episodic and continuous drinking trends (Sobell et al., 2003). Still, this remains a subjective measure, reliant on patient self-reporting recent alcohol use.

[0228] FC will be used as a biomarker both to guide treatment and for outcome variable. Curated tabular data are shared in 3 SQL databases of the coordinate-reporting literature: Task Activation, (TA DB); Voxel-based Morphometry (VBMDB); and Voxel-based Physiology494929-9580-4034v.1(VBPDB). BCPTM resources were used to generate both the AUD map (network nodes) and the AUD connectome (network edges)

[0229] Image-Guided Treatment Planning. TMS Stimulation will be targeted using our AUD MACM. Respective nodes of the MACM will be used to develop patient-specific treatment plans for either the DLPFC, DMPFC, or OFC. Each patient’s preprocessed rs-fMRI will be used to determine their cortical target to NAc connectivity map. A spherical ROI (r = 6 mm) will be placed at that patient’s maximum absolute response in the NAc. That downstream target will be used to determine the patient’s optimized NAc-to-target FC map. A scaling algorithm will use the coil-to- cortex depth to determine the TMS stimulator intensity setting (e.g. % of machine output) needed to deliver a depth-corrected, suprathreshold E-field strength, e.g. 95 Volts / meter (V / m), at that patient’s target site. Each patient’s structural MRI, target FC map, and optimized target site coordinates will then be loaded into our C3 Plan® software to determine the optimal TMS coil position, orientation, and intensity setting required to deliver the maximal effective E-field at the patient’s target site.

[0230] In 150 IR-TMS patients and 50 Naltrexone patients, six time points will be assessed: baseline, mid-treatment, post-treatment, 1 month, 3 months, and 6 months. Two types of response-predictive models will be created for each time point: a seed-to- whole brain FC map and MACM node-and-edge models. The seed-to-whole brain FC maps will be determined using FSL’s FEAT toolbox. A mixed-effects, repeated measures Analysis of Variance will be used to capture both week-to-week FC changes and baseline comparisons across all timepoints. Covariates of no-interest (age, gender, etc.) will be regressed out for each analysis. Post-hoc tests will be used to determine if specific time points significantly differ from each other. The MACM node-edge models will be determined using the ALE map of PTSD. The time series for each of the significant ALE clusters will be calculated and their cross-correlations will be estimated. R-to-Z transforms will be applied to the correlations which will be combined to create the AUD MACM. This MACM will be used to determine the nodes and edges for each patient at each time point. Edge quantification will be computed using structural equation modeling (SEM; AMOS, 7th edition). Nodes-and-edge models for each network will be compared using: the chi-square (c² ) statistic, RMSEA, and the Bayes information criterion (BIC).

[0231] Analyses will be intent to treat and include all participants who completed a baseline regardless of the extent of study participation. Prior to developing statistical models, variables will be examined with the use of frequency distributions, scatter plots, and histograms.504929-9580-4034v.1Statistics such as means or proportions, standard errors, ranges, and estimates of skewness and kurtosis will be computed and used as guidelines in the application analyses. The study design is an equivalence approach that will include 3 different IR-TMS montages compared to naltrexone as a standard of care (n = 50 per group; N total = 200).Results

[0232] The results of this ongoing trial are depicted in Figs. 12A-12B. Fig. 12A depicts changes in phosphatidylethanol from the end of IR-TMS treatment to 1 month in accordance with embodiments of the present invention. Fig. 12B depicts changes in phosphatidylethanol from the end of IR-TMS treatment to 6 months in accordance with embodiments of the present invention.

[0233] Participant ID is shown by color in the legend on the right, line shape reflects change in PEth from the previous assessment (solid=decrease in PEth or dashed=increase in PEth) as described. Visit identification is described on the x-axis, and PEth concentration is expressed in ng / ml as indicated (in whole blood). These data appear to show a potential emerging treatment effect during the 1 month after TMS. The slope of the change at 1 -month follow-up is -74.9 ng / ml and across 1,3, and 6 moths of follow-up is -10.7 ng / ml. This reflects the occurrence of relaspe in some participants after 1 month. Still, the majority of individual changes are negative from treatment to 3 months, which is promising. Relapse could likely be mitigated by the addition of a behavioral component to the therapy, which would be typical if this w ere to move into clinical use.

[0234] Fig. 25 depicts a dimensionality estimation for alcohol use disorder, in accordance with embodiments of the present invention.Example - Coordinate-based meta-analysis of Alzheimer’s

[0235] The Amyloid / Tau / Neurodegeneration (A / T / N) biomarker framework is used for in vivo pathological profiling of Alzheimer's disease. A binary (+ / -) label is assigned for each pathology (A, T, N) based on its presence or absence as determined by imaging or fluid biomarkers. Using imaging, neurodegeneration is confirmed by either structural MRI (indicating atrophy) or 18F- FDG PET (indicating hypometabolism), implicitly assuming that both modalities identify equivalent pathology. Preliminary evidence suggests that atrophy and hypometabolism do not spatially co- localize and are likely caused by distinct underlying pathologies. Further, while MRI is widely available, many sites lack access to PET. Recent evidence indicates that hemodynamic indices, identified via functional MRI or SPECT, may514929-9580-4034v.1be spatially concordant with hypometabolism identified by 18F- FDG- PET and able to serve as reliable proxies. Coordinate- based meta- analyses of voxel- based morphometry and voxelbased physiology reports in Alzheimer's disease, collectively analyzing 139 articles (4218 individuals) were performed to test the following hypotheses: (1) that atrophy and hypometabolism in AD are spatially dissociated; (2) that hemodynamic indices exhibit the same spatial distributions as hypometabolism identified by PET; (3) that regions of atrophy and hypometabohsm involve different functional systems and cognitive operations. Results confirmed all three hypotheses. Separation of atrophy and hypometabolism into two distinct subcategories of neurodegeneration as well as the development of regionally specific biomarkers for each are in order. See Dang et al. 2025.

[0236] Figs. 13A-13B depict node-and-edge network representations of Alzheimer’s disease (AD)-related atrophy (VBM) and hypometabolic (VBP) changes in accordance with embodiments of the present invention. Nodes represent regions of VBM and VBP alteration in AD patients compared to controls. Edges represent healthy functional connections between nodes. Edges surviving p < 0.01 thresholding are depicted with a thick line, while edges with 0.01 < p < 0.05 are depicted with a thinner line. For visualization purposes, connections are shown as undirected and indicate co-activation of the two nodes in at least one direction.

[0237] The VBM network model is comprised of 19 nodes, with all nodes connected to at least one other node. The VBP model is comprised of 14 nodes, including 3 nodes without significant connections to other nodes. L = left; R = right; Ant = anterior; Post = posterior.

[0238] Fig. 24 depicts a dimensionality estimation for Alzheimer’s disease / mild cognitive decline, in accordance with embodiments of the present invention.Example - Treating obesity as food addition via trans cranial magnetic stimulation informed by coordinate-based meta-analysisBackgroundProblem

[0239] Obesity is a global pandemic affecting > 600 million persons worldwide. US obesity prevalence is 42% and steadily rising. Obesity begins benignly, as gradual weight gain without comorbidities. As weight gain continues and body mass index (BMI) rises, a pathological cascade ensues leading to diabetes, hypertension, and dyslipidemia; these four commonly comorbid conditions are collectively termed metabolic syndrome (MetS). Persistent MetS damages many organs, leading to cardiovascular, renal, and hepatic failure and cognitive524929-9580-4034v.1impairment (“type 3 diabetes’'). The annual cost of treating the comorbidities of obesity is $200 billion in the US alone (cdc.gov).

[0240] Diet and exercise are universally advocated, no-cost, lifestyle interventions for healthy weight management. Yet, they are demonstrably ineffective for the millions of obese persons. Bariatric surgery is the most effective medical intervention for obesity and effectively ameliorates obesity-related comorbidities. Bariatric surgery, however, is costly, invasive, and medically recommended only for persons with BMI > 40 or BMI > 35 in the setting of serious obesity-related comorbidities. As well, weight regain is now known to be the norm in longterm follow up (> 2 years) after bariatric surgery. Nutritional, cognitive-behavioral and other lifestyle interventions are largely ineffective for long term weight reduction, either alone or in combination with bariatric surgery. Glucagon-like peptide- 1 receptor agonists (GLP-1 RA), the most recently developed class of weight-loss drugs, achieve weight reductions comparable to those of bariatric surgery, with concomitant ameliorations of MetS and more serious obesity comorbidities. Unfortunately, weight regain is the norm when GLP-1 RA medication is discontinued. Cost and side effects make chronic GLP-1 RA therapy untenable as a scalable solution. In view of the serious shortcomings of all current therapies, novel therapeutic approaches are desperately needed to manage the obesity epidemic. This experiment addresses this global need.Premises

[0241] Obesity as Food Addiction. A grounding premise of this proposal is that obesity is a behavioral disorder: a food addiction. Food intake is a behavior and, thereby, ultimately controlled by the brain. Feeding behaviors are motivated both by homeostatic systems, to meet metabolic demands, and by hedonic systems, seeking pleasure. Homeostatic networks respond to appetite hormones released from the GI tract (insulin, GLP-1, ghrelin) and adipose tissue (leptin). Hormonal homeostatic regulation and dysregulation is a field of study which proposal Co-PI DeFronzo has pioneered over several decades, being perhaps the field’s most prolific and impactful contributor. In the brain, homeostatic hormones are believed to operate chiefly on hypothalamus. Visceral signals (gut mechanical distention / contraction) convey the sensation of relative fullness / emptiness via the vagus nerve to hypothalamus and also to brainstem. Hedonic eating, by contrast, derives from the pleasure of eating food, and is independent of one’s energy status. Although homeostatic appetite hormones become dysregulated with obesity and lead to MetS, they do not cause obesity. Nor do visceral signals. The primary cause of excess weight gain and obesity is hedonic eating.534929-9580-4034v.1

[0242] Human functional brain mapping by task-activation (TA) imaging with positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) is a widely adopted, highly productive research domain, originally pioneered by proposal PI Fox. Proposal Co-PIs Fox and DeFronzo have collaborated for many years using PET and fMRI to investigate obesity and MetS. Through their efforts and those of numerous colleagues in this burgeoning field, PET and fMRI have provided fundamental insights into the neural systems mediating hedonic eating. Food cues — visual, olfactory, and gustatory — are powerful drivers of food cravings and hedonic consumption. In-scanner presentation of food cues robustly engages appetitive brain networks and does so more intensely and extensively in obese persons than in lean. A similar phenomenon occurs with pictorial presentation of abused substances to persons with substance use disorders. Not surprisingly, in both obesity and substance-use disorders, cue-driven craving engages brain regions that regulate reward seeking behaviors more generally, including monetary rewards. Thus, obesity, substance-use disorders (drug addictions), and behavioral addictions, such as gambling, all exhibit functional dysregulation of reward networks. Collectively, then, TA neuroimaging argues persuasively that obesity be regarded as a food addiction.

[0243] That some brain disorders exhibit atrophy patterns that recapitulate known physiological connectivity patterns was first noted by Seeley and colleagues (2009) in the context of age-related dementias. In that landmark study, gray-matter volume loss was visualized by voxel-based morphometric (VBM) analysis of structural MRI (sMRI); functional connectivity patterns were visualized by temporal covariance analysis of resting-state fMRI. Network- based spread of pathology was attributed to transneuronal propagation of degraded proteins (phosphorylated a-beta and tau). At that time, the network degeneration was thought to selectively affect higher cognitive processes, where plaques (a- beta) and tangles (tau) accumulated. Network degeneration has subsequently been demonstrated in host of brain disorders, including non-dementing neurological conditions, pyschopathologies, and developmental disorders. The most comprehensive and compelling demonstrations of network degeneration in non-dementing disorders have come from reports by proposal PI Fox, his trainees and collaborators using meta-analytic connectivity modeling, independent components analysis, graph theoretic modeling, structural equation modeling, and related advanced, multivariate, supervised and unsupervised analytics (Smith 2009; Fox 2014; Crossley 2014; Vanasse 2018; Cauda 2018). Graph theory modeling demonstrated that atrophy predominates in hubs of the connectome, leading Crossley to propose network hyperactivity and attendant544929-9580-4034v.1metabolic stress as the primary causative mechanism. For psychopathology, this postulate is consistent with Borsboonfs Network Theory of Mental Disorders (2017). Crossley's postulated was further supported by entropy modeling of disease and task networks by Vanasse (2021). Thus, the preponderance of evidence suggests that in many disorders gray- matter volume loss is network based and attributable to chronic, symptom-associated hyperactivity and attendant metabolic stress.Methods and results

[0244] Published reports of gray-matter atrophy in obesity were identified by filtered search (disease effects; obesity; gray-matter atrophy) of the BrainMap® VBM Database. Thirty-two, peer-reviewed publications reporting a total of 50 unique case-control sMRI contrasts (obese vs lean) and collectively representing 3,368 participants (1,781 obese; 1,587 lean) were identified. Searches of other literature sources (PubMed, Science Direct, Scopus, Web of Knowledge) found no other standard-compliant literature. Data were screened for exclusion criterae and duplication at the paper and experiment level. After exclusions, 23 experiments from 14 publications reported 920 metabolically healthy obese (MHO) subjects; 27 experiments from 18 publications reported 861 metabolically unhealthy obese (MUO) participants. Alteration likelihood estimation (ALE) statistical analysis (p < 0.01, FWE correction for multiple comparisons, 10,000 permutations) applied to the combined cohort (MHO + MHO) yielded the statistical parametric map illustrated in Fig. 14.

[0245] Fig. 14 depicts a meta-analytic statistical parametric map of gray -matter atrophy in obesity (vs. lean controls) in accordance with embodiments of the present invention. As shown in Fig. 14 atrophic regions engaged in hedonic (reward), executive, and salience mental operations, by comparison to normative task- activation data. MHO cohort analysis recovered exclusively cerebral alterations. MUO cohort analysis identified cerebellar alterations, confirming prior results from our group (Kotkotkowski 2018).

[0246] Network modeling of the results illustrated in Figure 15 was performed using the BrainMap® TA Database for Atrophy Functional Network (AFN) modeling, following Chiang (2021). Dense functional interconnection among cerebral nodes was confirmed, as shown in Figure 15.

[0247] Fig. 15 depicts Atrophy Functional Network modeling of network degeneration in obesity in accordance with embodiments of the present invention. Meta-analytic connectivity554929-9580-4034v.1modeling demonstrates normative functional connectivity patterns among obesity -altered brain regions. These motivate and inform network-based modulatory therapy with IR-TMS.

[0248] These results support the inference that the gray-matter volume loss in obesity is caused by chronic craving-associated hyperactivity and metabolic stress in the hedonic, executive and salience networks, as in other addiction disorders. In the work proposed, this model provides targets for network-based neuromodulation (Aim 1). It also serves as quantitative, biologically derived measure of treatment response (Aim 2).

[0249] Neuromodulation — repetitive electromagnetic stimulation of neural tissues — is a therapeutic approach with proven efficacy and safety in several neurologic and psychiatric disorders. Modification of network firing patterns is the hypothesized mechanism of action. This most likely is by disrupting maladaptive, symptom-associated, re- entrant neuronal firing patterns, thereby allowing re-emergence of physiological rhythms. Neuromodulation is most rigorously conceptualized connectomically, that is, as modifying multiregional neural circuits, rather than individual brain regions. Confirming this premise, PI Fox reported PET imaging of transcranial magnetic stimulation (TMS), showing propagation from the stimulated site along established neural pathways to engage distant network nodes (Fox 2004, 2006). As anticipated, network propagation of TMS-induced activation was in good agreement with network architecture computed meta-analytically, using the BrainMap® resources described above (Laird 2008; Narayana 2012). Furthermore, the precision with which specific networks can be targeted was enhanced by personalization, i.e., fitting an individual participant’s functional MRI data. To this end, we have developed TMS treatment planning tool (C3PlanTM ) illustrated in Figure 16. Fig. 16 depicts a personalized TMS treatment plan in accordance with embodiments of the present invention. To achieve treatment-delivery fidelity commensurate with treatment-planning precision (~ 1 mm), a robotic, coil-navigation device (C3NavTM ) was developed in-house, a combination of methods termed image-guided, robot- navigated TMS (IR-TMSTM ).

[0250] As a therapeutic proof of principle of our meta-analytic approach to TMS-treatment design, DoD / VA funding (W81XWH-02-2-0112) was obtained for a clinical trial in for posttraumatic stress disorder (PTSD) was obtained. An AFN model of network degeneration in PTSD was developed in the manner illustrated above for obesity (Fig. 15), adapted for IR-TMS treatment, and patented (US Patent 11,458,326). This PTSD-specific model was then used to target personalized IR-TMS in 119 active-duty service members in a sham-controlled, blinded, randomized (active vs sham) clinical trial (clinicaltrial.gov NCT02853032). Active 564929-9580-4034v.1IR-TMS was significantly superior to sham at end of treatment and during 3- month follow-up. Application for FDA clearance to marked IR-TMS combat-related PTSD is in preparation.

[0251] Fig. 17 depicts a dimensionality estimation for obesity, in accordance with embodiments of the present invention.Example - consensus mapping of structural brain abnormalities in addiction disorders: a coordinate-based meta-analysisBackground

[0252] Addictions are commonly dichotomized into two categories: substance-use disorders (SUD; APA, 2013) and non-substance-use disorders (Robinson & Adinoff, 2016), also termed behavioral addictions (BA). Both SUD and BA are characterized by behaviors being compulsively repeated in response to rewards, despite harmful consequences (APA, 2013). High prevalence rates of SUD and BA have led to devastating effects across personal and societal domains (SAMHSA, 2021; Sussman & Sussman, 2011). The negative impact of SUD and BA is compounded by the lack of effective interventions (Fleury et al., 2016) and high rates of relapse (Brecht et al., 2014).

[0253] SUD is recognized in the Diagnostic and Statistical Manual of Mental Disorders 3rd(DSM-3), 4th(DSM-4), and 5theditions (DSM-5; APA, 2013). In DSM-5, SUD covers 10 substance-specific diagnoses. As well. SUD is recognized in the International Classification of Disorders 10th(ICD-10) and 11theditions (ICD-11; WHO, 2018), also with numerous substance-specific diagnostic codes. Similarly, SUD is recognized by the Hierarchical Taxonomy of Psychopathology (HiTOP; Kotov et al., 2017), with substance-specific categories. Distinct scientific communities address SUD by investigating the best possible interventions for SUD (Connery et al., 2020), the pathophysiology of the disorder (Everitt, 2014), and animal models of SUD (Wu & Schulz, 2012). The advancement of neuroimaging methods now allows for non-invasive studies in humans to understand SUD-related imagingbased biomarkers (Heilig & Leggio, 2016; Heilig et al., 2021; Volkow et al., 2003).

[0254] BAs, by contrast, remain marginalized in the medical and scientific fields. In the DSM, BAs were recognized for the first time in the 5thedition (2013), with Pathological Gambling (PG) as the sole identified disorder. BAs were recognized for the first time in ICD-11 (WHO, 2018), with Internet Gaming (IG) disorder as the only behavioral-specific category. HiTOP excludes BAs, altogether. To this day, PG remains the only behavioral addiction recognized in the latest DSM (DSM-5-TR; APA, 2022), with ICD-11 recognizing both PG and574929-9580-4034v.1IG (WHO, 2025). Animal models of BAs exist but are limited (Proctor et al., 2014). To address these issues, clinician scientists have used advanced functional and structural medical-imaging modalities to study and elucidate diseases / disorders such as internet addiction (Yuan et al., 2011) long before being recognized or classified by behavioral health professionals using frameworks like the DSM or HiTOP, or by government organizations such as the WHO (Zhu et al.. 2015).

[0255] Voxel-based morphometry (VBM) is a whole-brain, voxel-wise neuroimaging method used to detect subtle but reliable disease-related brain alterations. VBM converts Tl-weighted MRI into images of gray-matter density, comparing cases to controls as statistical parametric images (SPI). Results are reported as atlas-referenced neuroanatomical coordinates of local maxima of statistically significant alterations (Wright et al., 1995; Ashbumer and Friston, 2000). VBM alterations (typically atrophy) have been reported in numerous neurological and psychiatric disorders (Vanasse et al., 2021). In neurological disorders, atrophy can be caused by a variety of underlying neuropathologies, notably including protein phosphorylation and misfolding (Tenreiro et al., 2014), inflammation (Chiang et al., 2019), and epileptic excitotoxicity (Armada-Moreira et al., 2020). In psychiatric disorders, the most likely pathophysiology is metabolic stress at high-traffic nodes associated with mental / behavioral stereotypies (Crossley et al., 2014; Cauda et al., 2018; Vanasse et al., 2021). In both neurological and psychiatric disorders, the observed alterations are strongly network based, confirming an extended scope of the network degeneration hypothesis (NDH) (Seeley et al., 2009; Crossley et al., 2014).

