Mapping of brain networks for detection or characterization of psychological conditions

Abstract

The disclosed approach uses imaging data from a scan of a brain of a subject to generate a metric indicative of a characteristic of one or more functional networks in the brain. The metric, or data based on the metric, may be provided to a machine learning classifier to generate a prediction related to a psychological condition such as depression.

Description

MAPPING OF BRAIN NETWORKS FOR DETECTION OR CHARACTERIZATION OF PSYCHOLOGICAL CONDITIONSCROSS-REFERENCE TO RELATED PATENT APPLICATIONS

[0001] This application claims the benefit of and priority to U.S. Provisional PatentApplication No. 63 / 517,836 filed August 4, 2023, and titled “Systems And Methods For Diagnosing Depression, Assessing Depression Risk, And Predicting Antidepressant Responses Using FMRI Network Mapping,” which is herein incorporated by reference in its entirety.STATEMENT REGARDING FEDERAL FUNDING

[0002] This invention was made with government support under F32MH120989 awarded by the National Institutes of Health. The government has certain rights in the invention.FIELD(0003[ This disclosure relates to generating and using one or more physical metrics (used interchangeably with “biomarker”) of a subject to detect, characterize, and / or make predictions regarding psychological conditions in the subject. For example, one or more biomarkers based on functional mapping of a subject’s brain can be used to detect, characterize, and / or make predictions related to the presence, progression, symptoms, and / or responses to therapies of depression and / or other psychological conditions in the subject.BACKGROUND

[0004] Biomarkers for detecting or characterizing psychiatric conditions, tracking and quantifying current symptoms, predicting future changes in specific symptom domains, and predicting responses to treatments are lacking and urgently needed.SUMMARY OF EXAMPLE EMBODIMENTS

[0005] Example embodiments of the disclosed approach relate to systems and methods for mapping functional networks in a brain of a subject, such as the human brain, to detect abnormal structural changes or features (e.g., expansion or contraction) in one or more brain networks. In various embodiments, the one or more brain networks is, or includes, oneor more salience networks, such as the frontostriatal salience network, and / or other networks. In various embodiments, changes in brain networks can be used to identify psychiatric conditions like depression in individuals, and / or other psychological conditions. In various embodiments, a machine learning classifier may be trained to make a prediction related to the psychological condition based on one or more features of a salience network and / or one or more other brain networks. In certain embodiments, one or more psychological conditions can be identified using significantly fewer metrics than prior approaches. In certain embodiments, depression can be identified based on a single metric corresponding to a functional brain map. Prior approaches did not use a biological measure to diagnose or make predictions regarding depression or other psychiatric conditions in individuals.

[0006] Example embodiments of the disclosed approach enable non-invasive identification of low-dimensional features of functional brain organization (e.g., how large or small a functional network is) that are associated with, for example, depression or other psychological conditions. In various embodiments, these features are stable (trait-like) and not sensitive to an individual’s current mood state, which is advantageous from a biomarker standpoint and personalized neuromodulation perspective. Whereas existing fMRI tools are highly sensitive to scanner-related artifacts, example embodiments of the disclosed biomarkers are not sensitive to scanner-related artifacts and can be used for clinical diagnostic and prognostic purposes without adjustment for scanner effects.

[0007] Example embodiments of the disclosed approach relate to features of functional brain organization involving metrics that are dependent on a mood state and could be used, for example, for tracking changes in symptoms and / or predicting symptoms before they occur (see, e.g., FIG. 6). For example: changes in functional connectivity between anterior cingulate and nucleus accumbens nodes of the salience network tracking current anhedonic symptoms (see, e.g., FIG. 6C); changes in functional connectivity between anterior cingulate and nucleus accumbens nodes of the salience network predicting future anhedonic symptoms (see, e.g., FIG. 6E); and / or changes in functional connectivity between anterior insula and nucleus accumbens nodes of the salience network tracking anxiety symptoms (see, e.g., FIG. 61). One clinical impact is that these could be used as outcome measures in industry trials of treatments for anhedonia or anxiety, could be used for guiding neuromodulation treatments to target these circuits, and / or could be used to predict future symptoms to intervene earlier in treating patients. Example embodiments of the disclosed approach further can use changes in networks (e.g., stable salience network expansion) topredict risk for becoming depressed in people who have not previously had depression (see, e g., FIG. 5G).