[0256] The first report of VBM effects in SUD appeared soon after the invention of the method (Franklin et al., 2002). The first VBM reports in BA appeared much later (Y uan et al., 2011; Zhou et al., 2011). As with all voxel -wise (mass-univariate) SPI methods, VBM requires correction for multiple comparisons. Insufficiently conservative corrections for multiple comparisons will predispose SPI analyses to false-positive results (Woo et al., 2014). The most reliable mean to overcome this limitation is via meta-analysis (Eklund et al., 2016). For coordinate-reporting literatures, coordinate-based meta-analysis (CBMA) is the most robust and widely applied approach.

[0257] CBMA uses the tabular results of a similarly themed body of published SPI reports as input to compute between-study consensus, expressed as an SPI of voxel-wise effect likelihood (Fox, Parsons, and Lancaster, 1998; Fox et al., 2014). Various CBMA algorithms have been developed and validated (Salo et al., 2023), including activation / alteration likelihood 584929-9580-4034v.1estimation (ALE; Turkeltaub et al., 2002; Eickhoff et al., 2009; Turkeltaub et al., 2012), multilevel kernel density analysis (MKDA; Wager et al., 2007), anisotropic effect-size signed differential mapping (AES-SDM; Radua et al., 2014), and seed-based d mapping with permutation of subject images (SDM-PSI; Albajes-Eizagirre et al., 2019). Of these, ALE is the most well-validated, and widely used (Acar et al., 2018; Tahmasian et al., 2019; Yeung et al., 2023). For these reasons. ALE was the method of choice for our analyses.

[0258] Several prior CBMAs have compared brain alterations in healthy controls versus SUD, BA, or both. Qin et al. (2020) analyzed 20 VBM studies (505 subjects) confined solely to BA using AES-SDM. Zhang et al. (2021) combined 36 VBM studies of SUD (1476 subjects) and 23 of BA (620 subjects), also using AES SDM. Yan et al. (2023) reported solely on SUD but combined 77 VBM studies (3457 subjects) with 39 voxel-based physiology (VBP) studies (1444 subjects), using SDM-PSI. Most recently, Mei et al. (2024) reported on BA alone, but combined 11 VBM studies (287 subjects) with 26 VBP studies (577 subjects), also using AES-SDM. Despite studying the same addiction disorders and using similar analytical methods, there is much variability in regional effects reported by these four addiction CBMAs. A consistent common inference was of reward system involvement, that emerged despite combining across disorder class (SUD & BA) and imaging modality (VBM & VBP). We note, however, that CBMAs combining VBM and VBP data have shown concordant results in some disorders (major depressive disorder, Gray et al., 2020; temporal lobe epilepsy, Towne et al., 2022), but discordant results in other disorders (Alzheimer’s disease, Dang et al., 2024). In addiction disorders, it remains to be demonstrated that VBM and VBP results are concordant.

[0259] In this study, our goal was to identify GM alterations associated with SUD and BA, both independently and jointly. To accomplish this goal, three separate CBMAs were conducted using the ALE algorithm on VBM studies. For Analysis 1, ALE was performed on SUD studies alone. In Analysis 2, ALE was performed on BA alone. In Analysis 3, ALE was implemented on the pooled (SUD + BA) VBM studies to explore shared GM alterations in SUD and BA, and their additive effects. Based on the results of previously published CBMAs, we hypothesized that: 1) SUD alone will show significant between-study spatial convergence; 2) BA alone will show significant between-study convergence; 3) comparison of SUD and BA alteration patterns will exhibit convergence (overlap); 4) SUD + BA jointly analyzed will exhibit additional regions of convergence; and 5) both SUD and BA will exhibit reward-system involvement.594929-9580-4034v.1Materials and methods

[0260] Our study protocol adhered to BrainMap Database (https: / / brainmap.org / ) data-quality standards (Fox et al., 2005, 2014), including only: 1) studies published in English-language, peer-reviewed journals; 2) studies performing whole-brain image acquisitions in all subjects; 3) study data transformed from native space into an anatomically standardized space, preferably Talairach Space or MNI space; 4) data analyzed as statistical parametric images in a whole-brain, voxel-wise (mass univariate) manner, without sub-sampling or regions-of-interest; 5) data expressed as statistical contrasts, either between condition (e.g. task-control contrasts) or between subject groups (e.g., case-control contrasts); 6) data are reported as x-y-z coordinates of statistically significant local maxima (center of mass or peak voxel) within the reference space. This study also complied with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA; Liberati et al., 2009).Data sources and searches

[0261] Standards-compliant data sets contrasting persons with addictions to healthy controls were identified using several strategies. The BrainMap VBM data sector (Vanasse et al., 2018) was searched using Sleuth (Version 3.0.4, https: / / brainmap.org / sleuth / ), as a source of relevant publications already coded for meta-analysis. The BrainMap website publications listing (https: / / brainmap.org / pubs / ) was searched for prior, standards-compliant coordinate based meta-analyses papers, which typically list all included papers. Literature repositories and search engines including PubMed, Google Scholar, ResearchGate, and ScienceDirect were also searched. Search terms applied in various combinations included: MRI, imaging, brain imaging, VBM, gray matter; and, addiction, addiction disorder, drug, drug addiction, substance-use disorder, SUD, drug addiction, polysubstance, alcohol, nicotine, smokers, smoking, stimulants, opioids, cannabis, heroin, marijuana; and, behavioral addiction, BA, pathological gambling, gambler, gambling, internet addiction, internet gaming addiction, and gaming addiction. The literature search was completed in January 2025.Study selection

[0262] VBM publications were systematically identified and reviewed for inclusion and exclusion criteria (Figure 18). Only studies that compared SUD / BA patients to healthy controls were included. Studies were included if the target population included individuals with a history of SUD or BA and an abstinence period of less than one year at the time of enrollment. Studies in which patients met criteria for comorbid medical illness (e.g., Alzheimer’s,604929-9580-4034v.1Depression, and Parkinson’s) or psychiatric comorbidities were excluded. Studies that reported either a decrease or increase in the gray matter of patients with addiction when compared to healthy controls were included in the analyses. Identifying data redundancy is crucial to avoid bias in meta-analytic findings. Thus, the data were screened for duplication at the paper and experiment level. Particular attention was paid to publications by the same authors and duplicative experiments to avoid redundancy of datasets. A list of the included studies and associated details are provided in supplementary tables SI and S2.Coordinate-data Harmonization

[0263] Data retrieved from the BrainMap VBM sector (Vanasse et al., 2018) were exported in MNI space. Data coded using Scribe (Version 3.6; https: / / brainmap.org / scribe / ) for BrainMap entry were coded in the as-published reporting space and converted into MNI space prior to analysis using the Lancaster transforms (Lancaster et al., 2007), to eliminate the spatial disparity between reference frames, thereby optimizing CBMA accuracy (Laird et al., 2010). Data Synthesis and AnalysisAnalysis 1 (Independent SUD) & Analysis 2 (Independent BA)

[0264] Independent ALE analyses were conducted for each addiction category: 1) SUD studies, including alcohol, nicotine, stimulants, opioids, and cannabis; and 2) BA studies, including pathological gambling, and internet gaming disorder (Table 10). Adhering to the recommended minimum number of experiments (23) for VBM ALE analysis (Frahm et al., 2023), 83 SUD experiments and 25 BA experiments were included in the analyses.Table 10. Demographics of Included Studies in the Meta-Analyses.Sample Size Clinical CharacteristicsPaper# Patient Control Diagnosis / Addiction (Experiment#) Subtype28 (27) 1613 1343 Alcohol SUD Total 83 (83)16 (16) 1086 1426 Nicotine20 (20) 916 780 Stimulants614929-9580-4034v.110 (11) 261 256 Opioid9 (9) 268 257 Cannabis19 (19) 592 578 Internet Gaming Disorder BA Total 25 (25) 6 (6) 182 220 Pathological Gambling108 (108) 4918 4860 Total

[0265] As shown in Table 10, A total of 83 SUD studies (alcohol, nicotine, stimulants, opioids, and cannabis) and 25 BA studies (internet gaming disorder and pathological gambling) were included for a cumulative total of 108 studies encompassing addiction. This resulted in a combined sample size of 4,918 patients and 4,860 controls. Abbreviations: BA, behavioral addiction; SUD, substance use disorder.

[0266] The CBMA was performed using the ALE algorithm implemented in the GingerALE software (Version 3.0.2, https: / / www.brainmap.org / ale / ) (Eickhoff et al., 2012; Eickhoff et al., 2009; Eickhoff et al., 2016; Laird et al., 2009; Turkeltaub et al., 2002; Turkeltaub et al., 2012). The ALE algorithm treats each x-y-z coordinate as a Gaussian probability distribution to account for spatial uncertainty, weighted by sample size (subjects per study), to form modeled alteration maps (MA maps) per-experiment; ALE then computes a union of probabilities across experiments as ALE values. Significance is determined relative to a spatially random null distribution.

[0267] Correction for family-wise error rate at the cluster-level (cFWE) was the approach of choice (Frahm et al., 2023). The selected method for the present study, cluster-level inference, generates a simulated data set of randomly distributed foci based on characteristics of the input data set for testing the null hypothesis. Results were thresholded for significance using cluster-level inference of p<0.05, with a modified cluster-forming threshold of p<0.01 due to sample size limitations (Moring et al., 2022; Muller et al., 2018; Tahmasian et al., 2019; Yeung, 2023). GingerALE (Version 3.0.2, https: / / www.brainmap.org / ale / ) produces output files for each ALE analysis, including an ALE map image (.nii) and a peaks spreadsheet (.xls). The peaks spreadsheet summarizes the analysis results, listing each cluster number, local ALE 624929-9580-4034v.1maxima (x, y, z coordinates in MNI space), the maximum ALE value, and a label of the nearest gray matter region within 5 mm (Anatomic Region). Mango software (Version 4.1, https: / / ric.uthscsa.edu / mango / ) was used to view ALE map results, determine the volume of each cluster, and to identify overlapping regions between the ALE maps of SUD and BA.Analysis 3: Pooled SUD + BA

[0268] The third meta-analysis was performed to identify a pattern of GM alteration across all addiction disorders (SUD and BA combined) using the ALE algorithm implemented in the GingerALE software (Version 3.0.2, https: / / www.brainmap.org / ale / ) (Turkeltaub et al., 2002; Eickhoff et al., 2009-2012-2016; Laird et al., 2009). This analysis consisted of 108 experiments, far exceeding the recommended minimum (Frahm et al., 2023). The same thresholds previously used were applied: the ALE algorithm was used with a cluster-level family -wise error (FWE) threshold of 0.05 and p<0.01 to correct for multiple comparisons (Eickhoff et al., 2009). GingerALE (Version 3.0.2, https: / / wTvw.brainmap.org / ale / ) produces output files for each ALE analysis, including an ALE map image (.nii) and a peaks spreadsheet (.xls). The peaks spreadsheet summarizes the analysis results, listing each cluster number, local ALE maxima (x. y, z coordinates in MNI space), the maximum ALE value, and a label of the nearest gray matter region within 5 mm (Anatomic Region). Mango software (Version 4.1, https: / / ric.uthscsa.edu / mango / ) was used to determine the volume of each cluster and to visualize ALE results by overlaying them onto a standardized brain template (T1 weighted image in MNI space).Noise Simulation for Estimation of File-Drawer Effect

[0269] Despite CBMA robustness, publication bias is a threat of validity to many meta-analytic studies that focus on finding patterns from published studies. Meta-analyses relying on effect size (e.g., using Cohen’s d or similar) are highly susceptible to missing data (filedrawer effect), while effect-location meta-analyses (Fox et al., 1998), as performed by ALE, are minimally susceptible to missing data. The fail-safe N (FSN; Rosenthal, 1979) method was introduced to correct for missing data in effect-size meta-analysis. Acar et al. (2018) proposed an ALE adaptation of the FSN method, which assesses the impact of null findings on the observed results, by introducing noise into the ALE algorithm. A modified version of the FSN method was implemented, which introduced 6% noise (Samartsidis et al., 2020). Each identified cluster was retested in the meta-analyses with added noise (null findings) to assess robustness against potentially unpublished findings. If a cluster survived the initial 6% noise, increased noise w as introduced up to 30% or until failure (Gray et al., 2020). In general, clusters 634929-9580-4034v.1that were not replicated when adding noise indicate that these clusters are less robust against the simulation.Spatial Cross Correlation

[0270] Similarity among the independent ALE statistical maps of SUD and BA was quantified using spatial cross-correlation analysis with FSL’s fslcc tool (Smith et al., 2004). The computed correlation coefficient of similarity between the two ALE maps ranges from -1.0 to +1.0 with higher coefficients indicating greater degree of similar! ty / dissimilarity.Percent Contribution Analysis

[0271] A percent contribution analysis was performed for the pooled SUD + BA analysis to confirm that each cluster had contribution from both SUD and BA studies. Briefly, the number of foci that contributed to each cluster from each study (SUD or BA) was quantified and a percent contribution of SUD and BA studies was calculated for each cluster.Network Analysis

[0272] Network analysis was performed by manually checking the overlap of ALE maxima results with functional connectivity (FC) independent component analysis (ICA) network masks (Vanasse et al., 2021) using the Mango software (Version 4.1, https: / / ric.uthscsa.edu / mango / ). Briefly, ALE results of each analysis and the FC ICA network masks from Vanasse et al. (2021) were overlayed onto a sample brain image (Talairach Colin27) using Mango; and the network that each ALE maxima most strongly overlapped with (based on z-score) was documented. A z-score of 3.0 or greater is considered statistically significant.Paradigm Class Analysis

[0273] Paradigm class analyses were completed using Paradigm Class plugins (Version 4.1, https: / / ric.uthscsa.edu / mango / ) in Mango software (115 paradigm classes), which reference task-activation studies in healthy controls from the BrainMap database (22,303 experiments; 112,160 subjects). The quantitative regional behavior analysis was performed on masks created from selected brain ROIs that are anatomically altered in addiction, to highlight how these regions relate to healthy brain functions (Crossley et al., 2014; Kotkowski et al., 2019; Smith et al., 2009). This is accomplished by comparing masks of disease states to a spatially uniform random distribution derived from a collection of task-based studies in healthy controls from the BrainMap database (Lancaster et al., 2012; Kotkowski et al.. 2019). ALE map images of each analysis were loaded onto Mango (Version 4.1,644929-9580-4034v.1https: / / ric.uthscsa.edu / mango / ) and Paradigm Class plugins were used to acquire whole image (Figure 21A-C) and cluster specific (Figures 22A-B and 23A-F) paradigm class analyses results. Paradigm classes with the operative threshold z-score of 3.0 or greater, as comparable to a group p-value of 0.05, are reported as statistically significant.ResultsStudy Inclusion and Characteristics

[0274] A total of 108 individual VBM experiments (108 papers) comprising results from 4,918 patients and 4,860 healthy controls were identified for inclusion in this meta-analysis (Table 10). SUD (83 experiments / papers) includes substances such as alcohol (1,613 patients), nicotine (1,086 patients), stimulants (916 patients), opioids (261 patients), and cannabis (268 patients) (Table 10). BA (25 experiments / papers) encompasses internet gaming disorder (592 patients) and pathological gambling (182 patients) (Table 10). Supplementary tables SI and S2 in the online supplement list all included studies and characteristics of the included subjects. ALE AnalysesAnalysis 1: Independent SUD

[0275] The SUD meta-analysis comprised a total of 650 coordinates (x-y-z reported as foci) created from summed results from 83 experiments. These per-experiment regional effects were reduced by ALE to four clusters (Table 12), each containing at least one ALE maximum. Patterns of convergence were found in several regions including, the right thalamus (medial dorsal nucleus), left caudate head, bilateral medial frontal gyrus (MFG) (Brodmann 6, 9, 10, and 11), left cingulate gyrus (Brodmann 32), left superior frontal gyrus (SFG) (Brodmann 6), and bilateral anterior cingulate cortex (ACC) (Brodmann 32) (Table 12 and Figure 19).Table 12. ALE Clusters. Meta-Analyses Results of SUD and BA Independently and the Overlap Region.654929-9580-4034v.1Cluster# Volume Side Anatomic Maximum MNI FSN Network Z-score (mm3) Region ALE coordinates of (%) (network) value local maxima(x IO3) X y zSUD1 4784 R Thalamus, 29.03 6 -10 8 FSN> Thalamus 24.5 Medial 30% (IC-6)DorsalNucleusL Caudate, 21.44 -2 2 0 Fronto- 10.4 Caudate Striatal (IC- Head 14)2 3992 L Cingulate 24.13 0 22 36 FSN> Cognitive 74.2 Gyrus, 30% Control Brodmann (IC-17)32R Medial 21.69 4 20 46 Cognitive 47.5 Frontal ControlGyrus, (IC-17)Brodmann6L Cingulate 21.37 2 32 28 Cognitive 30.5 Gyrus, Control Brodmann (IC-17)32L Medial 16.79 -2 44 26 Default16.7 Frontal mode (IC- Gyrus, 13)Brodmann9L Superior 16.70 -2 16 60 Visuomotor 12.1 Frontal Coordination Gyrus, 1 (IC-5) Brodmann63 3032 R Anterior 32.22 6 48 _2 FSN> Default16.3 Cingulate, 30% mode (IC- Brodmann 13)32R Anterior 21.23 4 44 12 Default17.6 Cingulate, mode (IC- Brodmann 13)324 2824 R Anterior 19.61 2 36 -14 FSN< Default6.1 Cingulate, 6% mode (IC- Brodmann 13)32664929-9580-4034v.1L Anterior 19.18 -6 44 -12 Default9.5 Cingulate, mode (IC- Brodmann 13)32R Anterior 1857 8 20 -14 Default1 8 Cingulate, mode (IC- Brodmann 13)32L Medial 15.94 0 50 -16 Default3.7 Frontal mode (IC- Gyrus, 13) Brodmann10R Medial 15.14 8 36 -20 Tempero- 0.8 Frontal Limbic (IC- Gyrus, 18) Brodmann11L Medial 14.61 -2 46 -18 Default3.7 Frontal mode (IC- Gyrus, 13) Brodmann10BA1 3464 L Anterior 1641 -10 30 20 FSN> Cognitive 125Cingulate, 30% Control Brodmann (IC-17)32R Cingulate 12.30 4 18 26 Cognitive 34.8 Gyrus, Control Brodmann (IC-17)24L Caudate, 11.01 -10 8 4 Fronto- 31.0 Caudate Striatal (IC- Body 14)L Caudate, 10.51 -8 20 0 Fronto- 8.3 Caudate Striatal (IC- Head 14)L Anterior 09.91 -4 22 16 Cognitive 9.1 Cingulate, Control Brodmann (IC-17)33L Lentiform 09.49 -18 4 6 Fronto- 32.8 Nucleus, Striatal (IC- Putamen 14) Overlap1 56 L Caudate, 12.16 -6 8 2 Fronto- 21.5Caudate Striatal (IC-Head 14)674929-9580-4034v.1

[0276] As shown in Table 12, Convergent findings of localized gray matter alteration in addiction were demonstrated as clusters. In SUD. ALE results demonstrated a convergent pattern of gray matter alteration in 4 clusters, while BA demonstrated convergence in only one cluster. Each cluster contained one or more local ALE maxima (x, y, z, MNI coordinates along with a maximum ALE value) in distinct anatomic regions (MNI space). The volume of each cluster is reported in mm3 and the location (left or right side of brain) of each local ALE maxima is also reported. The one region of overlap (left caudate head) between SUD and BA independent ALE maps is also reported in the same format. FSN percentage of additional noise that must be added to each independent ALE analysis to result in failure of convergence for each identified cluster is reported. Network associations and their strength (z-score) are also presented. Abbreviations: ALE. activation / alteration likelihood estimation; BA, behavioral addiction; FSN, fail-safe N; L, left; MNI, Montreal Neurological Institute; R, right; SUD, substance use disorder.Analysis 2: Independent BA

[0277] The BA meta-analysis comprised a total of 127 coordinates (foci) created from summed results of 25 experiments. Compared to SUD, ALE maxima for BA had less identified regions. ALE results demonstrated, convergent abnormality in only one cluster (Table 12) which included local maxima in the: left ACC (Brodmann 32 and 33), right cingulate gy rus (Brodmann 24), left caudate body and head, and left lentiform nucleus (putamen) (Table 12 and Figure 19). The caudate head was the only region of overlap between the independent SUD and BA ALE maps that was identified using the overlay logical analysis tool in Mango software (Table 12 and Figure 19).Analysis 3: Pooled SUD + BA

[0278] The pooled (SUD + BA) addiction meta-analysis showed localized GM alterations among cortical and subcortical structures. The meta-analysis comprised a total of 777 foci groups created from summed results of all 108 experiments. In addition to regions found in the two independent meta-analyses, the pooled analysis yielded many (n = 24) alteration foci not identified in either independent CBMA. A percent contribution analysis confirmed that all additional foci were contributed to by both conditions (SUD and BA) and are thus termed '“interactive” (Table 13). These interactive areas include the right and left insula (Brodmann 13), bilateral claustrum, left SFG (Brodmann 10), right inferior frontal gyrus (Brodmann 13), bilateral lentiform nucleus (lateral globus pallidus), bilateral parahippocampal gyrus684929-9580-4034v.1(amygdala), right parahippocampal gyrus (hippocampus), left sub-gyral middle temporal gyrus (Brodmann 21), and left middle temporal gyrus (Brodmann 21) (Table 13 and Figure 20). Table 13. ALE Clusters. Results of Pooled SUD + BA Meta-Analysis.Cluste % Volum Inter Anato Maximu MNI coordinates of FSN Networ z r# Cont. e Sid mic m local maxima (%) k(SUD (mm3) activ e Region ALE y z / BA) e valueRegio (x IO3)n1 80 / 20 10216 R Anterio 32.24 6 48 -2 FSN Default16.r mode 3 Cingula 30% (IC-13) te,Brodmann 32* L Superio 32.15 -8 60 -16 Default0.4 r mode Frontal (IC-13) Gyrus,Brodmann 10L Anterio 2469 -10 30 20 Cogniti 12 r ve 5 Cingula Control te, (IC-17) Brodmann 32L Cingula 24.42 0 22 36 Cogniti 74.te ve 2 Gyrus, Control Brodma (IC-17) nn 32❖ R Cingula 23.56 4 30 34 Cogniti 40.te ve 9 Gyrus, Control Brodma (IC-17) nn 32I, Cingula 2225 2 32 28 Cogniti 30 te ve 5 Gyrus, Control Brodma (IC-17) nn 32R Medial 21.70 4 20 46 Cogniti 47.Frontal ve 5 Gyrus, Control Brodma (IC-17) nn 6L Medial 21.58 -12 62 -2 Default2.6 Frontal mode Gyrus, (IC-13) Brodmann lOR Anterio 21.24 4 44 12 Default17.r mode 6 Cingula (IC-13)te,694929-9580-4034v.1Brodmann 32* L Medial 19.81 -4 50 26 Default20.Frontal mode 6 Gyrus, (IC-13) Brodmann 92 81 / 19 5208 * R Inferior 32.20 32 4 -18 FSN Temper 12.Frontal >30 0- 9 Gyrus, % Limbic Brodma (IC-18) nnl3* R Lentifor 21.15 18 6 -16 Temper 15.m 0- 6 Nucleus Limbic (IC-18) Putamen* R Lentifor 20.10 26 -16 -8 Temper 22.m 0- 3 Nucleus Limbic, Lateral (IC-18) GlobusPallidus* R Parahip 17.42 34 -8 -18 Temper 12.pocamp 0- 8 al Limbic Gyrus, (IC-18) Amygdala* R Lentifor 17 18 36 -14 -6 Temper 10 m 0- 2 Nucleus Limbic (IC-18) PutamenR Parahip 13.91 38 -14 -28 Temper 3.3 pocamp 0- al Limbic Gyrus, (IC-18) Hippocampus3 83 / 17 4704 * L Insula, 24.92 -44 -8 -2 FSN Auditor 15.Brodma <6% y (ic-7) 8 nn l3* L Parahip 23.04 -34 -8 -20 Temper 12.pocamp 0- 8 al Limbic Gyrus, (IC-18) Amygdala* L Lentifor 21.81 -26 -14 -8 Temper 28.m 0- 9 Nucleus Limbic, Lateral(IC-18) GlobusPallidus704929-9580-4034v.1❖ L Lcntifor 21.19 -28 -4 6 Somcst 21.m hesis 0 Nucleus (IC-2) Putamen❖ L Middle 16.75 -56 -2 -14 Default6.6Temper mode al (IC-13) Gyrus,Brodmann 21* L Claustr 16.39 -36 -12 2 Somest 12.um hesis 3(IC-2) * L Sub- 14.54 -40 -14 -14 Temper 4.9Gyral 0- Middle Limbic Temper(IC-18) alGyrus,Brodmann21* L Claustr 13.64 -40 -20 0 Auditor 14.um y (ic-7) 2 4 95 / 5 4624 R Thalam 29.03 6 -10 8 FSN Thalam 24.us, >30 us (IC- 5 Medial % 6) DorsalNucleusL Caudate 24.19 -6 8 2 Fronto- 21.Striatal 5 Caudate (IC-14) Head5 89 / 11 3176 * L Anterio 24.47 0 36 -14 FSN Default6.8 r <6% mode Cingula (IC-13) te,Brodmann 32* L Anterio 22.02 -6 42 -12 Default10.r mode 0 Cingula (IC-13) te,Brodmann 32R Anterio 19.00 8 20 -14 Default1.8 r mode Cingula (IC-13) to,Brodmann 32L Medial 1608 0 50 -16 Default37 Frontal mode Gyrus, (IC-13) Brodmann 10714929-9580-4034v.16 85 / 15 2536 ❖ R Insula, 24.01 38 6 8 FSN Somest 29.Brodma <6% hesis 3 nn 13 (IC-2) * R Claustr 20.30 34 14 -2 Salienc 25.um e (IC-4) 8 * R Insula, 1647 36 22 0 Salienc 31Brodma e (IC-4) 5 nn 13❖ R Insula, 16.25 46 10 -4 Ventral 14.Brodma Attentio 5 nn 13 n(IC-16)

[0279] As show n in Table 13, convergent findings of localized gray matter alteration in addiction were demonstrated as clusters. For the combined SUD + BA analysis, ALE results demonstrated a convergent pattern of gray matter alteration in 6 clusters. Each cluster contained one or more local ALE maxima (x, y, z, MNI coordinates along with a maximum ALE value) in distinct anatomic regions (MNI space). The volume of each cluster is reported in mm3 and the location (left or right side of brain) of each local ALE maxima is also reported. In addition to the regions found in the two independent meta-analyses, more regional effects were demonstrated in the pooled meta-analysis. These Interactive Regions (*) are regions not present in either independent analysis. The percent contribution of SUD and BA to each cluster is also provided. FSN percentage of additional noise that must be added to the pooled SUD + BA ALE analysis to result in failure of convergence for each identified cluster is reported. Network associations and their strength (z-score) are also presented.Abbreviations: ALE, activation / alteration likelihood estimation; BA. behavioral addiction; FSN, fail-safe N; L, left; MNI, Montreal Neurological Institute; R, right; SUD, substance use disorder.Noise Simulation

[0280] From the independent ALE analyses, SUD clusters 1, 2, and 3 survived 30% added noise, while cluster 4 did not survive 6% noise (Table 12). BA consisted of only one cluster which survived 30% added noise (Table 12). From the pooled ALE analysis (SUD + BA), clusters 1, 2, and 4 survived 30% noise, while clusters 3, 5, and 6 did not survive 6% noise (Table 13).Spatial Cross Correlation

[0281] The correlation between SUD and BA independent ALE maps was low (r = 0.0045).724929-9580-4034v.1Percent Contribution Analysis

[0282] From the pooled SUD + BA analysis, Cluster 1 had an 80% contribution from SUD studies and a 20% contribution from BA studies (Table 13). Cluster 2 had an 81% contribution from SUD studies and 19% contribution from BA studies (Table 13). Cluster 3 had an 83% contribution from SUD studies and 17% contribution from BA studies (Table 13). Cluster 4 had a 95% contribution from SUD studies and a 5% contribution from BA studies (Table 13). Cluster 5 had an 89% contribution from SUD studies and an 11% contribution from BA studies (Table 13). Cluster 6 had an 85% contribution from SUD studies and a 15% contribution from BA studies (Table 13). Results from this analysis confirmed that each cluster from the pooled SUD + BA analysis had contribution from both SUD and BA, confirming the interactive effects not seen in the independent analyses (Table 13). The larger percent contribution from SUD studies is likely due to the 3.3-fold more included studies.Network Analysis

[0283] Each cluster was observed to significantly overlap (z > 3) with at least one canonical FC network. From the independent SUD analysis, Cluster 1 was most associated with the Thalamus (IC-6) and the Fronto-Striatal (IC-14) networks (z = 24.5 and 10.4 respectively) (Table 12). SUD cluster 2 was most associated with Cognitive Control (IC-17), Default-mode (1C-13), and Visuomotor Coordination 1 (1C-5) networks (z = 74.2, 16.7, and 12.1 respectively) (Table 12). SUD clusters 3 and 4 were solely associated with the Default-mode (IC-13) network (z = 16.3 and 6.1 respectively) (Table 12). From the independent BA analysis, cluster 1 was associated with both the Cognitive Control (IC-17) and Fronto-Striatal (IC-14) networks (z = 12.5 and 31 respectively) (Table 12). The only region of overlap between the two independent analyses (caudate head) was associated with the Fronto-Striatal (IC-14) network (z = 21.5) (Table 12). From the SUD + BA pooled analysis, cluster 1 was most associated with the Default-mode (IC-13) and Cognitive Control (IC-17) networks (z = 16.3 and 12.5 respectively) (Table 13). Cluster 2 was solely associated with the Tempero-Limbic (IC-18) network (z = 12.9) (Table 13). Cluster 3 was associated with Auditory (IC-7), Tempero-Limbic (IC-18), Somesthesis (IC-2), and Default-mode (IC-13) networks (z = 15.8, 12.8, 21, and 6.6 respectively) (Table 13). Cluster 4 was associated with the Thalamus (IC-6) and Fronto-Striatal (IC-14) networks (z = 24.5 and 21.5 respectively) (Table 13). Cluster 5 was only associated with the Default-mode (IC-13) network (z = 6.8) (Table 13); while Cluster 6 was associated with the Somesthesis (IC-2), Salience (IC-4), and Ventral Attention (IC-16) networks (z = 29.3, 25.8, and 14.5 respectively) (Table 13).734929-9580-4034v.1Paradigm Class Analysis

[0284] Regional-alteration clusters spatially associated with a task-activation behavioral paradigm are reported based on a z-score > 3 (Figures 21A-C, 22A-B, 23A-F). Whole image regions exhibiting GM alteration from independent SUD meta-analysis (n = 83) and BA metaanalysis (n = 25) were significantly associated with reward (z = 9.08 and 3.83 respectively) (Figure 21A-C). This association with reward was also found in regions exhibiting GM alteration from the pooled SUD + BA meta-analysis (n = 108) (z = 9.68) (Figure 21 A-C). Other significant categories of association are shown in Figure 21A-C. All four SUD-specific clusters, 1 (z = 4.23), 2 (z = 5.28), 3 (z = 4.99), and 4 (z = 3.31) were significantly associated with reward (Figure 22A-B). For the pooled SUD + BA ALE analysis, clusters 1 (z = 6.92), 4 (z = 4.71), and 6 (z = 4.15) were significantly associated with reward (Figure 23A-F). Additional paradigm classes that were associated with altered GM regions are illustrated in Figures 22A-B and 23A-F.Discussion

[0285] Our findings confirmed all hypotheses: 1) SUD alone showed significant between-study convergence; 2) BA alone also showed significant between-study convergence; 3) SUD-alone and BA-alone alteration patterns showed overlap (minimal); 4) SUD + BA jointly analyzed exhibited additional regions of interactive convergence; and 5) both SUD and BA exhibited reward-system involvement. Although independent ALE analyses of SUD and BA only demonstrated overlap in the caudate head, additional anatomical regions (n = 24 / 34), not previously identified in either independent meta-analysis, were discovered in the pooled meta-analysis (interactive). Moreover. SUD alone presented with more identified foci (n = 15) when compared to BA alone (n = 6), likely due to SUD having 3.3 times more included studies. In all three meta-analyses the altered regions were loaded most heavily on reward-paradigm activation patterns, as determined by the paradigm class analyses. While there was low spatial cross correlation (r = 0.0045) between independent SUD and BA ALE maps, the pooled SUD + BA analysis demonstrated that addiction disorders (SUD and BA) share a similar pattern of gray-matter alteration, while consistently being implicated in reward system dysregulation based on Paradigm Class Analyses results.Identified Regions in Independent Meta-Analyses

[0286] Two independent ALE analyses were conducted based on the two categories of addiction: 1) SUD studies (including alcohol, nicotine, stimulants, opioids, and cannabis) and744929-9580-4034v.12) BA studies (including pathological gambling and internet gaming disorder). The convergent regions of SUD-related GM volume pathology included the right thalamus (medial dorsal nucleus), left caudate head, bilateral MFG (Brodmann 6, 9, 10 and 11), left cingulate gyrus (Brodmann 32), left SFG (Brodmann 6), and bilateral ACC (Brodmann 32) (Table 12 and Figure 19). Convergent regions of GM volume pathology in BA included the left ACC (Brodmann 32 and 33), right cingulate gyrus (Brodmann 24). left caudate body and head, and the left putamen (Table 12 and Figure 19). The cingulate gyms, including the ACC (Brodmann 32), was a convergent region in both independent analyses, although their thresholded clusters did not overlap.

[0287] Separation of data into categories of addiction revealed distinct neuroanatomical GM changes between SUD and BA, with direct overlap only within the caudate nucleus (caudate head). The caudate nucleus is a brain region involved in reward processing, learning, and motor control (Robinson et al., 2012). This was further confirmed in the network analysis, as the caudate head showed a strong association (z = 21.5) with the Fronto-Striatal (IC-14) network, which has been shown to be heavily loaded on behaviors such as positive emotions (reward) and motor learning (Vanasse et al., 2021). Moreover, the caudate is intricately linked to pleasure and motivation, which can, at least partially, explain the compulsive and repetitive cycle of maladaptive behaviors (Haber & Knutson, 2010; Watanabe & Hikosaka, 2005). These morphological effects of the rew ard circuitry appear to be invariant across both SUD and BA, and may be an appropriate target for neuromodulatoiy therapies, including deep brain stimulation (DBS), transcranial magnetic stimulation (TMS), and transcranial direct cunent stimulation (tDCS). However, alterations within the caudate alone, do not fully explain SUD or BA pathology. From our findings, it appears that other structural alterations are significantly associated with both SUD and BA.

[0288] Discrepant regions were found between SUD and BA when studied independently. The specific regions that were associated with SUD, and not BA, included the MFG (Brodmann 6, 9, 10, and 11), SFG (Brodmann 6), and the medial dorsal nucleus of the thalamus. These regions are associated with higher cognitive functions, such as decision making, impulsivity, and emotion regulation (Perry et al., 2023; Euston et al., 2012; El-Baba & Schury, 2023). Furthermore, these critical functions are proposed to be central to the development and maintenance of addiction (Koob & Volkow, 2010; Volkow & Baler, 2014). This association was evident from the network analysis as the MFG (Brodmann 6. 9, 10, and 11) demonstrated a strong association with the Cognitive Control (IC-17), Tempero-Limbic (IC-18), and Default-754929-9580-4034v.1mode (IC-13) networks, which are involved in memory, cognition, and emotion (Vanasse et al., 2021). Moreover, impaired decision-making may serve to inhibit abilities of individuals to distinguish between adaptive and maladaptive behavioral options, as well as the ability to consider short- and long-term consequences of actions (Bechara, 2005; Bolla et al., 2005; Zhang & Volkow, 2019). Simultaneously, changes in dopamine signaling can increase the desire for drugs and contribute to impulsive drug-seeking behavior (Everitt & Robbins, 2016). Emotional dysregulation can further impair decision-making abilities and increase impulsivity, leading to continued drug use, despite negative consequences (Fox et al., 2008). Understanding these interactions and their underlying neurobiological mechanisms is important for the development of interventions to target the underlying dysregulation related to decision-making, impulsivity, and emotion regulation.

[0289] Alternatively, the regions associated with BA, and not SUD, included the caudate body and putamen. Results of the whole image BA (Cluster 1) paradigm class analysis indicated that these regions are significantly associated with reward. Additionally, results from the network analysis reinforced this association with reward, as the caudate body and putamen were both associated with the Fronto-Striatal (IC-14) network, shown to be strongly associated with reward (Vanasse et al., 2021). These findings indicate that the neurobiological effects related to reward may serve as maintenance factors of BA due to the reinforcing effects of addictive behavior, similar to the underlying effect in SUD (Koob & Volkow, 2016). Behavioral approaches for BA, such as differential reinforcement strategies, may be especially warranted for BA.Identified Regions in Pooled SUD + BA Meta-Analysis

[0290] The pooled meta-analysis indicated consistent disease-related effects of addiction in GM across various addiction types (SUD and BA). The convergent regions of GM volume pathology in the pooled analyses were predominately (>70%) interactive regions (n = 24 / 34) between SUD and BA, and consisted of the bilateral insula (Brodmann 13), bilateral claustrum, left SFG(Brodmann 10), right inferior frontal gyrus (Brodmann 13), bilateral lentiform nucleus (lateral globus pallidus), bilateral parahippocampal gyrus (amygdala), right parahippocampal gyrus (hippocampus), left sub-gyral middle temporal gy rus (Brodmann 21), and left middle temporal gyrus (Brodmann 21) (Table 13 and Figure 20). Non-interactive regions (also present in the independent SUD / BA analyses) included the bilateral ACC (Brodmann 32), left cingulate gyrus (Brodmann 32), right MFG (Brodmann 6), left MFG (Brodmann 10), right thalamus (medial dorsal nucleus), and left caudate head. These regions are located within764929-9580-4034v.1canonical FC networks, including the default mode network (DMN). the executive control network (ECN), salience network (SN). and the limbic networks (Laird et al., 2011; Smith et al., 2009). The ECN is also referred to as the Cognitive Control (IC-17) network (Vanasse et al., 2021). Observed alterations in numerous neurological and psychiatric disorders have been shown to be strongly network based (Crossley et al., 2014). Therefore, we aim to relate the anatomical alterations seen in our study to the canonical networks to which they belong (Buckner & Krienen, 2013; Moring et al., 2022).

[0291] Significant interactions between SUD and BA regional alterations were found in the ACC (Brodmann 32), MFG (Brodmann 9 and 10), MTG (Brodmann 21), and cingulate gyms (Brodmann 32); all of winch are associated with the DMN (Camchong et al., 2019; Damai et al., 2019; Kuo et al., 2019; Li et al., 2015). This association was also confirmed by the network analysis, as clusters 1 and 5 from the pooled SUD + BA analysis had a strong association with the Default-mode (1C-13) network. The DMN is a large-scale functional brain network that exhibits high activation during rest and less activation during task performance (Laird et al., 2009). Beyond SUD and BA (Uddin et al., 2009; Wang et al., 2017; Zhang & Volkow. 2019), the DMN is implicated in multiple psychiatric disorders, such as depression, anxiety, and posttraumatic stress disorder (Mohan et al., 2016; Zhang & Volkow, 2019).

[0292] The DMN also strongly interacts with subcortical areas and other large-scale networks, most prominently with the executive control network (ECN) and the salience network (SN) (Zhang & Volkow^, 2019). This interaction was evident in the network analysis, as cluster 1 of the SUD + BA analysis was strongly associated with both the Default-mode (IC-13) network and the Cognitive Control (IC-17) networks. Reduced efficiency of ECN and DMN reflect diminished cognitive control and self-monitoring (Sridharan et al.. 2008; Sutherland et al., 2012). Lack of cognitive control may be indicative of cognitive fusion, in which individuals lack discernment between internal thought processes and behavioral responses (Xiong et al., 2021).

[0293] Other interactive effects of SUD and BA were found within the ACC (Brodmann 32), MFG (Brodmann 6), SFG (Brodmann 10), IFG (Brodmann 13), medial dorsal nucleus of the thalamus, and insula (Brodmann 13), which are associated with the ECN. This was confirmed by results from the network analysis, as cluster 1 from the pooled SUD + BA analysis had a strong association with the Cognitive Control (IC-17) network. The ECN is implicated in addiction and is responsible for various higher cognitive functions (Fox et al., 2006; Jentsch & Taylor, 1999; Menon, 2011). Overall, these regions are associated with 774929-9580-4034v.1regulating limbic reward regions and higher-order executive function; as well as working memory, decision-making, and problem-solving in goal-directed behavior (Damoiseaux et al., 2006; Goldstein et al., 2009; Menon, 2011). Thus, urges to engage in maladaptive behavior, lack of insight, disregard for long-term consequences, and other common symptoms could be associated with dysregulation in these regions (Ashare et al., 2013; Cogdell-Brooke et al., 2020; Le et al., 2021; Loughead et al., 2009; Crockford et al., 2005; Goudriaan et al., 2010).