[0008] Various embodiments relate to a method. The method comprises: acquiring imaging data from a scan of a brain of a subject, the scan lasting a minimum time; generating, based on the imaging data, a functional connectivity model with correlations among, or between, cortical vertices and subcortical voxels in the imaging data; generating, based on the functional connectivity model, a metric indicative of a relative occupancy of one or more functional networks in the brain; providing the metric to a machine learning classifier to generate an output from the classifier, the output providing a prediction related to a psychological condition; and providing the prediction, or information generated from the prediction, to one or more users, wherein providing the prediction or the information comprises at least one of transmitting the prediction or the information to a user device, displaying the prediction or the information on a display device, and / or storing the prediction or the information in a non-transitory computer-readable memory of a networked computing system that is accessible, through the network, to one or more user devices of the one or more users.

[0009] In example embodiments, the scan is a functional magnetic resonance imaging (fMRI) scan of the brain of the subject. In example embodiments, the minimum time is 10 minutes. In example embodiments, the minimum time is 15 minutes. In example embodiments, the scan is a muti-echo resting-state fMRI (ME-rsfMRI) scan of the brain of the subject. In example embodiments, the scan is a single-echo resting-state fMRI scan of the brain of the subject. In example embodiments, the single echo resting-state fMRI scan of the brain has a duration of at least 45 minutes, such as between 60 minutes and 120 minutes or between 90 minutes and 150 minutes, or greater than 120 minutes. In example embodiments, the correlations of the functional connectivity model are or comprise correlations between a blood-oxygen-level-dependent (BOLD) signal time series of the cortical vertices and / or the subcortical voxels. In example embodiments, the correlations of the functional connectivity model are or comprise correlations between cortical vertices, between subcortical voxels, and / or between cortical vertices and subcortical voxels. In example embodiments, the correlations of the functional connectivity model are or comprise correlations between pairs of cortical vertices, or between pairs of subcortical voxels, or between one cortical vertex and one subcortical voxel. In example embodiments, the machine learning classifier is a support vector machine (SVM) classifier. In example embodiments, the machine learning classifier istrained using (a) diagnosis status as class labels and / or (b) functional brain network size as features. In example embodiments, generating the functional connectivity model comprises setting correlations between nodes that are less than a threshold distance apart to zero. In example embodiments, generating the functional connectivity model comprises setting correlations between nodes that are greater than the threshold distance apart to zero. In example embodiments, the method comprises identifying one or more functional networks in the imaging data using a community detection algorithm. In example embodiments, generating the metric comprises determining a surface area for each vertex in a region of the brain. In example embodiments, generating the metric comprises determining a relative contribution of each functional network to a total cortical surface area. In example embodiments, the metric is or comprises an occupancy measure for a network, such as a salience network. In example embodiments, the metric is indicative of an expansion or contraction of one or more networks. In example embodiments, the metric is indicative of an expansion of a salience network of the subject, such as a frontostriatal salience network of the subject. In example embodiments, the metric is indicative of a contraction of one or more networks (see, e.g., FIG. 3). In example embodiments, the metric is indicative of a contraction of a plurality of networks, such a plurality of adjacent or otherwise neighboring networks, such as three adjacent networks. In example embodiments, the metric is indicative of a contraction of one or more networks comprising at least one of a default mode network, a frontoparietal control network, and / or a cingulo-opercular control network. In example embodiments, the metric is indicative of a contraction of a default mode network, a frontoparietal control network, and a cingulo-opercular control network. In example embodiments, the method comprises using the prediction or the information to (a) determine a treatment for the psychological condition, (b) predict a response to the treatment, (c) quantify symptoms, and / or (d) predict changes in symptoms. In example embodiments, the metric is a biomarker that is not sensitive to scanner-related artifacts. In example embodiments, the method does not comprise adjusting for scanner effects. In example embodiments, the metric is stable such that the metric does not necessitate a mood state.