[0294] The SN (including the insula, cingulate gyrus, IFG, thalamus, lateral globus pallidus, claustrum, and ACC) is crucial for segregating the most relevant stimuli to guide behavior, and for mediating dynamic activity between the DMN and ECN to direct attention toward internal or external events (Fedota et al., 2018; Li et al., 2017; Menon, 2011; Menon & Uddin, 2010; Seeley et al., 2007; Sridharan et al., 2008). Based on the network analysis, the claustrum and insula (Brodmann 13) do have a very’ strong association (z = 25.8 and 31.5 respectively) with the Salience (1C-4) network. In addition, the insula and ACC are functionally connected with other regions, including subregions of the PFC, inferior parietal lobe (Oosterwijk et al., 2012; Peters et al., 2016; Seeley et al., 2007; Williams, 2016), subcortical regions of the extended amygdala, and the ventral striatum (Sylvester et al., 2012). Based on this knowledge, the possibility exists that an individuals’ inability to regulate attention and emotion, due to dysfunction in the SN, may lead to increased focus on internal events, such as drug cravings, rather than the logical consequences of maladaptive behavior. This hypothesis is consistent with previous studies that suggest the insula and ACC are crucial for psychological well-being and adaptive functioning (Goodkind et al., 2015; McTeague et al., 2017; Sha et al., 2019; Zhang et al., 2021; Zhao et al., 2021).

[0295] Our study identified alterations in the amygdala, lateral globus pallidus, hippocampus, putamen, IFG (Brodmann 13), and MTG (Brodmann 21) indicating dysfunction in the limbic and reward networks. The association of these regions with the limbic system was confirmed through the network analysis, as clusters 2 and 3 from the pooled SUD + BA analysis showed a strong association with the Tempero-Limbic (IC-18) network. The limbic system, which includes the amygdala, is responsible for processing emotions, motivation, and memory, all of which play7a significant role in addiction (Koob & Volkow, 2010). The reward system, which includes the ventral striatum (putamen and caudate head) and the prefrontal cortex (ACC, SFG, MFG, and IFG), is responsible for reinforcing behavior associated with positive outcomes (Koob & Volkow. 2010). The amygdala is a part of the reward network through its connections to the frontal cortex, nucleus accumbens, and hippocampus (Lesage & Stein,784929-9580-4034v.12016). These networks are highly interconnected and work together to reinforce addiction-related behavior (Goldstein & Volkow, 2011). Alterations in these networks may contribute to the persistent drug-seeking behavior and difficulty in abstaining from drugs or addictive behaviors seen in individuals with addiction. Morphological changes within these regions may also represent an individuals’ increased attention toward the rewarding internal or social effects of the problematic substance or behavior, and a decreased motivation toward other goal-directed behavior. Cognitive and behavioral therapies may help to clarify values, bring saliency to important and relevant goals, and alter associations with substances or maladaptive behaviors.

[0296] Overall, these results may suggest a shared underlying mechanism for SUD and BA, emphasizing the role of common neural structural biomarkers across addiction-related disorders. Furthermore, the network associated alterations seen in our study may be similar to what has been reported in neurodegenerative disorders associated with aging, specifically the NDH (Seeley et al., 2009). The network-based degeneration phenomenon has also been reported in many psychiatric and neurologic disorders, with evidence pointing to hubs of the human connectome being the target of degeneration (Crossley et al., 2014). Nevertheless, we view the results of this study as a motivation to clarify the neurobiological similarities and differences between these two types of addiction. Future research utilizing structural modeling may help clarify the role of these specific regions and how they interact with other SUD- and BA-related regions and networks.Figures

[0297] Fig. 18 depicts flow diagram of the study selection process for addiction VBM studies in accordance with embodiments of the present invention. As shown in Fig. 18, a literature search identified an overall dataset of 108 independent publications reporting 108 experiments. Published VBM studies were systematically reviewed for inclusion in the metaanalysis. Study selection adhered to standard quality criteria of BrainMap coordinate-based meta-analysis in addition to the Preferred Reporting Items for Systematic Reviews and MetaAnalyses statement. For more information, visit: htps: Z / w w w. brat n map or g / Abbreviations: BA, behavioral addiction; CTh, cortical thickness; HC, healthy control; ROI, region of interest; SUD, substance use disorder; VBM, voxel-based morphometry.

[0298] Fig. 19 depicts independent SUD and BA ALE maps and overlap region in accordance with embodiments of the present invention. As shown in Fig. 19, the SUD meta-analysis (A) demonstrated a convergent pattern of gray matter alteration in regions including:794929-9580-4034v.1the right thalamus (medial dorsal nucleus), left caudate head, bilateral medial frontal gyrus (MFG) (Brodmann 6, 9, 10, and 11), left cingulate gyrus (Brodmann 32), left superior frontal gyrus (SFG) (Brodmann 6), and bilateral anterior cingulate cortex (ACC) (Brodmann 32). For the BA meta-analysis (B), a convergent pattern of gray matter alteration in only one cluster included the regions of: the left ACC (Brodmann 32, and 33), right cingulate gyrus (Brodmann 24), left caudate body and head, and left lentiform nucleus (putamen). These two independent meta-analyses demonstrated largely distinct anatomical-alteration patterns, overlapping (Pink Color, Orange Arrow) only in the left caudate head (C). Abbreviations: ALE, activation / alteration likelihood estimation; BA, behavioral addiction; SUD, substance use disorder.

[0299] Fig. 20 depicts pooled SUD + BA ALE map and interactive regions in accordance with embodiments of the present invention. (A) A convergent pattern of gray matter alteration was identified in the pooled SUD + BA meta-analysis, in regions including the anterior cingulate, inferior frontal g rus, insula, and thalamus. (B) Regions of gray matter alteration identified in the pooled SUD + BA meta-analysis but not in either independent SUD or BA meta-analyses (interactive regions), included the right and left insula (Brodmann 13), bilateral claustrum, left SFG (Brodmann 10). right inferior frontal gyrus (Brodmann 13). bilateral lentiform nucleus (lateral globus pallidus), bilateral parahippocampal gyrus (amygdala), right parahippocampal gyrus (hippocampus), left sub-gyral middle temporal gyrus (Brodmann 21), and left middle temporal gy rus (Brodmann 21). Abbreviations: ALE, activation / alteration likelihood estimation; BA, behavioral addiction; SUD, substance use disorder.

[0300] Figs. 21 A-C depicts results of whole image paradigm class analysis, in accordance with embodiments of the present invention. Whole image regions exhibiting gray matter alteration from the independent SUD meta-analysis (A) (n = 83) and the independent BA meta-analysis (B) (n == 25) were significantly associated with reward (z::::9.08 and 3.83 respectively). This association with rew ard w as also found in regions exhibiting gray matter alteration from the pooled SUD + BA meta-analysis (C) (n = 108) (z = 9.68). Abbreviations: BA. behavioral addiction; SUD, substance use disorder.

[0301] Figs. 22A-B depicts results of SUD cluster specific paradigm class analysis in accordance with embodiments of the present invention. SUD clusters 1 (z = 4.23), 2 (z = 5.28), 3 (z = 4.99), and 4 (z = 3.31) were significantly associated with reward (A, B, C, and D respectively). Abbreviations: SUD, substance use disorder.804929-9580-4034v.1

[0302] Figs. 23A-F depicts results of pooled SUD + BA cluster specific paradigm class analysis. In the pooled SUD + BA analysis, clusters 1 (z = 6.92), 4 (z = 4.71), and 6 (z = 4.15) were significantly associated with reward (A, D, and F respectively). Abbreviations: BA, behavioral addiction; SUD, substance use disorder.

[0303] Fig. 26 depicts a dimensionality estimation for substance use disorder / behavioral addiction, in accordance with embodiments of the present invention.ALTERNATIVE SYSTEMS AND METHODS

[0304] A method for improving effectiveness of a neurosurgery on a brain of a subject, the method comprising:(a) imaging brain activity of the subject prior to and / or during the neurosurgery: (b) comparing the imaged brain activity in real time to a database of virtual representations of brain networks associated with a neurologic disorder or a psychiatric disorder, the network comprising a morphologic property’ and a physiologic property and, optionally, a temporal property, wherein the neurosurgery is being performed for said neurological or psychiatric disorder, wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information comprising:(i) coordinate-based quantified brain activity data related to specific human subject tasks stored in a task-activation database (TA DB);(ii) voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); and(iii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); andwherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and (c) determining, based on the comparison, if the neurosurgery in progress is tending to revert the brain network associated w ith a neurologic disorder or a psychiatric disorder imaged in the subject to a brain activity associated with an ameliorated disease state or not;where not, adjusting the neurosurgery; and814929-9580-4034v.1(d) further determining if the adjusted neurosurgery in progress is tending to revert the brain network associated with a neurologic disorder or a psychiatric disorder imaged in the subject to a brain activity associated with an ameliorated disease state or not.

[0305] A method of treating a subject for a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes the method comprising:a) imaging brain activity of said subject;b) comparing said imaged brain activity to a database of virtual representations of brain networks associated with a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes, each network comprising a morphologic property and a physiologic property and, optionally, a temporal property’,wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information comprising:i) coordinate-based quantified brain activity’ data related to specific human subject tasks and stored in a task-activation database (TA DB);ii) voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); andiii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); andwherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and c) determining, based on the comparison, if the subject exhibits a brain network associated with a specific neurologic disorder, a specific psychiatric disorder, a specific development dis-order, or specific age-associated changes, wherein if the subject does exhibit said brain network associated with the specific neuro-logic disorder, the specific psychiatric disorder, the specific development disorder, or specific age-associated changes, administering a treatment to said subject for said neurologic disorder, psychiatric disorder, development disorder or age-associated changes.824929-9580-4034v.1

[0306] A method of determining if a neuromodulatory therapy is effective in ameliorating a neurologic disorder or a psychiatric disorder in a subject comprising:a) providing a virtual representation of a human brain network associated with a neurologic disorder or a psychiatric disorder,wherein the virtual representation was generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information comprising:i) coordinate-based quantified brain activity data related to specific human subject tasks and stored in a task-activation database (TA DB); ii) comprising voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); andiii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); andwherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and b) identifying, from imaging brain activity, if the subject has a brain network associated with a neurologic disorder or a psychiatric disorder, wherein, if the brain network is associated with a neurologic disorder or a psychiatric disorder, further performing the steps of:c) administering the neuromodulatory therapy to said subject; andd) further imaging brain activity of said subject so as to determine if the neuromodulatory therapy tends to revert the brain network associated with a neurologic disorder or a psychiatric disorder imaged in the subject to a brain activity associated with a non-disease state,wherein a neuromodulatory therapy which tends to revert the brain network associated with a neurologic disorder or a psychiatric disorder imaged in the subject to a brain activity associated with a non-disease state is determined as effective, and a neuromodulatory therapy which does not revert the brain network associated with a neurologic disorder or a psychiatric disorder imaged in the subject to a brain activity associated with a non-disease state is not determined as effective.834929-9580-4034v.1ADDITIONAL DATA TABLESTable 14. Node coordinates for GTMSeed_idx Seed_Label x_i Y_1 z_i 1 Parahippocampal Gyrus (Hippocampus) -15 -8 -9 2 Thalamus (Medial Dorsal Nucleus) _2 -8 4 3 Superior Temporal Gyrus (Brodmann area 38) -20 2 -16 4 Cerebellar Tonsil (*) 16 -32 -17 5 Caudate (Caudate Body) -3 8 3 6 Precentral Gyrus (Brodmann area 4) 16 -13 32 7 Supramarginal Gyrus (Brodmann area 40) -28 -27 18 8 Middle Occipital Gyrus (Brodmann area 18) -20 -42 -3 9 Precentral Gyrus (Brodmann area 4) -30 -4 15 10 Lingual Gyrus (Brodmann area 18) -2 -44 2 11 Sub-Gyral (Hippocampus) -14 -17 -1 12 Medial Frontal Gyrus (Brodmann area 10) -6 33 1 13 Middle Temporal Gyrus (Brodmann area 37) -27 -30 -4 14 Postcentral Gyrus (Brodmann area 3) -11 -13 32 15 Inferior Parietal Lobule (Brodmann area 40) -23 -16 21 16 Precuneus (Brodmann area 31) -1 -34 16 17 Cerebellar Tonsil (*) -19 -29 -21 18 Cingulate Gyrus (Brodmann area 31) 5 -3 24 19 Middle Frontal Gyrus (Brodmann area 10) -16 24 3 20 Inferior Occipital Gyrus (Brodmann area 19) 21 -41 -1 21 Superior Frontal Gyrus (Brodmann area 8) -18 12 25 22 Medial Frontal Gyrus (Brodmann area 9) 0 25 9 23 Inferior Frontal Gyrus (Brodmann area 47) 22 17 -3 24 Precuneus (Brodmann area 19) -16 -42 21 25 Caudate (Caudate Head) 6 12 -4 26 Precuneus (Brodmann area 7) 5 -20 28844929-9580-4034v.127 Precentral Gyrus (Brodmann area 6) 16 -32 -17 28 Middle Frontal Gyrus (Brodmann area 8) -12 21 18 29 * (Medial Geniculum Body) 11 -12 -3 30 Inferior Semi-Lunar Lobule (*) -10 -38 -22 31 Superior Frontal Gyrus (Brodmann area 6) 12 3 33 32 Fusiform Gyrus (Brodmann area 20 ) -27 -22 -13 33 Cingulate Gyrus (Brodmann area 31) 0 -20 16 34 Superior Temporal Gyrus (Brodmann area 22 ) -35 -22 9 35 Middle Temporal Gyrus (Brodmann area 21) 28 2 -8 36 Parahippocampal Gyrus (Amygdala) 11 -2 -11 37 Middle Frontal Gyrus (Brodmann area 6) -15 -3 24 38 - _2 -12 -13 39 Superior Parietal Lobule (Brodmann area 7 ) -14 -27 31 40 Claustrum (*) -17 3 -4 41 Superior Temporal Gyrus (*) -23 -16 9 42 Parahippocampal Gyrus (Brodmann area 36) 17 -13 -14 43 Middle Frontal Gyrus (Brodmann area 46) -25 16 12 44 Anterior Cingulate (Brodmann area 32) 1 25 -5 45 Inferior Parietal Lobule (Brodmann area 40) 26 -23 25 46 Inferior Temporal Gyrus (Brodmann area 20) 32 -22 -11 47 Superior Temporal Gyrus (Brodmann area 22 ) 25 -28 10 48 Medial Frontal Gyrus (Brodmann area 6) -7 6 34 49 Superior Occipital Gyrus (Brodmann area 19) 20 -38 16 50 Caudate (Caudate Body) 9 5 8 51 Parahippocampal Gyrus (Brodmann area 30) -7 -26 3 52 Superior Temporal Gyrus (Brodmann area 22 ) -30 5 -4 53 Inferior Temporal Gyrus (Brodmann area 20) -31 -9 -11 54 Superior Frontal Gyrus (Brodmann area 10) 14 26 -1 55 Insula (Brodmann area 13) -21 -4 6854929-9580-4034v.156 Precentral Gyrus (Brodmann area 9) 17 6 17 57 Superior Frontal Gyrus (Brodmann area 8) 10 20 24 58 Middle Frontal Gyrus (Brodmann area 9) 19 24 11 59 Superior Frontal Gyrus (Brodmann area 6) -8 18 29 60 Medial Frontal Gyrus (Brodmann area 25) -6 12 -11 61 Superior Temporal Gyrus (Brodmann area 22) 27 -7 0 62 Inferior Frontal Gyrus (Brodmann area 9) 31 9 12 63 Posterior Cingulate (Brodmann area 31 ) 12 -30 12 64 Cerebellar Tonsil (*) 25 -30 -26 65 Cingulate Gyrus (Brodmann area 24) -4 0 16 66 Middle Temporal Gyrus (Brodmann area 38) 18 4 -22 67 - -6 -27 -28 68 Middle Frontal Gyrus (Brodmann area 47) -22 17 -5 69 Lingual Gyrus (Brodmann area 18) 6 -34 2 70 Cingulate Gyrus (Brodmann area 32) -1 14 14 71 Insula (Brodmann area 13) 18 -11 10 72 Middle Temporal Gyrus (Brodmann area 39) -25 -33 8 73 Superior Frontal Gyrus (Brodmann area 9) 8 33 13 74 Cuneus (Brodmann area 19) 8 -43 19 75 Middle Occipital Gyrus (Brodmann area 18) -12 -47 8 76 Parahippocampal Gyrus (Brodmann area 30) 8 -22 3 77 Precuneus (Brodmann area 7) 5 -34 29 78 Superior Parietal Lobule (Brodmann area 7) 16 -31 25 79 Tuber (*) -11 -47 -14 80 * (Hypothalamus) -2 -1 -6 81 Uvula (*) 16 -43 -12 82 - 9 -32 -27 83 Precuneus (Brodmann area 7) -7 -36 27 84 Middle Temporal Gyrus (Brodmann area 21) -29 -17 -2864929-9580-4034v.185 Declive (*) -14 -28 -7 86 Insula (Brodmann area 13) -16 8 7 87 Cuneus (*) 7 -52 3 88 - -15 -24 13 89 Inferior Temporal Gyrus (Brodmann area 20) -25 -8 -22 90 Superior Temporal Gyrus (Brodmann area 22) 26 -19 0 91 Inferior Frontal Gyrus (Brodmann area 47) 16 11 -12 92 Postcentral Gyrus (Brodmann area 7) 16 -24 37 93 Superior Frontal Gyrus (Brodmann area 6) 3 15 33 94 Superior Frontal Gyrus (Brodmann area 11) -7 32 -11 95 - 0 -9 -25 96 Superior Frontal Gyrus (Brodmann area 9) -9 32 14 97 Inferior Semi-Lunar Lobule (*) 2 -42 -23 98 Tuber (*) -27 -36 -15 99 - -19 14 -17 100 Precuneus (Brodmann area 7) -3 -30 38 101 Claustrum (*) 18 1 1 102 Precentral Gyrus (Brodmann area 6) -18 -4 36 103 Inferior Parietal Lobule (Brodmann area 40) 30 -17 15 104 * (*) -4 -24 -14 105 Cuneus (Brodmann area 19) -3 -48 16 106 Inferior Temporal Gyrus (Brodmann area 20) 24 -7 -22 107 - 15 -19 21 108 Middle Temporal Gyrus (Brodmann area 21) 30 -10 -12 109 Superior Frontal Gyrus (Brodmann area 10) 7 36 1 110 - 19 -19 -28 111 Culmen (*) -13 -16 -18 112 - 1 -9 40 113 Rectal Gyrus (Brodmann area 11) 5 17 -15874929-9580-4034v.1114 Lingual Gyrus (Brodmann area 18 ) 1 -50 -8 115 Inferior Temporal Gyrus (Brodmann area 37) 33 -30 -2 116 - 0 1 -18 117 - -19 -20 -29 118 - _2 -18 -35Table 15. Edge coordinates for GTMSeed Seed_Labe Seed_Label_ Target Target_La Targe t_Label X Y Z X Y Z _idx 1 Unique _idx bel _Unique _1 _1 _1 2 _2 _2 1 Parahippoc Parahippoca 2 Thalamus Thalamus -8 -9 _2 -8 4 ampal mpal Gyrus (Medial (Medial 15Gyrus (Hippocamp Dorsal Dorsal(Hippocam us) - Node 1 Nucleus) Nucleus)pus) Node 21 Parahippoc Parahippoca 3 Superior Superior -8 -9 2 ampal mpal Gyrus Temporal Temporal 15 20 16 Gyrus (Hippocamp Gyrus Gyrus(Hippocam us) - Node 1 (Brodman (Brodmannpus) n area 38) area 38)Node 32 Thalamus Thalamus 3 Superior Superior -2 -8 4 2 (Medial (Medial Temporal Temporal 20 16 Dorsal Dorsal Gyrus GyrusNucleus) Nucleus) (Brodman (BrodmannNode 2 n area 38) area 38)Node 31 Parahippoc Parahippoca 4 Cerebellar Cerebellar -8 -9 16 ampal mpal Gyrus Tonsil (*) Tonsil (*) - 15 32 17 Gyrus (Hippocamp Node 4(Hippocam us) - Node 1pus)2 Thalamus Thalamus 4 Cerebellar Cerebellar -2 -8 4 16 (Medial (Medial Tonsil (*) Tonsil (*) - 32 17 Dorsal Dorsal Node 4Nucleus)884929-9580-4034v.1Nucleus)Node 21 Parahippoc Parahippoca 5 Caudate Caudate -8 -9 -3 8 3 ampal mpal Gyrus (Caudate (Caudate 15Gyrus (Hippocamp Body) Body) - Node(Hippocam us) - Node 1 5pus)2 Thalamus Thalamus 5 Caudate Caudate -2 -8 4 -3 8 3 (Medial (Medial (Caudate (CaudateDorsal Dorsal Body) Body) - NodeNucleus) Nucleus) 5Node 23 Superior Superior 5 Caudate Caudate 2 -3 8 3 Temporal Temporal (Caudate (Caudate 20 16Gyrus Gyrus Body) Body) - Node(Brodman (Brodmann 5n area 38) area 38) - Node 31 Parahippoc Parahippoca 6 Precentral Precentral -8 -9 16 32 ampal mpal Gyrus Gyrus Gyrus 15 13 Gyrus (Hippocamp (Brodman (Brodmann(Hippocam us) - Node 1 n area 4) area 4) - Nodepus) 62 Thalamus Thalamus 6 Precentral Precentral -2 -8 4 16 32 (Medial (Medial Gyrus Gyrus 13 Dorsal Dorsal (Brodman (BrodmannNucleus) Nucleus) n area 4) area 4) - NodeNode 2 63 Superior Superior 6 Precentral Precentral 2 16 32 Temporal Temporal Gyrus Gyrus 20 16 13 Gyrus Gyrus (Brodman (Brodmann(Brodman (Brodmann n area 4) area 4) - Noden area 38) area 38) - 6Node 32 Thalamus Thalamus 7 Supramarg Supramargina -2 -8 4 18 (Medial (Medial inal Gyrus 1 Gyrus 28 27 Dorsal (Brodmann894929-9580-4034v.1Dorsal Nucleus) (Brodman area 40)Nucleus) Node 2 n area 40) Node 74 Cerebellar Cerebellar 7 Supramarg Supramargina 16 18 Tonsil (*) Tonsil (*) - inal Gyrus 1 Gyrus 32 17 28 27 Node 4 (Brodman (Brodmannn area 40) area 40)Node 71 Parahippoc Parahippoca 8 Middle Middle -8 -9 -3 ampal mpal Gyrus Occipital Occipital 15 20 42 Gyrus (Hippocamp Gyrus Gyrus(Hippocam us) - Node 1 (Brodman (Brodmannpus) n area 18) area 18)Node 82 Thalamus Thalamus 8 Middle Middle -2 -8 4 -3 (Medial (Medial Occipital Occipital 20 42 Dorsal Dorsal Gyms GymsNucleus) Nucleus) (Brodman (BrodmannNode 2 n area 18) area 18)Node 81 Parahippoc Parahippoca 9 Precentral Precentral -8 -9 -4 15 ampal mpal Gyrus Gyrus Gyrus 15 30 Gyrus (Hippocamp (Brodman (Brodmann(Hippocam us) - Node 1 n area 4) area 4) - Nodepus) 92 Thalamus Thalamus 9 Precentral Precentral -2 -8 4 -4 15 (Medial (Medial Gyrus Gyrus 30 Dorsal Dorsal (Brodman (BrodmannNucleus) Nucleus) n area 4) area 4) - NodeNode 2 97 Supramarg Supramargin 9 Precentral Precentral 18 -4 15 inal Gyrus al Gyrus Gyrus Gyrus 28 27 30 (Brodman (Brodmann (Brodman (Brodmannn area 40) area 40) - n area 4) area 4) - NodeNode 7 92 Thalamus Thalamus 10 Lingual Lingual Gyms -2 -8 4 -2 2 (Medial (Medial Gyms (Brodmann 44Dorsal904929-9580-4034v.1Dorsal Nucleus) (Brodman area 18)Nucleus) Node 2 n area 18) Node 104 Cerebellar Cerebellar 10 Lingual Lingual Gyrus 16 -2 2 Tonsil (*) Tonsil (*) - Gyrus (Brodmann 32 17 44 Node 4 (Brodman area 18)n area 18) Node 107 Supramarg Supramargin 10 Lingual Lingual Gyrus 18 -2 2 inal Gyrus al Gyrus Gyrus (Brodmann 28 27 44 (Brodman (Brodmann (Brodman area 18)n area 40) area 40) - n area 18) Node 10Node 71 Parahippoc Parahippoca 11 Sub-Gyral Sub-Gyral -8 -9 -1 ampal mpal Gyrus (Hippocam (Hippocampu 15 14 17 Gyrus (Hippocamp pus) s) - Node 11(Hippocam us) - Node 1pus)2 Thalamus Thalamus 11 Sub-Gyral Sub-Gyral -2 -8 4 -1 (Medial (Medial (Hippocam (Hippocampu 14 17 Dorsal Dorsal pus) s) - Node 11Nucleus) Nucleus)Node 21 Parahippoc Parahippoca 12 Medial Medial -8 -9 -6 33 1 ampal mpal Gyrus Frontal Frontal Gyrus 15Gyrus (Hippocamp Gyrus (Brodmann(Hippocam us) - Node 1 (Brodman area 10)pus) n area 10) Node 123 Superior Superior 12 Medial Medial 2 -6 33 1 Temporal Temporal Frontal Frontal Gyrus 20 16Gyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 10)n area 38) area 38) - n area 10) Node 12Node 34 Cerebellar Cerebellar 12 Medial Medial 16 -6 33 1 Tonsil (*) Tonsil (*) - Frontal Frontal Gyrus 32 17Node 4 Gyrus (Brodmann(Brodman area 10)n area 10) Node 12914929-9580-4034v.16 Precentral Precentral 12 Medial Medial 16 32 -6 33 1 Gyrus Gyrus Frontal Frontal Gyrus 13(Brodman (Brodmann Gyrus (Brodmannn area 4) area 4) (Brodman area 10)Node 6 n area 10) Node 127 Supramarg Supramargin 12 Medial Medial 18 -6 33 1 inal Gyrus al Gyrus Frontal Frontal Gyrus 28 27(Brodman (Brodmann Gyrus (Brodmannn area 40) area 40) - (Brodman area 10)Node 7 n area 10) Node 128 Middle Middle 12 Medial Medial -3 -6 33 1 Occipital Occipital Frontal Frontal Gyrus 20 42Gyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 10)n area 18) area 18) - n area 10) Node 12Node 89 Precentral Precentral 12 Medial Medial -4 15 -6 33 1 Gyrus Gyrus Frontal Frontal Gyrus 30(Brodman (Brodmann Gyrus (Brodmannn area 4) area 4) (Brodman area 10)Node 9 n area 10) Node 1210 Lingual Lingual 12 Medial Medial -2 2 -6 33 1 Gyrus Gyrus Frontal Frontal Gyrus 44(Brodman (Brodmann Gyrus (Brodmannn area 18) area 18) - (Brodman area 10)Node 10 n area 10) Node 121 Parahippoc Parahippoca 13 Middle Middle -8 -9 -4 ampal mpal Gyrus Temporal Temporal 15 27 30 Gyrus (Hippocamp Gyrus Gyrus(Hippocam us) - Node 1 (Brodman (Brodmannpus) n area 37) area 37)Node 132 Thalamus Thalamus 13 Middle Middle -2 -8 4 -4 (Medial (Medial Temporal Temporal 27 30 Dorsal Dorsal Gyrus GyrusNucleus) Nucleus) (Brodman (BrodmannNode 2 n area 37)924929-9580-4034v.1area 37)Node 134 Cerebellar Cerebellar 13 Middle Middle 16 -4 Tonsil (*) Tonsil (*) - Temporal Temporal 32 17 27 30 Node 4 Gyrus Gyrus(Brodman (Brodmannn area 37) area 37)Node 138 Middle Middle 13 Middle Middle -3 -4 Occipital Occipital Temporal Temporal 20 42 27 30 Gyrus Gyrus Gyrus Gyrus(Brodman (Brodmann (Brodman (Brodmannn area 18) area 18) - n area 37) area 37)Node 8 Node 139 Precentral Precentral 13 Middle Middle -4 15 -4 Gyrus Gyrus Temporal Temporal 30 27 30 (Brodman (Brodmann Gyrus Gyrusn area 4) area 4) (Brodman (BrodmannNode 9 n area 37) area 37)Node 1312 Medial Medial 13 Middle Middle -6 33 1 -4 Frontal Frontal Temporal Temporal 27 30 Gyrus Gyrus Gyrus Gyrus(Brodman (Brodmann (Brodman (Brodmannn area 10) area 10) - n area 37) area 37)Node 12 Node 131 Parahippoc Parahippoca 14 Postcentral Postcentral -8 -9 32 ampal mpal Gyrus Gyrus Gyrus 15 11 13 Gyrus (Hippocamp (Brodman (Brodmann(Hippocam us) - Node 1 n area 3) area 3) - Nodepus) 142 Thalamus Thalamus 14 Postcentral Postcentral -2 -8 4 32 (Medial (Medial Gyrus Gyrus 11 13 Dorsal Dorsal (Brodman (BrodmannNucleus) Nucleus) n area 3) area 3) - NodeNode 2 14934929-9580-4034v.13 Superior Superior 14 Postcentral Postcentral 2 32 Temporal Temporal Gyrus Gyrus 20 16 11 13 Gyrus Gyrus (Brodman (Brodmann(Brodman (Brodmann n area 3) area 3) - Noden area 38) area 38) - 14Node 36 Precentral Precentral 14 Postcentral Postcentral 16 32 32 Gyrus Gyrus Gyrus Gyrus 13 11 13 (Brodman (Brodmann (Brodman (Brodmannn area 4) area 4) n area 3) area 3) - NodeNode 6 1412 Medial Medial 14 Postcentral Postcentral -6 33 1 32 Frontal Frontal Gyrus Gyrus 11 13 Gyrus Gyrus (Brodman (Brodmann(Brodman (Brodmann n area 3) area 3) - Noden area 10) area 10) - 14Node 121 Parahippoc Parahippoca 15 Inferior Inferior -8 -9 21 ampal mpal Gyrus Parietal Parietal 15 23 16 Gyrus (Hippocamp Lobule Lobule(Hippocam us) - Node 1 (Brodman (Brodmannpus) n area 40) area 40)Node 152 Thalamus Thalamus 15 Inferior Inferior -2 -8 4 21 (Medial (Medial Parietal Parietal 23 16 Dorsal Dorsal Lobule LobuleNucleus) Nucleus) (Brodman (BrodmannNode 2 n area 40) area 40)Node 152 Thalamus Thalamus 16 Precuneus Precuneus -2 -8 4 -1 16 (Medial (Medial (Brodman (Brodmann 34 Dorsal Dorsal n area 31 ) area 31)Nucleus) Nucleus) Node 16Node 24 Cerebellar Cerebellar 16 Precuneus Precuneus 16 -1 16 Tonsil (*) Tonsil (*) - (Brodman (Brodmann 32 17 34 Node 4 n area 31 )944929-9580-4034v.1area 31)Node 167 Supramarg Supramargin 16 Precuneus Precuneus 18 -1 16 inal Gyrus al Gyrus (Brodman (Brodmann 28 27 34 (Brodman (Brodmann n area 31) area 31)n area 40) area 40) - Node 16Node 710 Lingual Lingual 16 Precuneus Precuneus -2 2 -1 16 Gyrus Gyrus (Brodman (Brodmann 44 34 (Brodman (Brodmann n area 31) area 31)n area 18) area 18) - Node 16Node 1012 Medial Medial 16 Precuneus Precuneus -6 33 1 -1 16 Frontal Frontal (Brodman (Brodmann 34 Gyrus Gyrus n area 31) area 31)(Brodman (Brodmann Node 16n area 10) area 10) - Node 121 Parahippoc Parahippoca 17 Cerebellar Cerebellar -8 -9ampal mpal Gyrus Tonsil (*) Tonsil (*) - 15 19 29 21 Gyrus (Hippocamp Node 17(Hippocam us) - Node 1pus)2 Thalamus Thalamus 17 Cerebellar Cerebellar -2 -8 4(Medial (Medial Tonsil (*) Tonsil (*) - 19 29 21 Dorsal Dorsal Node 17Nucleus) Nucleus)Node 24 Cerebellar Cerebellar 17 Cerebellar Cerebellar 16Tonsil (*) Tonsil (*) - Tonsil (*) Tonsil (*) - 32 17 19 29 21Node 4 Node 177 Supramarg Supramargin 17 Cerebellar Cerebellar 18inal Gyrus al Gyrus Tonsil (*) Tonsil (*) - 28 27 19 29 21 (Brodman (Brodmann Node 17n area 40) area 40) - Node 7954929-9580-4034v.18 Middle Middle 17 Cerebellar Cerebellar -3 Occipital Occipital Tonsil (*) Tonsil (*) - 20 42 19 29 21 Gyrus Gyrus Node 17(Brodman (Brodmannn area 18) area 18) - Node 89 Precentral Precentral 17 Cerebellar Cerebellar -4 15Gyrus Gyrus Tonsil (*) Tonsil (*) - 30 19 29 21 (Brodman (Brodmann Node 17n area 4) area 4)Node 912 Medial Medial 17 Cerebellar Cerebellar -6 33 1Frontal Frontal Tonsil (*) Tonsil (*) - 19 29 21 Gyrus Gyrus Node 17(Brodman (Brodmannn area 10) area 10) - Node 1213 Middle Middle 17 Cerebellar Cerebellar -4 Temporal Temporal Tonsil (*) Tonsil (*) - 27 30 19 29 21 Gyrus Gyrus Node 17(Brodman (Brodmannn area 37) area 37) - Node 131 Parahippoc Parahippoca 18 Cingulate Cingulate -8 -9 5 24 ampal mpal Gyrus Gyrus Gyrus 15Gyrus (Hippocamp (Brodman (Brodmann(Hippocam us) - Node 1 n area 31) area 31)pus) Node 182 Thalamus Thalamus 18 Cingulate Cingulate -2 -8 4 5 -3 24 (Medial (Medial Gyrus GyrusDorsal Dorsal (Brodman (BrodmannNucleus) Nucleus) n area 31 ) area 31)Node 2 Node 183 Superior Superior 18 Cingulate Cingulate 2 5 -3 24 Temporal Temporal Gyrus Gyrus 20 16Gyrus Gyrus (Brodman (Brodmann(Brodmann n area 31 )964929-9580-4034v.1(Brodman area 38) - area 31)n area 38) Node 3 Node 185 Caudate Caudate 18 Cingulate Cingulate -3 8 3 5 -3 24 (Caudate (Caudate Gyrus GyrusBody) Body) (Brodman (BrodmannNode 5 n area 31) area 31)Node 1814 Postcentral Postcentral 18 Cingulate Cingulate 32 5 -3 24 Gyrus Gyrus Gyrus Gyrus 11 13(Brodman (Brodmann (Brodman (Brodmannn area 3) area 3) n area 31) area 31)Node 14 Node 181 Parahippoc Parahippoca 19 Middle Middle -8 -9 24 3 ampal mpal Gyrus Frontal Frontal Gyrus 15 16 Gyrus (Hippocamp Gyrus (Brodmann(Hippocam us) - Node 1 (Brodman area 10)pus) n area 10) Node 192 Thalamus Thalamus 19 Middle Middle -2 -8 4 24 3 (Medial (Medial Frontal Frontal Gyrus 16 Dorsal Dorsal Gyrus (BrodmannNucleus) Nucleus) (Brodman area 10)Node 2 n area 10) Node 193 Superior Superior 19 Middle Middle 2 24 3 Temporal Temporal Frontal Frontal Gyrus 20 16 16 Gyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 10)n area 38) area 38) - n area 10) Node 19Node 35 Caudate Caudate 19 Middle Middle -3 8 24 (Caudate (Caudate Frontal Frontal Gyrus 16 Body) Body) Gyrus (BrodmannNode 5 (Brodman area 10)n area 10) Node 1918 Cingulate Cingulate 19 Middle Middle 5 24 24 J Gyrus Gyrus Frontal Frontal Gyrus 16 (Brodman (Brodmann Gyrus (Brodmannn area 31)974929-9580-4034v.1area 31) - (Brodman area 10)Node 18 n area 10) Node 191 Parahippoc Parahippoca 22 Medial Medial -8 -9 0 25 9 ampal mpal Gyrus Frontal Frontal Gyrus 15Gyrus (Hippocamp Gyrus (Brodmann(Hippocam us) - Node 1 (Brodman area 9) - Nodepus) n area 9) 222 Thalamus Thalamus 22 Medial Medial -2 -8 4 0 25 9 (Medial (Medial Frontal Frontal GyrusDorsal Dorsal Gyrus (BrodmannNucleus) Nucleus) (Brodman area 9) - NodeNode 2 n area 9) 223 Superior Superior 22 Medial Medial 2 0 25 9 Temporal Temporal Frontal Frontal Gyrus 20 16Gyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 9) - Noden area 38) area 38) - n area 9) 22Node 35 Caudate Caudate 22 Medial Medial -3 8 0 25 9 (Caudate (Caudate Frontal Frontal GyrusBody) Body) Gyrus (BrodmannNode 5 (Brodman area 9) - Noden area 9) 2218 Cingulate Cingulate 22 Medial Medial 5 24 0 25 9 Gyrus Gyrus Frontal Frontal Gyrus(Brodman (Brodmann Gyrus (Brodmannn area 31) area 31) - (Brodman area 9) - NodeNode 18 n area 9) 2219 Middle Middle 22 Medial Medial 24 0 25 9 Frontal Frontal Frontal Frontal Gyrus 16Gyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 9) - Noden area 10) area 10) - n area 9) 22Node 192 Thalamus Thalamus 23 Interior Interior -2 -8 4 22 17 -3 (Medial (Medial Frontal Frontal GyrusDorsal Gyrus (Brodmann984929-9580-4034v.1Dorsal Nucleus) (Brodman area 47)Nucleus) Node 2 n area 47) Node 234 Cerebellar Cerebellar 23 Inferior Inferior 16 22 17 -3 Tonsil (*) Tonsil (*) - Frontal Frontal Gyrus 32 17Node 4 Gyrus (Brodmann(Brodman area 47)n area 47) Node 237 Supramarg Supramargin 23 Inferior Inferior 18 22 17 -3 inal Gyrus al Gyrus Frontal Frontal Gyrus 28 27(Brodman (Brodmann Gyrus (Brodmannn area 40) area 40) - (Brodman area 47)Node 7 n area 47) Node 239 Precentral Precentral 23 Inferior Inferior -4 15 22 17 -3 Gyrus Gyrus Frontal Frontal Gyrus 30(Brodman (Brodmann Gyrus (Brodmannn area 4) area 4) (Brodman area 47)Node 9 n area 47) Node 2312 Medial Medial 23 Inferior Inferior -6 33 1 22 17 -3 Frontal Frontal Frontal Frontal GyrusGyrus Gyrus Gyrus (Brodmann(Brodman (Brodmann (Brodman area 47)n area 10) area 10) - n area 47) Node 23Node 121 Parahippoc Parahippoca 25 Caudate Caudate -8 -9 6 12 -4 ampal mpal Gyrus (Caudate (Caudate 15Gyrus (Hippocamp Head) Head) - Node(Hippocam us) - Node 1 25pus)2 Thalamus Thalamus 25 Caudate Caudate -2 -8 4 6 12 -4 (Medial (Medial (Caudate (CaudateDorsal Dorsal Head) Head) - NodeNucleus) Nucleus) 25Node 25 Caudate Caudate 25 Caudate Caudate -3 8 3 6 12 -4 (Caudate (Caudate (Caudate (CaudateBody) Body) Head) Head) - NodeNode 5 25994929-9580-4034v.16 Precentral Precentral 25 Caudate Caudate 16 32 6 12 -4 Gyrus Gyrus (Caudate (Caudate 13(Brodman (Brodmann Head) Head) - Noden area 4) area 4) 25Node 64 Cerebellar Cerebellar 27 Precentral Precentral 16 26 -5 17 Tonsil (*) Tonsil (*) - Gyrus Gyrus 32 17Node 4 (Brodman (Brodmannn area 6) area 6) - Node277 Supramarg Supramargin 27 Precentral Precentral 18 26 -5 17 inal Gyrus al Gyrus Gyrus Gyrus 28 27(Brodman (Brodmann (Brodman (Brodmannn area 40) area 40) - n area 6) area 6) - NodeNode 7 279 Precentral Precentral 27 Precentral Precentral -4 15 26 -5 17 Gyrus Gyrus Gyrus Gyrus 30(Brodman (Brodmann (Brodman (Brodmannn area 4) area 4) n area 6) area 6) - NodeNode 9 2710 Lingual Lingual 27 