[0010] In example embodiments, the metric is dependent on a mood state. In example embodiments, the metric is used for tracking changes in symptoms and / or for predicting symptoms before they occur. In example embodiments, current anhedonic symptoms are tracked based on changes in functional connectivity between or among nodes, such as between anterior cingulate and nucleus accumbens nodes of the salience network. Inexample embodiments, future anhedonic symptoms are predicted based on changes in functional connectivity between or among nodes, such as between anterior cingulate and nucleus accumbens nodes of the salience network. In example embodiments, anxiety symptoms are tracked based at least in part on changes in functional connectivity between or among nodes, such as between anterior insula and nucleus accumbens nodes of the salience network tracking. In example embodiments, metrics are used as outcome measures in treatments for psychological conditions such as anhedonia or anxiety. In example embodiments, metrics are used for guiding neuromodulation treatments to target one or more circuits in the brain. In example embodiments, metrics are used to predict future symptoms, so as to, for example, intervene earlier in treating patients. In example embodiments, changes in networks (e.g., stable salience network expansion) are used to predict risk for becoming depressed in people who have not previously had depression (see, e.g., FIG. 5G).

[0011] Various embodiments relate to a computing system. The computing system comprises one or more processors and is configured to perform any of the methods disclosed herein. In example embodiments the computing system is configured to communicate with a medical imaging system, such as an fMRI scanner, to acquire the imaging data. In example embodiments, the computing system is accessible to one or more other computing systems and / or to one or more user computing devices for exchange of information related to subjects, such as imaging data, images, results of analysis, etc.(0012[ Various embodiments relate to an imaging system comprising one or more detectors. The imaging system may comprise one or more processors configured to perform any of the methods disclosed herein. The imaging system may, alternatively or additionally, communicate with one or more computing systems and / or with one or more user devices. The one or more computing systems and / or one or more devices may perform one or more steps, or all steps, of any of the methods disclosed herein. The user devices may send commands for control of functionality (such as scanning of subjects and performance of analyses) and / or receive results of imaging and / or analyses.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Fig. 1 is an example system comprising components capable of implementing various illustrative embodiments of the disclosure.]0014[ Figs. 2A-2B depict example processes for implementing various embodiments of the disclosed approach, according to various illustrative embodiments of the disclosure.

[0015] Figs. 3A-3H illustrate frontostriatal salience network expanded nearly twofold in the cortex of highly-sampled individuals with depression, according to various illustrative embodiments of the disclosure. Fig. 3A: The salience network (SAL, black) has representation in lateral prefrontal (LPFC), anterior cingulate (ACC), and anterior insular cortex (Al). Figs. 3B, 3C: The salience network was 73% larger on average in n = 6 highly- sampled individuals with depression (a dataset referred to here as Serial Imaging of Major Depression, SIMD, significance assessed using a permutation test, *P = 0.001, Bonferroni correction, Z-score = 6.19). This effect was replicated thrice (two-tailed independent sample t-tests, n = 48 from Weill Cornell Medicine, MDD-1 : T = 3.54, *P = 0.01, Bonferroni correction, Cohen’s d = 0.72; another sample of n = 45 from Weill Cornell Medicine, MDD- 2: T = 4.17, *P = 0.002, Bonferroni correction, = 0.84; n = 42 from Stanford University, MDD-3: T = 3.68, *P = 0.008, Bonferroni correction, Cohen’s d= 0.77). Data are presented as mean values + / - SD. Fig. 3D: No significant group differences in salience network representation in the striatum were observed in either the discovery (two-tailed permutation test, P = 0.07, uncorrected) or replication datasets (two-tailed independent sample Ltests, all ’s > 0.43, uncorrected). Data are presented as mean values + / - SD. Fig. 3E: Density maps confirm spatial locations of salience network nodes were similar in healthy controls and individuals with depression, but that network borders extended further outwards from their centroids in each cortical zone in depression (red boxes). Fig. 3F: A support vector machine (SVM) classifier distinguished individuals with depression from healthy controls above chance (accuracy = 78.4%, significance assessed using a permutation test, P = 0.001) using the size of each functional network as features. Fig. 3G, 3H: Linear predictor coefficients (P) associated with the trained model and the change in accuracy after network exclusion both indicated that salience network size was the most important feature.