Precentral Precentral -2 2 26 -5 17 Gyrus Gyrus Gyrus Gyrus 44(Brodman (Brodmann (Brodman (Brodmannn area 18) area 18) - n area 6) area 6) - NodeNode 10 2723 Inferior Inferior 27 Precentral Precentral 22 17 -3 26 -5 17 Frontal Frontal Gyrus GyrusGyrus Gyrus (Brodman (Brodmann(Brodman (Brodmann n area 6) area 6) - Noden area 47) area 47) - 27Node 236 Precentral Precentral 29 * (Medial * (Medial 16 32 11 -3 Gyrus Gyrus Geniculum Geniculum 13 12 (Brodman (Brodmann Body) Body) - Noden area 4) area 4) 29Node 61004929-9580-4034v.128 Middle Middle 33 Cingulate Cingulate 21 18 0 16 Frontal Frontal Gyrus Gyrus 12 20 Gyrus Gyrus (Brodman (Brodmann(Brodman (Brodmann n area 31) area 31)n area 8) area 8) Node 33Node 281 Parahippoc Parahippoca 36 Parahippoc Parahippocam -8 -9 11 -2 ampal mpal Gyrus ampal pal Gyrus 15 11 Gyrus (Hippocamp Gyrus (Amygdala) - (Hippocam us) - Node 1 (Amygdala Node 36pus) )2 Thalamus Thalamus 37 Middle Middle -2 -8 4 -3 24 (Medial (Medial Frontal Frontal Gyrus 15 Dorsal Dorsal Gyrus (BrodmannNucleus) Nucleus) (Brodman area 6) - NodeNode 2 n area 6) 3736 Parahippoc Parahippoca 42 Parahippoc Parahippocam 11 -2 17 ampal mpal Gyrus ampal pal Gyrus 11 13 14 Gyrus (Amygdala) Gyrus (Brodmann(Amygdala - Node 36 (Brodman area 36)) n area 36) Node 421 Parahippoc Parahippoca 44 Anterior Anterior -8 -9 1 25 -5 ampal mpal Gyrus Cingulate Cingulate 15Gyrus (Hippocamp (Brodman (Brodmann(Hippocam us) - Node 1 n area 32) area 32)pus) Node 4420 Inferior Inferior 21 -1 Occipital Occipital 41Gyrus Gyrus(Brodman (Brodmannn area 19) area 19) - Node 2021 Superior Superior 12 25Frontal Frontal 18Gyrus Gyrus(Brodman (Brodmannn area 8)1014929-9580-4034v.1area 8)Node 2124 Precuneus Precuneus 21 (Brodman (Brodmann 16 42 n area 19) area 19) - Node 2426 Precuneus Precuneus 5 28 (Brodman (Brodmann 20 n area 7) area 7)Node 2630 Inferior InferiorSemi- Semi-Lunar 10 38 22 Lunar Lobule (*) - Lobule (*) Node 3031 Superior Superior 12 3 33 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 6) area 6)Node 3132 Fusiform FusiformGyrus Gyrus 27 22 13 (Brodman (Brodmannn area 20) area 20) - Node 3234 Superior Superior 9 Temporal Temporal 35 22 Gyrus Gyrus(Brodman (Brodmannn area 22) area 22) - Node 3435 Middle Middle 28 2 -8 Temporal TemporalGyrus Gyrus(Brodman (Brodmannn area 21) area 21) - Node 351024929-9580-4034v.138 - - - Node 38 -212 13 39 Superior Superior 31 Parietal Parietal 14 27 Lobule Lobule(Brodman (Brodmannn area 7) area 7)Node 3940 Claustrum Claustrum 3 -4 (*) (*) - Node 40 17 41 Superior Superior 9 Temporal Temporal 23 16 Gyrus (*) Gyrus (*) - Node 4143 Middle Middle 16 12 Frontal Frontal 25 Gyrus Gyrus(Brodman (Brodmannn area 46) area 46) - Node 4345 Inferior Inferior 26 25 Parietal Parietal 23 Lobule Lobule(Brodman (Brodmannn area 40) area 40) - Node 4546 Inferior Inferior 32 Temporal Temporal 22 11 Gyrus Gyrus(Brodman (Brodmannn area 20) area 20) - Node 4647 Superior Superior 25 10 Temporal Temporal 28 Gyrus Gyrus(Brodman (Brodmannn area 22)1034929-9580-4034v.1area 22) - Node 4748 Medial Medial -7 6 34 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 6) area 6)Node 4849 Superior Superior 20 16 Occipital Occipital 38 Gyrus Gyrus(Brodman (Brodmannn area 19) area 19) - Node 4950 Caudate Caudate 9 5 8 (Caudate (CaudateBody) Body)Node 5051 Parahippoc Parahippoca -7 ampal mpal Gyrus 26 Gyrus (Brodmann(Brodman area 30) - n area 30) Node 5152 Superior Superior 5 -4 Temporal Temporal 30 Gyrus Gyrus(Brodman (Brodmannn area 22) area 22) - Node 5253 Inferior Inferior -9 Temporal Temporal 31 11 Gyrus Gyrus(Brodman (Brodmannn area 20) area 20) - Node 5354 Superior Superior 14 26 -1 Frontal Frontal1044929-9580-4034v.1Gyrus Gyrus(Brodman (Brodmannn area 10) area 10) - Node 5455 Insula Insula -4 6 (Brodman (Brodmann 21 n area 13) area 13) - Node 5556 Precentral Precentral 17 6 17 Gyrus Gyrus(Brodman (Brodmannn area 9) area 9)Node 5657 Superior Superior 10 20 24 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 8) area 8)Node 5758 Middle Middle 19 24 11 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 9) area 9)Node 5859 Superior Superior -8 18 29 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 6) area 6)Node 5960 Medial Medial -6 12 Frontal Frontal 11 Gyrus Gyrus(Brodman (Brodmannn area 25) area 25) - Node 601054929-9580-4034v.161 Superior Superior 27 -7 0 Temporal TemporalGyrus Gyrus(Brodman (Brodmannn area 22) area 22) - Node 6162 Inferior Inferior 31 9 12 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 9) area 9)Node 6263 Posterior Posterior 12 12 Cingulate Cingulate 30 (Brodman (Brodmannn area 31) area 31) - Node 6364 Cerebellar Cerebellar 25 Tonsil (*) Tonsil (*) - 30 26Node 6465 Cingulate Cingulate -4 0 16 Gyrus Gyrus(Brodman (Brodmannn area 24) area 24) - Node 6566 Middle Middle 18 4 Temporal Temporal 22 Gyrus Gyrus(Brodman (Brodmannn area 38) area 38) - Node 6667 - - - Node 67 -627 28 68 Middle Middle 17 -5 Frontal Frontal 22 Gyrus Gyrus(Brodmann1064929-9580-4034v.1(Brodman area 47) - n area 47) Node 6869 Lingual Lingual 6 2 Gyrus Gyrus 34 (Brodman (Brodmannn area 18) area 18) - Node 6970 Cingulate Cingulate -1 14 14 Gyrus Gyrus(Brodman (Brodmannn area 32) area 32) - Node 7071 Insula Insula 18 10 (Brodman (Brodmann 11 n area 13) area 13) - Node 7172 Middle Middle 8 Temporal Temporal 25 33 Gyrus Gyrus(Brodman (Brodmannn area 39) area 39) - Node 7273 Superior Superior 8 33 13 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 9) area 9)Node 7374 Cuneus Cuneus 8 19 (Brodman (Brodmann 43 n area 19) area 19) - Node 7475 Middle Middle 8 Occipital Occipital 12 47 Gyrus Gyrus(Brodman (Brodmannn area 18)1074929-9580-4034v.1area 18) - Node 7576 Parahippoc Parahippoca 8 3 ampal mpal Gyrus 22 Gyrus (Brodmann(Brodman area 30) - n area 30) Node 7677 Precuneus Precuneus 5 29 (Brodman (Brodmann 34 n area 7) area 7)Node 7778 Superior Superior 16 25 Parietal Parietal 31 Lobule Lobule(Brodman (Brodmannn area 7) area 7)Node 7879 Tuber (*) Tuber (*) - Node 79 11 47 14 80 * * _2 -1 -6 (Hypothala (Hypothalammus) us ) - Node 8081 Uvula (*) Uvula (*) - 16Node 81 43 12 82 - - - Node 82 932 27 83 Precuneus Precuneus -7 27 (Brodman (Brodmann 36 n area 7) area 7)Node 8384 Middle Middle -2 Temporal Temporal 29 17 Gyrus Gyrus(Brodman (Brodmannn area 21) area 21) - Node 841084929-9580-4034v.185 Declive (*) Declive (*) - -7 Node 85 14 28 86 Insula Insula 8 7 (Brodman (Brodmann 16 n area 13) area 13) - Node 8687 Cuneus (*) Cuneus (*) - 7 3Node 87 52 88 - - - Node 88 1315 24 89 Inferior Inferior -8 Temporal Temporal 25 22 Gyrus Gyrus(Brodman (Brodmannn area 20 ) area 20) - Node 8990 Superior Superior 26 0 Temporal Temporal 19 Gyrus Gyrus(Brodman (Brodmannn area 22) area 22) - Node 9091 Inferior Inferior 16 11 Frontal Frontal 12 Gyrus Gyrus(Brodman (Brodmannn area 47) area 47) - Node 9192 Postcentral Postcentral 16 37 Gyrus Gyrus 24 (Brodman (Brodmannn area 7) area 7)Node 9293 Superior Superior 3 15 33 Frontal FrontalGyrus Gyrus(Brodmann1094929-9580-4034v.1(Brodman area 6)n area 6) Node 9394 Superior Superior -7 32 Frontal Frontal 11 Gyrus Gyrus(Brodman (Brodmannn area 11) area 11) - Node 9495 - - - Node 95 0 -925 96 Superior Superior -9 32 14 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 9) area 9)Node 9697 Inferior Inferior 2 Semi- Semi-Lunar 42 23 Lunar Lobule (*) - Lobule (*) Node 9798 Tuber (*) Tuber (*) - Node 98 27 36 15 99 - - - Node 99 1419 17 100 Precuneus Precuneus -3 38 (Brodman (Brodmann 30 n area 7) area 7)Node 100101 Claustrum Claustrum 18 1 1 (*) (*) - Node101102 Precentral Precentral -4 36 Gyrus Gyrus 18 (Brodman (Brodmannn area 6) area 6)Node 1021104929-9580-4034v.1103 Inferior Inferior 30 15 Parietal Parietal 17 Lobule Lobule(Brodman (Brodmannn area 40) area 40) - Node 103104 * (*) * (*) - Node -4104 24 14 105 Cuneus Cuneus -3 16 (Brodman (Brodmann 48 n area 19) area 19) - Node 105106 Inferior Inferior 24 -7 Temporal Temporal 22 Gyrus Gyrus(Brodman (Brodmannn area 20) area 20) - Node 106107 - - - Node 107 15 2119 108 Middle Middle 30 Temporal Temporal 10 12 Gyrus Gyrus(Brodman (Brodmannn area 21) area 21) - Node 108109 Superior Superior 7 36 1 Frontal FrontalGyrus Gyrus(Brodman (Brodmannn area 10) area 10) - Node 109110 - - - Node 110 1919 28 111 Culmen(*) Culmen (*) - Node 111 13 16 181114929-9580-4034v.1112 - - -Node 112 1 -9 40 113 Rectal Rectal Gyrus 5 17 Gyrus (Brodmann 15 (Brodman area 11) - n area 11) Node 113114 Lingual Lingual 1 -8 Gyrus Gyrus 50 (Brodman (Brodmannn area 18) area 18) - Node 114115 Inferior Inferior 33 -2 Temporal Temporal 30 Gyrus Gyrus(Brodman (Brodmannn area 37) area 37) - Node 115116 - - - Node 116 0 118 117 - - -Node 11719 20 29 118 - - - Node 118 -218 351124929-9580-4034v.1REFERENCESDang, Annie, et al. " Spatial Dissociation of Atrophy and Hypophysiology in Alzheimer's Disease: Implications for Biomarkers." Human Brain Mapping 46.16 (2025): e70402.Barron, D. S., Fox, P. M., Laird, A. R., Robinson, J. L., & Fox, P. T. (2012). Thalamic medial dorsal nucleus atrophy in medial temporal lobe epilepsy: A VBM meta-analysis. NeuroImage. Clinical, 2, 25-32. https: / / doi. Org / 10.1016 / j.nicl.2012.ll.004Bassett, D. S., & Sporns. O. (2017). Network neuroscience. Nature neuroscience, 20(3), 353-364. https: / / doi.org / 10.1038 / nn.4502Beckmann, C. F., & Smith, S. M. (2004). Probabilistic independent component analysis for functional magnetic resonance imaging. IEEE transactions on medical imaging, 23(2), 137— 152. https: / / doi.org / 10.1109 / TMI.2003.822821Chiang, F. L., Feng, M., Romero, R. S., Price, L., Franklin, C. G., Deng, S., Gray, J. P., Yu, F. F., Tantiwongkosi, B., Huang, S. Y., & Fox, P. T. (2021). Disruption of the Atrophy-based Functional Network in Multiple Sclerosis Is Associated with Clinical Disability: Validation of a Meta- Analytic Model in Resting-State Functional MRI. Radiology, 299(1), 159-166. https: / / doi.org / 10.1148 / radiol.2021203414Correa, N., Adali, T., & Calhoun, V. D. (2007). Performance of blind source separation algorithms for fMRI analysis using a group ICA method. Magnetic resonance imaging, 25(5), 684-694. https: / / doi. Org / 10.1016 / j.mri.2006.10.017Crossley, N. A., Mechelli, A., Vertes, P. E., Winton-Brown, T. T., Patel, A. X., Ginestet, C. E., McGuire, P., & Bullmore, E. T. (2013). Cognitive relevance of the community structure of the human brain functional coactivation network. Proceedings of the National Academy of Sciences of the United States of America, 110(28), 11583-11588. https: / / doi.org / 10.1073 / pnas.1220826110Eickhoff, S. B., Nichols, T. E., Laird, A. R., Hoffstaedter, F., Amunts, K., Fox, P. T., Bzdok, D., & Eickhoff, C. R. (2016). Behavior, sensitivity, and power of activation likelihood estimation characterized by massive empirical simulation. NeuroImage, 137, 70-85. https: / / doi.org / 10.1016 / j.neuroimage.2016.04.0721134929-9580-4034v.1Fox, P. T., Lancaster, J. L., Laird, A. R., & Eickhoff, S. B. (2014). Meta-analysis in human neuroimaging: computational modeling of large-scale databases. Annual review of neuroscience, 37, 409-434. https: / / doi.org / 10.1146 / annurev-neuro-062012-170320 Fox, P. T. (2023, September 28-30). BrainMap Community Portal: A Meta-Connectomics Resource [Keynote Talk], TACC Symposium for Texas Researchers, Austin, TXhttps: / / www. tacc. utexas. edu / taccster / 2022Frahm, L., Satterthwaite, T. D., Fox. P. T., Langner, R., & Eickhoff, S. B. (2023). ALE metaanalyses of voxel-based morphometry studies: Parameter validation via large-scale simulations. NeuroImage, 281, 120383. https: / / doi. Org / 10.1016 / j.neuroimage.2023.120383 Hunt, K. J., Knight, L. K., & Depue, B. E. (2023). Related neural networks underlie suppression of emotion, memory, motor processes as identified by data-driven analysis. BMC neuroscience, 24(1), 44. https: / / doi.org / 10.1186 / sl2868-023-00812-5Isvoranu A-M, Epskamp S, Waldorp LJ, Borsboom D. (2022). Network Psychometrics with R. Rutledge, New York, NY.Lu, Q. Y., Towne. J. M., Lock, M., Jiang, C., Cheng, Z. X., Habes, M., Zuo, X. N., & Zang, Y. F. (2022). Toward Coordinate-based Cognition Dictionaries: A BrainMap and Neurosynth Demo. Neuroscience, 493, 109-118. https: / / doi. Org / 10.1016 / j.neuroscience.2022.02.016 Marshall, C. R., Hardy, C. J. D., Russell, L. L, Bond, R. L., Sivasathiaseelan, H., Greaves, C., Moore, K. M., Agustus, J. L., van Leeuwen, J. E. P., Washing, S. J., Rohrer, J. D., Kilner, J. M., & Warren, J. D. (2019). The functional neuroanatomy of emotion processing in frontotemporal dementias. Brain: a journal of neurology, 142(9), 2873-2887.https: / / doi. org / 10.1093 / brain / awz204McKeown, M. J., Makeig, S., Brown. G. G., Jung, T. P., Kindermann, S. S., Bell, A. J., & Sejnowski. T. J. (1998). Analysis of fMRI data by blind separation into independent spatial components. Human brain mapping, 6(3), 160-188. https: / / doi.org / 10.1002 / (SICI)1097-0193(1998)6:3<160:: AID-HBM5>3.0. CO;2-1Nedic, S., Stufflebeam, S. M., Rondinoni, C., Velasco, T. R., dos Santos, A. C., Leite, J. P., Gargaro, A. C., Mujica-Parodi, L. R., & Ide, J. S. (2015). Using network dynamic fMRI for detection of epileptogenic foci. BMC neurology, 15, 262. https: / / doi.org / 10.1186 / sl2883-015-0514-y1144929-9580-4034v.1Seeley, W. W., Crawford, R. K., Zhou, J., Miller, B. L., & Greicius, M. D. (2009).Neurodegenerative diseases target large-scale human brain networks. Neuron, 62(1). 42-52. https: / / doi. org / 10.1016 / j.neuron.2009.03.024Smith, S. M., Fox, P. T., Miller, K. L., Glahn, D. C., Fox, P. M., Mackay, C. E., Filippini, N., Watkins, K. E., Toro, R., Laird, A. R., & Beckmann, C. F. (2009). Correspondence of the brain's functional architecture during activation and rest. Proceedings of the National Academy of Sciences of the United States of America, 106(31), 13040-13045. https: / / doi.org / 10.1073 / pnas.0905267106Towne, J. M., Eslami, V., Fox, P. M., Cavazos, J. E., & Fox, P. T. (2022). Discovering Pathophysiologic Networks of Temporal Lobe Epilepsy Using the BrainMap Community Portal through the Texas Advanced Computing Center. The University of Texas at Austin. https: / / doi.org / 10.26153 / TSW / 43750Vanasse. T. J., Fox, P. T., Fox, P. M., Cauda, F., Costa. T., Smith, S. M.. Eickhoff. S. B., & Lancaster, J. L. (2021). Brain pathology recapitulates physiology: A network meta-analysis. Communications biology, 4(1), 301. https: / / doi.org / 10.1038 / s42003-021-01832-9 Wagenmakers E. J. (2007). A practical solution to the pervasive problems of p values.Psychonomic bulletin & review, 14(5), 779-804. https: / / doi.org / 10.3758 / bf03194105 Acar. F., Seurinck, R., Eickhoff, S. B., & Moerkerke, B. (2018). Assessing robustness against potential publication bias in Activation Likelihood Estimation (ALE) meta-analyses for fMRI. PloS one, 13(11), e0208177.Albajes-Eizagirre, A., Solanes, A., Fullana, M. A., loannidis, J. P. A., Fusar-Poli, P., Torrent, C., Sole, B., Bonnin, C. M., Vieta, E., Mataix-Cols, D., & Radua, J. (2019). Meta-analysis of Voxel-Based Neuroimaging Studies using Seed-based d Mapping with Permutation of Subject Images (SDM-PS1). Journal of visualized experiments: JoVE, (153), 10.3791 / 59841. Albajes-Eizagirre, A., Solanes, A., Vieta, E., & Radua, J. (2019). Voxel-based meta-analysis via permutation of subject images (PSI): Th eon and implementation for SDM. NeuroImage, 186, 174-184.APA. (2013). Diagnostic and Statistical Manual of Mental Disorders: DSM-5 (5th Edition). American Psychiatric Publishing, a division of American Psychiatric Association.1154929-9580-4034v.1APA. (2022). Diagnostic and Statistical Manual of Mental Disorders: DSM-5-TR (5th Edition Text Revision). American Psychiatric Publishing, a division of American Psychiatric Association.Armada-Moreira, A., Gomes, J. I., Pina, C. C., Savchak, O. K., Gonpalves-Ribeiro, J., Rei, N., Pinto, S., Morais, T. P., Martins, R. S., Ribeiro, F. F., Sebastiao, A. M., Crunelli, V., & Vaz, S. H. (2020). Going the Extra (Synaptic) Mile: Excitotoxicity as the Road Toward Neurodegenerative Diseases. Frontiers in cellular neuroscience. 14. 90.Ashare, R. L., Wileyto, E. P., Ruparel, K.. Goelz, P. M., Hopson. R. D., Valdez, J. N., Gur, R. C., Loughead, J., & Lerman, C. (2013). Effects of tolcapone on working memory and brain activity in abstinent smokers: a proof-of-concept study. Drug and alcohol dependence, 133(3), 852-856.Ashburner, J., & Friston. K. J. (2000). Voxel-based morphometry— the methods. NeuroImage, 11(6 Pt 1), 805-821.Bechara, A. (2005). Decision making, impulse control and loss of willpower to resist drugs: a neurocognitive perspective. Nature neuroscience, 8(11), 1458-1463.Bolla, K. I., Eldreth, D. A., Matochik, J. A., & Cadet, J. L. (2005). Neural substrates of faulty decision-making in abstinent marijuana users. Neuroimage, 26(2), 480-492.Brecht, M. L., & Herbeck, D. (2014). Time to relapse following treatment for methamphetamine use: a long-term perspective on patterns and predictors. Drug and alcohol dependence, 139, 18-25.Button, K. S., loannidis, J. P., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S., & Munafd, M. R. (2013). Power failure: why small sample size undermines the reliability of neuroscience. Nature reviews neuroscience, 14(5), 365-376.Buckner, R. L., & Krienen, F. M. (2013). The evolution of distributed association networks in the human brain. Trends in cognitive sciences, 17(12), 648-665.Camchong, J., Collins, P. F., Becker, M. P., Lim, K. O., & Luciana, M. (2019). Longitudinal alterations in prefrontal resting brain connectivity in non-treatment-seeking young adults with cannabis use disorder. Frontiers in Psychiatry, 10, 514.1164929-9580-4034v.1Cauda, F, Nani, A., Manuello, J., Premi, E., Palermo, S., Tatu, K., Duca, S., Fox, P. T., & Costa, T. (2018). Brain structural alterations are distributed following functional, anatomic and genetic connectivity. Brain: a journal of neurology, 141(11), 3211-3232.Chiang, F. L., Wang, Q, Yu, F. F., Romero, R. S., Huang, S. Y., Fox, P. M., Tantiwongkosi, B., & Fox, P. T. (2019). Localised grey matter atrophy in multiple sclerosis is network-based: a coordinate-based meta-analysis. Clinical radiology, 74(10), 816.el9-816.e28.Cogdell-Brooke, L. S., Sowden, P. T., Violante, I. R., & Thompson, H. E. (2020). A meta-analysis of functional magnetic resonance imaging studies of divergent thinking using activation likelihood estimation. Human brain mapping, 41(17), 5057-5077.Cole, D. M., Smith, S. M., & Beckmann, C. F. (2010). Advances and pitfalls in the analysis and interpretation of resting-state FMRI data. Frontiers in systems neuroscience, 4, 8.Connery, H. S., McHugh, R. K., Reilly, M., Shin, S., & Greenfield, S. F. (2020). Substance use disorders in global mental health delivery: epidemiology, treatment gap, and implementation of evidence-based treatments. Harvard review of psychiatry, 28(5), 316-327. Costa, T., Ferraro, M., Manuello, J., Camasio, A., Nani, A., Mancuso, L., Cauda, F., Fox, P. T., & Liloia, D. (2024). Activation Likelihood Estimation Neuroimaging Meta-Analysis: a Powerful Tool for Emotion Research. Psychology research and behavior management, 17, 2331-2345.Crockford, D. N., Goodyear. B.. Edwards. J., Quickfall. J., & El-Guebaly, N. (2005). Cue-induced brain activity in pathological gamblers. Biological psychiatry, 58(10), 787-795. Crossley, N. A., Mechelli, A., Scott, J., Carletti, F., Fox, P. T., McGuire, P., & Bullmore, E. T. (2014). The hubs of the human connectome are generally implicated in the anatomy of brain disorders. Brain, 137(8), 2382-2395.Dalal, P. K., & Sivakumar, T. (2009). Moving towards ICD-11 and DSM-V: Concept and evolution of psychiatric classification. Indian journal of psychiatry. 51(4), 310-319.Damoiseaux, J. S., Rombouts, S., Barkhof, F., Scheltens, P., Stam, C. J., Smith, S. M., & Beckmann, C. F. (2006). Consistent resting-state networks across healthy subjects.Proceedings of the national academy of sciences, 103(37), 13848-13853.1174929-9580-4034v.1Dang, A., Wang, D., Towne, J., Habes, M., & Fox, P. T. (2024). Dissociation of Structural and Functional Changes in Alzheimer's Disease and Mild Cognitive Impairment. Alzheimer's Dement., 20: e087663.Damai, G., Perlaki, G., Zsido, A. N., Inhof, O., Orsi, G., Horvath, R, Nagy, S. A., Labadi, B., Tenyi, D., & Kovacs, N. (2019). Internet addiction and functional brain networks: task-related fMRI study. Scientific Reports, 9(1), 15777.Deng, S., Franklin. C. G., O'Boyle, M., Zhang, W., Heyl, B. L., Jerabek, P. A., Lu, H., & Fox, P. T. (2022). Hemodynamic and metabolic correspondence of resting-state voxel-based physiological metrics in healthy adults. NeuroImage, 250, 118923.Eickhoff, S. B., Bzdok, D., Laird, A. R., Kurth, F., & Fox, P. T. (2012). Activation likelihood estimation meta-analysis revisited. Neuroimage, 59(3), 2349-2361.Eickhoff, S. B., Laird, A. R., Grefkes, C., Wang, L. E., Zilles, K., & Fox, P. T. (2009).Coordinate-based activation likelihood estimation meta-analysis of neuroimaging data: A random-effects approach based on empirical estimates of spatial uncertainty. Human brain mapping, 30(9), 2907-2926.Eickhoff, S. B., Nichols, T. E., Laird, A. R., Hoffstaedter, F., Amunts, K., Fox, P. T., Bzdok, D., & Eickhoff, C. R. (2016). Behavior, sensitivity', and power of activation likelihood estimation characterized by massive empirical simulation. Neuroimage, 137, 70-85.Eklund, A., Nichols, T. E.. & Knutsson, H. (2016). Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates. Proceedings of the national academy of sciences, 113(28), 7900-7905.El-Baba, R. M., & Schury, M. P. (2023). Neuroanatomy, Frontal Cortex. Treasure Island (FL): StatPearls Publishing.Euston, D. R, Gruber, A. J., & McNaughton, B. L. (2012). The role of medial prefrontal cortex in memory and decision making. Neuron, 76(6), 1057-1070.Everitt, B. J. (2014). Neural and psychological mechanisms underlying compulsive drug seeking habits and drug memories-indications for novel treatments of addiction. European Journal of Neuroscience, 40(1), 2163-2182.Everitt, B. J., & Robbins, T. W. (2016). Drug addiction: updating actions to habits to compulsions ten years on. Annual Review of Psychology. 67. 23-50.1184929-9580-4034v.1Fedota, J. R., Ding, X., Matous, A. L., Salmeron, B. J., McKenna, M. R., Gu, H., Ross, T. J., & Stein, E. A. (2018). Nicotine abstinence influences the calculation of salience in discrete insular circuits. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 3(2), 150-159.Fleury, M. J., Djouini, A., Huynh, C., Tremblay, J., Ferland, F., Menard, J. M., & Belleville, G. (2016). Remission from substance use disorders: A systematic review and meta-analysis. Drug and alcohol dependence. 168, 293-306.Fox, M. D., Corbetta, M., Snyder, A. Z.. Vincent. J. L., & Raichle, M. E. (2006).Spontaneous neuronal activity distinguishes human dorsal and ventral attention systems. Proceedings of the national academy of sciences, 103(26), 10046-10051.Fox, H., Hong, K., & Sinha, R. (2008). Difficulties in emotion regulation and impulse control in recently abstinent alcoholics compared with social drinkers. Addictive Behaviors, 33(2), 388-394.Fox, P. T., Laird, A. R., Fox, S. P., Fox, P. M., flecker, A. M., Crank, M., Koenig, S. F., & Lancaster, J. L. (2005). BrainMap taxonomy of experimental design: description and evaluation. Human brain mapping, 25(1), 185-198.Fox, P. T., Lancaster, J. L., Laird, A. R., & Eickhoff, S. B. (2014). Meta-analysis in human neuroimaging: computational modeling of large-scale databases. Annual review of neuroscience, 37, 409-434.Fox, P. T., Parsons, L. M., & Lancaster, J. L. (1998). Beyond the single study: function / location metanalysis in cognitive neuroimaging. Current opinion in neurobiology, 8(2), 178-187.Frahm, L., Cieslik, E. C., Hoffstaedter, F., Satterthwaite, T. D., Fox, P. T., Langner, R., & Eickhoff. S. B. (2022). Evaluation of thresholding methods for activation likelihood estimation meta-analysis via large-scale simulations. Human brain mapping, 43(13), 3987-3997.Frahm, L., Satterthwaite, T. D., Fox, P. T., Langner, R., & Eickhoff, S. B. (2023). ALE metaanalyses of voxel-based morphometry studies: Parameter validation via large-scale simulations. NeuroImage, 281, 120383.1194929-9580-4034v.1Franklin, T. R., Acton, P. D., Maldjian, J. A., Gray, J. D., Croft, J. R., Dackis, C. A., O'Brien, C. P.. & Childress. A. R. (2002). Decreased gray matter concentration in the insular, orbitofrontal, cingulate, and temporal cortices of cocaine patients. Biological psychiatry, 51(2), 134-142.Goldstein, R. Z., Bechara, A., Garavan, H., Childress, A. R., Paulus, M. P., & Volkow, N. D. (2009). The neurocircuitry of impaired insight in drug addiction. Trends in cognitive sciences, 13(9), 372-380.Goldstein, R. Z., & Volkow. N. D. (2011). Dysfunction of the prefrontal cortex in addiction: neuroimaging findings and clinical implications. Nature reviews neuroscience, 12(11), 652-669.Goodkind, M., Eickhoff, S. B., Oathes, D. J., Jiang, Y., Chang, A., Jones-Hagata, L. B., Ortega, B. N., Zaiko, Y. V., Roach, E. L., & Korgaonkar, M. S. (2015). Identification of a common neurobiological substrate for mental illness. JAMA psychiatry, 72(4). 305-315. Goudriaan, A. E., De Ruiter, M. B., Van Den Brink, W., Oosterlaan, J., & Veltman, D. J. (2010). Brain activation patterns associated with cue reactivity and craving in abstinent problem gamblers, heavy smokers and healthy controls: an fMRI study. Addiction biology, 15(4), 491-503.Gray, J. P., Muller. V. I.. Eickhoff, S. B., & Fox, P. T. (2020). Multimodal Abnormalities of Brain Structure and Function in Major Depressive Disorder: A Meta- Analysis of Neuroimaging Studies. The American journal of psychiatry, 177(5), 422-434.Haber, S. N., & Knutson, B. (2010). The reward circuit: linking primate anatomy and human imaging. Neuropsychopharmacology, 35(1), 4-26.He, Y., Wang, L., Zang, Y., Tian, L., Zhang, X., Li, K., & Jiang, T. (2007). Regional coherence changes in the early stages of Alzheimer's disease: a combined structural and resting-state functional MRI study. NeuroImage, 35(2), 488-500.Heilig, M., & Leggio, L. (2016). What the alcohol doctor ordered from the neuroscientist: theragnostic biomarkers for personalized treatments. Progress in brain research, 224, 401-418.1204929-9580-4034v.1Heilig, M., MacKillop, J., Martinez, D., Rehm, J., Leggio, L., & Vanderschuren, L. J. (2021). Addiction as a brain disease revised: why it still matters, and the need for consilience.Neuropsychopharmacology, 46(10), 1715-1723.Heikkinen, N., Niskanen, E., Kononen, M., Tolmunen, T., Kekkonen, V., Kivimaki, P, Tanila, H., Laukkanen, E., Vanninen, R. (2017). Alcohol consumption during adolescence is associated with reduced grey matter volumes. Addiction, 112(4), 604-613.Ide, J. S., Zhang, S., Hu, S., Sinha, R., Mazure, C. M., Li, C. R. (2014). Cerebral gray matter volumes and low-frequency fluctuation of BOLD signals in cocaine dependence: duration of use and gender difference. Drug and Alcohol Dependence, 134, 51-62.Jentsch, J. D., & Taylor, J. R. (1999). Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli.Psychopharmacology, 146, 373-390.Koob, G. F., & Volkow, N. D. (2010). Neurocircuitry of addiction.Neuropsychopharmacology, 35(1), 217-238.Koob, G. F., & Volkow, N. D. (2016). Neurobiology of addiction: a neurocircuitry analysis. The Lancet Psychiatry, 3(8), 760-773.Kotkowski, E., Price, L. R., Franklin, C., Salazar, M., Woolsey, M., DeFronzo, R. A., Blangero, J., Glahn, D. C., & Fox, P. T. (2019). A neural signature of metabolic syndrome. Human brain mapping, 40(12), 3575-3588.Kotov, R., Krueger, R. F., Watson, D., Achenbach, T. M., Althoff, R. R.. Bagby. R. M., Brown, T. A., Carpenter, W. T., Caspi, A., Clark, L. A., Eaton, N. R., Forbes, M. K, Forbush, K. T., Goldberg, D., Hasin, D., Hyman, S. E., Ivanova, M. Y., Lynam, D. R., Markon, K, Miller, J. D.,... Zimmerman, M. (2017). The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of abnormal psychology, 126(4), 454-477.Kuo, L.-W., Lin, P.-S., Lin, S.-Y., Liu, M.-F., Jan, H., Lee, H.-C., & Wang, S.-C. (2019). Functional correlates of resting-state connectivity in the default mode netw ork of heroin users on methadone treatment and medication-free therapeutic community program. Frontiers in Psychiatry. 10. 381.1214929-9580-4034v.1Laird, A. R., Eickhoff, S. B., Kurth, F., Fox, P. M., Uecker, A. M., Turner, J. A., Robinson, J. L., Lancaster, J. L., & Fox, P. T. (2009). ALE meta-analysis workflows via the brainmap database: progress towards a probabilistic functional brain atlas. Frontiers in neuroinformatics, 23.Laird, A. R., Eickhoff, S. B., Li, K., Robin, D. A., Glahn, D. C., & Fox, P. T. (2009).Investigating the functional heterogeneity of the default mode network using coordinate-based meta-analytic modeling. Journal of neuroscience. 29(46). 14496-14505.Laird, A. R., Fox, P. M.. Eickhoff. S. B., Turner, J. A., Ray, K. L., McKay. D. R., Glahn, D. C., Beckmann, C. F., Smith, S. M., & Fox, P. T. (2011). Behavioral interpretations of intrinsic connectivity networks. Journal of cognitive neuroscience, 23(12), 4022-4037.Laird, A. R., Robinson, J. L., McMillan, K. M., Tordesillas-Gutierrez, D., Moran, S. T., Gonzales, S. M., Ray, K. L., Franklin, C., Glahn, D. C., Fox, P. T., & Lancaster, J. L. (2010). Comparison of the disparity between Talairach and MNI coordinates in functional neuroimaging data: validation of the Lancaster transform. NeuroImage, 51(2), 677-683. Lancaster, J. L., Laird, A. R., Eickhoff, S. B., Martinez, M. J., Fox, P. M., & Fox, P. T.(2012). Automated regional behavioral analysis for human brain images. Frontiers in neuroinformatics, 6, 23.Lancaster, J. L., Tordesillas-Gutierrez, D.. Martinez, M., Salinas, F., Evans, A., Zilles, K, Mazziotta, J. C.. & Fox, P. T. (2007). Bias between MNI and Talairach coordinates analyzed using the ICBM-152 brain template. Human brain mapping, 28(11), 1194-1205.Le, T. M., Potvin, S., Zhomitsky, S., & Li, C.-S. R. (2021). Distinct patterns of prefrontal cortical disengagement during inhibitory control in addiction: A meta-analysis based on population characteristics. Neuroscience & Biobehavioral Reviews, 127, 255-269.Lesage, E.. & Stein, E. A. (2016). Networks Associated with Reward. In: Pfaff. D., Volkow, N. (eds) Neuroscience in the 21st Century. Springer, New York, NY.Li, S., Demenescu, L. R., Sweeney-Reed, C. M., Krause, A. L, Metzger, C. D., & Walter, M. (2017). Novelty seeking and reward dependence-related large-scale brain networks functional connectivity variation during salience expectancy. Human brain mapping, 38(8), 4064-4077.1224929-9580-4034v.1Li, W., Li, Y., Yang, W., Zhang. Q., Wei, D., Li, W., Hitchman, G., & Qiu, J. (2015). Brain structures and functional connectivity associated with individual differences in Internet tendency in healthy young adults. Neuropsychologia, 70, 134-144.Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gotzsche, P. C, loannidis, J. P., Clarke, M., Devereaux, P. J., Kleijnen, J., & Moher, D. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS medicine, 6(7), el 000100.Loughead, J., Wileyto. E., Valdez. J., Sanborn, P., Tang, K., Strasser. A.. Ruparel, K., Ray, R., Gur, R., & Lerman, C. (2009). Effect of abstinence challenge on brain function and cognition in smokers differs by COMT genotype. Molecular psychiatry, 14(8), 820-826. McTeague, L. M., Huemer, J., Carreon, D. M., Jiang, Y., Eickhoff, S. B., & Etkin, A. (2017). Identification of common neural circuit disruptions in cognitive control across psychiatric disorders. American Journal of Psychiatry, 174(7), 676-685.Mei, B., Tao. Q., Dang, J., Niu, X., Sun, J., Zhang, M., Wang, W., Han, S., Zhang, Y., & Cheng, J. (2024). Meta-analysis of structural and functional abnormalities in behavioral addictions. Addictive behaviors, 157, 108088.Menon, V. (2011). Large-scale brain networks and psychopathology: a unifying triple network model. Trends in cognitive sciences, 15(10), 483-506.Menon, V.. & Uddin, L. Q. (2010). Saliency, switching, attention and control: a network model of insula function. Brain Structure and Function, 214, 655-667.Mohan, A., Roberto, A. J., Mohan, A., Lorenzo, A., Jones, K., Camey, M. J., Liogier-Weyback, L., Hwang, S., & Lapidus, K. A. (2016). Focus: the aging brain: the significance of the default mode network (DMN) in neurological and neuropsychiatric disorders: a review. The Yale journal of biology and medicine, 89(1), 49.Moring, J. C., Husain, F. T.. Gray. J., Franklin, C., Peterson, A. L., Resick. P. A., Garrett, A., Esquivel, C., & Fox, P. T. (2022). Invariant structural and functional brain regions associated with tinnitus: A meta-analysis. PloS one, 17(10), e0276140.Muller, V. I., Cieslik, E. C., Laird, A. R., Fox, P. T., Radua, J., Mataix-Cols, D., Tench, C. R., Yarkoni, T., Nichols, T. E., & Turkeltaub, P. E. (2018). Ten simple rules for neuroimaging meta-analysis. Neuroscience & Biobehavioral Reviews, 84, 151-161.1234929-9580-4034v.1Oosterwijk, S., Lindquist, K. A., Anderson, E., Dautoff, R., Moriguchi, Y., & Barrett, L. F. (2012). States of mind: Emotions, body feelings, and thoughts share distributed neural networks. Neuroimage, 62(3), 2110-2128.Parvaz, M. A., Rabin, R. A., Adams, F., & Goldstein, R. Z. (2022). Structural and functional brain recovery in individuals with substance use disorders during abstinence: A review of longitudinal neuroimaging studies. Drug and alcohol dependence, 232, 109319.Peng, P., Wang, Z., Jiang, T., Chu, S., Wang, S., Xiao, D. (2017). Brain-volume changes in young and middle-aged smokers: a DARTEL-based voxel-based morphometry study.Clinical Respiratory Journal, 11(5), 621-631.Peng, P., Li, M., Liu, H., Tian, Y. R., Chu, S. L., Van Halm-Lutterodt, N., Jing, B., Jiang, T. (2018). Brain structure alterations in respect to tobacco consumption and nicotine dependence: a comparative voxel-based morphometry study. Frontiers in Neuroanatomy, 12, 43.Perry, B. A. L., Mendez, J. C., & Mitchell, A. S. (2023). Cortico-thalamocortical interactions for learning, memory and decision-making. The Journal of physiology, 601(1), 25-35.Peters, S. K., Dunlop, K., & Downar, J. (2016). Cortico-striatal-thalamic loop circuits of the salience network: a central pathway in psychiatric disease and treatment. Frontiers in systems neuroscience, 10, 104.Proctor, D., Williamson. R. A., Latzman, R. D.. de Waal, F. B.. & Brosnan, S. F. (2014). Gambling primates: reactions to a modified Iowa Gambling Task in humans, chimpanzees and capuchin monkeys. Animal cognition, 17(4), 983-995.Qin, K., Zhang, F., Chen, T., Li, L., Li, W., Suo, X., Lei, D., Kemp, G. J., & Gong, Q.(2020). Shared gray matter alterations in individuals with diverse behavioral addictions: A voxel-wise meta-analysis. Journal of behavioral addictions, 9(1), 44-57.Radua. J., Rubia. K. Canales-Rodriguez, E. J., Pomarol-Clotet. E., Fusar-Poli, P., & Mataix-Cols, D. (2014). Anisotropic kernels for coordinate-based meta-analyses of neuroimaging studies. Frontiers in psychiatry, 5, 13.Robinson, S. M., & Adinoff, B. (2016). The Classification of Substance Use Disorders: Historical, Contextual, and Conceptual Considerations. Behavioral sciences (Basel, Switzerland), 6(3). 18.1244929-9580-4034v.1Robinson, J. L., Laird, A. R., Glahn, D. C., Blangero, J., Sanghera, M. K., Pessoa, L., Fox, P. M.. Uecker, A., Friehs, G.. & Young, K. A. (2012). The functional connectivity of the human caudate: an application of meta-analytic connectivity modeling with behavioral filtering. Neuroimage, 60(1), 117-129.Rosenthal, R. (1979). The file drawer problem and tolerance for null results. Psychological Bulletin, 86(3). 638-641Salo, T., Yarkoni, T, Nichols, T. E., Poline, J. P., Bilgel, M., Bottenhorn, K. L., Eickhoff, S. B., Jarecka, D., Kent, J. D., Kimbler, A.. Nielson, D. M.. Oudyk, K. M., Peraza, J. A., Perez. A., Reeders, P. C., Yanes, J. A., & Laird, A. R. (2023). NiMARE: Neuroimaging Meta- Analysis Research Environment. Aperture Neuro, 3, 1-32SAMHSA. (2021). 2021 National Survey on Drug Use and Health (NSDUH). Substance Abuse and Mental Health Services Administration (SAMHSA).Samartsidis, P., Montagna, S., Laird, A. R., Fox, P. T., Johnson, T. D., & Nichols, T. E. (2020). Estimating the prevalence of missing experiments in a neuroimaging meta-analysis. Research synthesis methods, 11(6), 866-883.Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., Reiss, A. L., & Greicius, M. D. (2007). Dissociable intrinsic connectivity networks for salience processing and executive control. Journal of neuroscience, 27(9), 2349-2356.Seeley. W. W.. Crawford, R. K.. Zhou, J., Miller, B. L.. & Greicius, M. D. (2009).Neurodegenerative diseases target large-scale human brain networks. Neuron, 62(1), 42-52. Sha, Z., Wager, T. D., Mechelli, A., & He, Y. (2019). Common dysfunction of large-scale neurocognitive networks across psychiatric disorders. Biological psychiatry, 85(5), 379-388. Smith, S. M., Fox, P. T., Miller, K. L., Glahn, D. C., Fox, P. M., Mackay, C. E., Filippini, N., Watkins, K. E., Toro, R., & Laird, A. R. (2009). Correspondence of the brain's functional architecture during activation and rest. Proceedings of the national academy of sciences, 106(31), 13040-13045.Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E., Johansen-Berg, H., Bannister, P. R., De Luca, M., Drobnjak, L, & Flitney, D. E. (2004). Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage, 23, S208-S219.1254929-9580-4034v.1Sprooten, E., Rasgon, A., Goodman. M., Carlin, A., Leibu, E., Lee, W. H., & Frangou, S. (2017). Addressing reverse inference in psychiatric neuroimaging: Meta-analyses of task-related brain activation in common mental disorders. Human brain mapping, 38(4), 1846- 1864.Sridharan, D., Levitin, D. J., & Menon, V. (2008). A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. Proceedings of the national academy of sciences, 105(34), 12569-12574.Sussman, S., & Sussman, A. N. (2011). Considering the definition of addiction. International journal of environmental research and public health, 8(10), 4025-4038.Sutherland, M. T., McHugh, M. J., Pariyadath, V., & Stein, E. A. (2012). Resting state functional connectivity in addiction: Lessons learned and a road ahead. Neuroimage, 62(4), 2281-2295.Sylvester, C. M.. Corbetta. M., Raichle, M., Rodebaugh, T., Schlaggar, B., Sheline, Y., Zorumski, C., & Lenze, E. (2012). Functional network dysfunction in anxiety and anxiety disorders. Trends in neurosciences, 35(9), 527-535.Tahmasian, M., Sepehry, A. A., Samea, F., Khodadadifar, T., Soltaninejad, Z., Javaheripour, N., Khazaie, H., Zarei, M., Eickhoff, S. B., & Eickhoff, C. R. (2019). Practical recommendations to conduct a neuroimaging meta-analysis for neuropsychiatric disorders. Human brain mapping, 40(17), 5142-5154.Tenreiro, S., Eckermann, K., & Outeiro, T. F. (2014). Protein phosphorylation in neurodegeneration: friend or foe?. Frontiers in molecular neuroscience, 7, 42.Towne, J., Eslami, V., Cavazos, J. E., & Fox, P. T. (2022). Resting-State Voxel-Based Pathophysiologic Changes in Medial Temporal Lobe Epilepsy (P4-8.004). Neurology 98 (18 Supplement), 3643.Turkeltaub, P. E., Eden, G. F.. Jones, K. M.. & Zeffiro, T. A. (2002). Meta-analysis of the functional neuroanatomy of single-word reading: method and validation. Neuroimage, 16(3), 765-780.Turkeltaub, P. E., Eickhoff, S. B., Laird, A. R., Fox, M., Wiener, M., & Fox, P. (2012). Minimizing within-experiment and within-group effects in activation likelihood estimation meta-analyses. Human brain mapping, 33(1), 1-13.1264929-9580-4034v.1Uddin, L. Q., Clare Kelly, A., Biswal, B. B., Xavier Castellanos, F., & Milham, M. P. (2009). Functional connectivity of default mode network components: correlation, anticorrelation, and causality. Human brain mapping, 30(2), 625-637.Vanasse, T. J., Fox, P. M., Barron, D. S., Robertson, M., Eickhoff, S. B., Lancaster, J. L., & Fox, P. T. (2018). BrainMap VBM: An environment for structural meta-analysis. Human brain mapping, 39(8), 3308-3325.Vanasse, T. J., Fox, P. T, Fox, P. M., Cauda, F., Costa, T., Smith, S. M., Eickhoff, S. B., & Lancaster, J. L. (2021). Brain pathology recapitulates physiology: A network meta-analysis. Communications biology, 4(1), 301.Volkow, N. D., Fowler, J. S., & Wang, G.-J. (2003). The addicted human brain: insights from imaging studies. The Journal of clinical investigation, 111(10), 1444-1451.Volkow, N. D., & Baler, R. D. (2014). Addiction science: Uncovering neurobiological complexity. Neuropharmacology, 76, 235-249.Wager. T. D., Lindquist. M.. & Kaplan, L. (2007). Meta-analysis of functional neuroimaging data: current and future directions. Social cognitive and affective neuroscience, 2(2), 150— 158.Wang, L., Shen, H., Lei, Y., Zeng, L.-L., Cao, F., Su, L., Yang, Z., Yao, S„ & Hu, D. (2017). Altered default mode, fronto-parietal and salience networks in adolescents with Internet addiction. Addictive Behaviors, 70, 1-6.Watanabe, K., & Hikosaka, O. (2005). Immediate changes in anticipators’ activity of caudate neurons associated with reversal of position-reward contingency. Journal of Neurophysiology, 94(3), 1879-1887.WHO. (2018). International Classification of Diseases 11th Revision. World Health Organization.WHO. (2025). International Classification of Diseases 11th Revision Update. World Health Organization.Williams, L. M. (2016). Precision psychiatry: a neural circuit taxonomy for depression and anxiety. The Lancet Psychiatry, 3(5), 472-480.1274929-9580-4034v.1Woo, C. W., Krishnan, A., & Wager, T. D. (2014). Cluster-extent based thresholding in fMRI analyses: pitfalls and recommendations. NeuroImage. 91, 412-419.Wright, I. C.. McGuire, P. K., Poline, J. B., Travere, J. M., Murray. R. M., Frith, C. D., Frackowiak, R. S., & Friston, K. J. (1995). A voxel-based method for the statistical analysis of gray and white matter density applied to schizophrenia. NeuroImage, 2(4), 244-252. Wu, P. H., & Schulz, K. M. (2012). Advancing addiction treatment: what can we leam from animal studies? ILAR journal, 53(1), 4-13.Xiong, A., Lai. X., Wu, S„ Yuan, X, Tang, J., Chen, J.. Liu, Y, & Hu. M. (2021).Relationship between cognitive fusion, experiential avoidance, and obsessive-compulsive symptoms in patients with obsessive-compulsive disorder. Frontiers in Psychology, 12, 655154.Yan, H., Xiao, S., Fu, S., Gong, J., Qi, Z., Chen, G., Chen, P., Tang, G., Su, T., Yang, Z., & Wang, Y. (2023). Functional and structural brain abnormalities in substance use disorder: A multimodal meta-analysis of neuroimaging studies. Acta psychiatrica Scandinavica, 147(4), 345-359.Yeung A. W. K. (2023). Do " Ten simple rules for neuroimaging meta-analysis" receive equal attention and accurate quotation? An examination on the quotations to an influential neuroimaging meta-analysis guideline. NeuroImage. Clinical. 39. 103496.Yeung. A. W. K., Robertson, M., Uecker, A., Fox, P. T., & Eickhoff. S. B. (2023). Trends in the sample size, statistics, and contributions to the BrainMap database of activation likelihood estimation meta-analyses: An empirical study of 10-year data. Human brain mapping, 44(5), 1876-1887.Yuan, K., Qin, W., Liu, Y., & Tian, J. (2011). Internet addiction: Neuroimaging findings. Communicative & integrative biology, 4(6). 637-639.Yuan, K., Qin. W.. Wang, G., Zeng. F., Zhao. L., Yang. X.. Liu, P., Liu, J.. Sun, J., von Deneen, K. M., Gong, Q., Liu, Y., & Tian, J. (2011). Microstructure abnormalities in adolescents with internet addiction disorder. PloS one, 6(6), e20708.Zang, Y., Jiang, T., Lu, Y., He, Y., & Tian, L. (2004). Regional homogeneity approach to IMRI data analysis. NeuroImage, 22(1), 394-400.1284929-9580-4034v.1Zhang, M., Gao, X., Yang, Z., Wen, M., Huang, H., Zheng, R., Wang, W., Wei, Y., Cheng, J., & Han, S. (2021). Shared gray mater alterations in subtypes of addiction: a voxel-wise meta-analysis. Psychopharmacology, 238, 2365-2379.Zhang, R., & Volkow, N. D. (2019). Brain default-mode network dysfunction in addiction. Neuroimage, 200, 313-331.Zhao, Y., Sallie, S. N., Cui, H., Zeng, N., Du, J., Yuan, T., Li, D., De Ridder, D., & Zhang, C. (2021). Anterior cingulate cortex in addiction: new insights for neuromodulation.Neuromodulation: Technology at the Neural Interface, 24(2). 187-196.Zhou, Y„ Lin, F. C, Du, Y. S„ Qin, L. D„ Zhao, Z. M„ Xu, J. R„ & Lei, H. (2011). Gray mater abnormalities in Internet addiction: a voxel-based morphometry study. European journal of radiology, 79(1), 92-95.Zhu, Y., Zhang, H., & Tian, M. (2015). Molecular and functional imaging of internet addiction. BioMed research international, 2015(1), 378675.Zou, Q. H.. Zhu, C. Z.. Yang, Y., Zuo, X. N., Long. X. Y, Cao, Q. J., Wang. Y. F., & Zang, Y. F. (2008). An improved approach to detection of amplitude of low-frequency fluctuation (ALFF) for resting-state fMRI: fractional ALFF. Journal of neuroscience methods, 172(1), 137-141.1294929-9580-4034v.1