[0016] Figs. 4A-4G illustrate three modes of salience network expansion in depression, according to various illustrative embodiments of the disclosure. Fig. 4A: Mode functional brain network assignments in cortex and striatum in healthy controls (HC). Fig. 4B, 4C: The parts of each depressed individual’s salience network that did and did not overlap with the healthy controls are referred to as non-encroaching and encroaching, respectively. Fig. 4D: Salience network expansion more often due to shifts in network borders than ectopic intrusions - isolated patches of salience network in atypical locations (two-tailed paired sample Ltest, *P < 0.001, n = 141). Data are presented as mean values + / - SD. Fig. 4E: The parietal subnetwork of the default mode (DMN, red), frontoparietal (FP,yellow), and cingulo-opercular (CO, purple) networks were most frequently displaced by salience network expansion. Fig. 4F: Salience network expansion affected different functional networks in different cortical zones. In the anterior insular cortex (Al), the frontoparietal network (T = 5.94, *P < 0.001) and cingulo-opercular (T = 6.42, *P < 0.001) networks were more affected than the default mode network. In the anterior cingulate cortex (ACC), the default mode network was more affected than either the frontoparietal (T = 17.53, *P < 0.001) or cingulo-opercular / action-mode (T = 15.25, *P < 0.001) networks. Finally, in lateral prefrontal cortex (LPFC), the frontoparietal network was more affected than either the default mode (T = 9.31, *P < 0.001) or cingulo-opercular (T = 6.33, *P < 0.001). Statistical significance was assessed using two-tailed two-sample t-tests, all p-values are Bonferroni corrected, n = 141. Data are presented as mean values + / - SD. Fig. 4G: Individuals with depression clustered using their encroachment profiles (the relative contribution of each functional network to the total surface area of the encroaching portion of their salience network) revealed three distinct modes of encroachment across individuals.

[0017] Figs. 5A-5G illustrate salience network expansion is stable over time and present prior to symptom onset, according to various illustrative embodiments of the disclosure. Fig. 5A: Cortical representation of the salience network was stable in repeatedly scanned healthy controls (left) and individuals with depression (right). The first ten study visits for each individual are shown for visualization purposes. Fig. 5B: Salience network in a representative individual with depression that was scanned longitudinally to sample different mood states. Figs. 5C, 5D: No significant correlation between the severity of depressive symptoms (Hamilton Depression Rating Scale, HDRS6) and salience network size in any repeatedly-sampled individual with depression (Pearson correlation, all p’s > 0.63, two-tail test), or before and after a course of either a traditional 6-week (two-tailed paired sample t- test, T = 0.58, P = 0.55, uncorrected, n = 90) or accelerated 1-week (two-tailed paired sample / -test, T = 0.58, P = 0.56, uncorrected, n = 45) repetitive transcranial magnetic stimulation therapy (rTMS). Data are plotted as mean values + / - SD. Figs. 5E, 5F: Individual differences in salience network size were not significantly correlated with depression severity (HDRS6, Pearson correlation, r = 0.04, P = 0.63, uncorrected, two-tailed test) or the number of depressive episodes experienced (inferred from the Mini-International Neuropsychiatric Interview). Tx = treatment. Data are plotted as mean values + / - SD. Fig. 5G: Children from the Adolescent Brain Cognitive Development study scanned prior to the onset of elevated depression symptoms were identified (ABCD-MDD). Depression symptoms wereoperationalized using the DSM-oriented scale for depression from the Child Behavior Checklist (CBCL, T-scores > 70 are in the clinical range). The salience network was significantly larger in children who later developed clinically elevated symptoms of depression compared to children did not (two-tailed independent sample t-test, T = 3.50, *P < 0.001, Cohen’s d = 0.62, n =114). Sx = symptom. Data are plotted as mean values + / - SD.