Claims

CLAIMSWhat is claimed is:

1. A method of treating a subject for a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes comprising:a) imaging brain activity of said subject;b) comparing the imaged brain activity to a database of virtual representations of brain networks associated with a neurologic disorder, a psychiatric disorder, a development disorder, or age-associated changes, each network comprising a morphologic property’ and a physiologic property and, optionally, a temporal property,wherein the virtual representations were generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information comprising:i) coordinate-based quantified brain activity data related to specific human subject tasks and stored in a task-activation database (TA DB);ii) voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); andiii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); andwherein the CBMA algorithm comprises a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm; and c) determining, based on the comparison, if the subject exhibits a brain network associated with a specific neurologic disorder, a specific psychiatric disorder, a specific development disorder, or specific age-associated changes, d) wherein if the subject does exhibit said brain network associated with the specific neurologic disorder, the specific psychiatric disorder, the specific development disorder, or specific age-associated changes, administering a treatment to said subject for said neurologic disorder, psychiatric disorder, development disorder or age-associated changes,wherein the treatment comprises a treatment device which has a spatial placement component, and the spatial placement component is determined by1304929-9580-4034v.1the locations in the imaged brain activity which correspond most strongly to said brain network.

2. The method of claim 1, wherein the treatment comprises transcranial magnetic brain stimulation, direct current brain stimulation, alternating current brain stimulation, deep brain stimulation, and / or focused ultrasound brain stimulation.

3. The method of claim 1, wherein the imaging brain activity of said subject comprises MRI, optionally fMRI.

4. The method of claim 1, wherein the human brain network is associated with a neurologic disorder.

5. The method of claim 1, wherein the human brain network is associated with a psychiatric disorder.

6. The method of claim 1, wherein the neurologic disorder is epilepsy.

7. The method of claim 1, wherein the neurologic disorder is medial temporal lobe epilepsy.

8. The method of claim 1, wherein the neurologic disorder is Alzheimer’s.

9. The method of claim 1, wherein the neurologic disorder is Tourette’s syndrome.

10. The method of claim 1, wherein the psychiatric disorder is an addiction disorder.

11. The method of claim 1, wherein the psychiatric disorder is methamphetamine addiction.

12. The method of claim 1, wherein the psychiatric disorder is alcohol use disorder.

13. The method of claim 1, wherein the psychiatric disorder is obsessive compulsive disorder.

14. The method of claim 1, wherein the psychiatric disorder is obesity.

15. The method of claim 1, wherein the virtual representations of brain networks are low dimensional.1314929-9580-4034v.

116. The method of claim 1, wherein the virtual representations of brain networks are of a dimensionality less than 20, less than 10, less than 7, or less than 5.

17. The method of claim 1, wherein virtual representations of brain networks do not contain sub-networks that are one or more of parcellation, noise, and / or canonical networks.

18. The method of claim 1, wherein the virtual representations of brain networks are comprised of brain sub-networks that are not parcellated networks, noise networks, and canonical networks.

19. The method of claim 1, wherein the virtual representations of brain networks are generated without separately identifying nodes and edges comprising the virtual representations of brain networks.

20. The method of claim 1, wherein the virtual representations of brain networks are derived from cross-sectional data, and are congruent with longitudinal network dynamics.

21. The method of claim 1, wherein the Low-d M-IC A algorithm is incrementally applied for different dimensions.

22. The method of claim 1, wherein the Low-d M-1CA algorithm contains stopping conditions comprising one or more of: three consecutive dimensions produce a subnetwork parcellation, noise component, and / or a canonical network; and / or insufficient statistical power (d > n / 23), where d is dimension and n is the number of experiments in the meta-analysis.

23. The method of claim 1, wherein the information associated with the neurologic disorder or the psychiatric disorder is modeled as a plurality of Gaussian probability distributions with full-width half-maximum values that are irrespective of the amount of information.

24. A method for generating a virtual representation of a human brain network associated with a selected neurologic disorder or psychiatric disorder, the human brain network comprising a morphologic property and a physiologic property and, optionally, a temporal property, the method comprising:1324929-9580-4034v.1a) selecting, by one or more computers of a high-performance computing portal, a neurologic disorder or psychiatric disorder;b) accessing, by the one or more computers, a plurality of databases storing group- averaged coordinate-based spatially-normalized data populated from a plurality' of human brain subject, each stored on one or more computer-readable media, comprising:i) a task-activation database (TA DB) comprising coordinate-based quantified brain activity data related to specific human subject tasks; ii) a voxel-based morphometry database (VBM DB) comprising voxelbased morphometric data; andiii) a voxel-based physiology database (VBP DB) comprising voxel-based physiological data;c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected neurologic disorder or psychiatric disorder;d) applying, by the one or more computers, a multivariate coordinate-based metaanalysis (CBMA) algorithm, comprising a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network comprising a morphologic property and a physiologic property and, optionally, a temporal property associated with the selected neurologic disorder or psychiatric disorder, and generating a virtual representation of the human brain network.

25. The method of claim 23, wherein the virtual representation of the human brain network is used as priors in machine learning, artificial intelligence, or statistical models.

26. The method of claim 24. wherein the human brain network is associated with a neurologic disorder.

27. The method of claim 24, wherein the human brain network is associated with a psychiatric disorder.

28. The method of claim 24, wherein the presenting step (E) comprises transmitting the virtual representation of said human brain network to a user device of the user.1334929-9580-4034v.

129. The method of claim 24, wherein the presenting step (E) comprises causing the virtual representation of said human brain network to be displayed on a user device of the user.

30. A method for generating a virtual representation of a human brain network associated with a selected developmental disorder, systemic disorder, or age-associated change, the human brain network comprising a morphologic property and a physiologic property and, optionally, a temporal property, the method comprising:a) obtaining, by one or more computers of a high-performance computing portal, the selected developmental disorder or systemic disorder;b) accessing, by the one or more computers, a plurality of databases storing group- averaged coordinate-based spatially-normalized data populated from a plurality of human brain subject, each stored on one or more computer-readable media, comprising:i) a task-activation database (TA DB) comprising coordinate-based quantified brain activity data related to specific human subject tasks; ii) a voxel-based morphometry database (VBM DB) comprising voxelbased morphometric data; andiii) a voxel-based physiology database (VBP DB) comprising voxel-based physiological data;c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected developmental disorder, systemic disorder, or age-associated changes;d) applying, by the one or more computers, a multivariate coordinate-based metaanalysis (CBMA) algorithm, comprising a Low-dimensionality’ Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network comprising a morphologic property and a physiologic property and, optionally, a temporal property associated with the selected developmental disorder, systemic disorder, or age- associated changes, and generating a virtual representation of the human brain network; ande) presenting, by the one or more computers, to a user the virtual representation of said human brain network.1344929-9580-4034v.

131. A system comprising:one or more computers of a high-performance computing portal comprising memory wherein the memory stores computer readable instructions comprising program code that, when executed, cause the one or more computers to perform the steps of:a) obtaining, by the one or more computers of the high-performance computing portal, a selected neurologic disorder or psychiatric disorder;b) accessing, by the one or more computers, a plurality of databases storing group- averaged coordinate-based spatially-normalized data populated from a plurality of human brain subjects, each stored on one or more computer- readable media, comprising:i) a task-activation database (TA DB) comprising coordinate-based quantified brain activity data related to specific human subject tasks; ii) a voxel-based morphometry database (VBM DB) comprising voxelbased morphometric data; andiii) a voxel-based physiology database (VBP DB) comprising voxel-based physiological data;c) obtaining, by the one or more computers, morphologic, physiologic, and optionally temporal, information from the plurality of databases, which information is associated with the selected neurologic disorder or psychiatric disorder;d) applying, by the one or more computers, a multivariate coordinate-based metaanalysis (CBMA) algorithm, comprising a Low-dimensionality Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm, to the obtained information so as to identify a human brain network comprising a morphologic property and a physiologic property and, optionally, a temporal property associated with the selected neurologic disorder or psychiatric disorder, and generating a virtual representation of the human brain network; and e) presenting, by the one or more computers, to a user the virtual representation of said human brain network.

32. The system of claim 30, wherein the human brain network is associated with a neurologic disorder.1354929-9580-4034v.

133. The system of claim 30, wherein the human brain network is associated with a psychiatric disorder.

34. The system of claim 30, wherein the presenting step (E) comprises transmitting the virtual representation of said human brain network to a user device of the user.

35. A method for delivery of transcranial magnetic stimulation (TMS) to a subject with a neurologic or psychiatric disorder, comprising:a) obtaining a virtual representation of a brain network associated with the neurologic disorder or a psychiatric disorder, wherein the virtual representation was generated by applying a multivariate coordinate-based meta-analysis (CBMA) algorithm to information associated with the neurologic disorder or the psychiatric disorder, the information comprising:i) coordinate-based quantified brain activity data related to specific human subject tasks stored in a task-activation database (TA DB);ii) voxel-based morphometric data stored in a voxel-based morphometry database (VBM DB); andiii) voxel-based physiological data stored in a voxel-based physiology database (VBP DB); andwherein the CBMA algorithm comprises a Low-dimensionality’ Meta-analytic Independent Component Analysis (Low-d M-ICA) algorithm;b) identifying, based at least in part on a connectivity -based parcellation derived from the virtual representation of a brain network, a first upstream target brain region;c) computing, in at least one computing system, a region-seeded connectivity' map of the first upstream target brain region of the subject based at least in part on the virtual representation of a brain network;d) identifying at least one downstream projection region of the subject associated with the first upstream target brain region of the subject;e) computing, in the at least one computing system, a region-seeded connectivity map of the at least one downstream projection region of the subject;f) identifying a first upstream projection region of the subject that demonstrates functional connectivity to the at least one downstream projection region of the subject;1364929-9580-4034v.1g) determining positioning for a TMS coil with respect to the first upstream projection region of the subject in accordance with a cortical column cosine aiming principle; andh) causing delivery of the TMS to the subject via the TMS coil according to the determined positioning.1374929-9580-4034v.1

Images