[0018] Figs. 6A-6H illustrate frontostriatal salience network connectivity predicts fluctuations in anhedonia and anxiety symptoms within deeply-sampled individuals with depression over time, according to various illustrative embodiments of the disclosure. Fig. 6A: Heat map summarizing fluctuations in individual items selected from a variety of clinical interviews and self-report scales related to anhedonia in a deeply-sampled individual with depression (MDD04). Clinical data was resampled to days for visualization purposes (black dots mark the study visits). Fig. 6B: Frontostriatal nodes of the salience network in MDD04. Figs. 6C, 6D: Salience network functional connectivity (FC), especially between nodes in the nucleus accumbens (NAc) and anterior cingulate cortex (ACC), tracked fluctuations in the severity of anhedonia related symptoms within both MDD04 (Pearson correlation, r = -0.37, P = 0.003) and MDD06 (Pearson correlation, r = -0.49, P = 0.001) across study visits. Statistical significance was assessed using two-tailed permutation tests with circular rotation to preserve temporal autocorrelation. Fig. 6E: Cross-correlation analyses indicated NAc < -ACC FC also predicted the severity of anhedonia related symptoms at the following study visit in MDD04 (Pearson correlation, significance tested via permutation test, r = -0.32, *P = 0.004), but not in MDD06. Fig. 6F: No significant correlation between individual differences in salience network NAc < - ACC FC strength and the severity of anhedonia related symptoms across individuals (assessed using the Snaith-Hamilton Pleasure Scale, SHAPS, Pearson correlation, r = 0.09, P = 0.41). Figs. 6G, 6H: Within MDD04 and MDD06, salience network NAc < - ACC FC was not significantly related to fluctuations in the severity of other depressive symptoms, such as anxiety, whereas FC between other pairs of nodes (nucleus accumbens and anterior insula) was (Pearson correlations, MDD04: r = -0.29, P = 0.02; MDD06: r = -0.45, P = 0.004, two-tailed tests), indicating different patterns of functional connectivity relate to different symptoms.

[0019] Fig. 7 provides a comparison to an alternative network parcellation algorithm, according to various illustrative embodiments of the disclosure. The multi-session hierarchical Bayesian modeling (MS-HBM) approach uses group-level spatial priors from training data to estimate individual-specific network parcellations. Functional network mapsfor representative individuals generated using our precision functional mapping approach (Infomap) and MS-HBM are shown. Overall, there was good correspondence between the two sets of maps. Similarity was quantified using Normalized Mutual Information (NMI), a measure of the similarity between two sets of functional brain networks. The average NMI was 0.63 ± 0.03 (statistical significance confirmed using a permutation test, P < 0.001, n = 43). Group difference (MDD > HC) in salience network size remained statistically significant when using MS-HBM (two-tailed independent sample t-test, T= 3.99, P = 0.005, Bonferroni correction, Cohen’s d = 0.74, n = 43). ± = SD.[0020| Figs. 8A-8F illustrate the effect of global signal regression, according to various illustrative embodiments of the disclosure. Global signal regression (GSR) is a debated processing step. Mean gray matter time-series regression (MGTR) is effectively equivalent to GSR, and used in the present study. Fig. 8A: Group differences in salience network size remain statistically significant when MGTR is not performed (independent sample t-test, T= 5.06, P < 0.001, Bonferroni correction, Cohen’s d = 1.71). Fig. 8B: Individual differences in salience network size with (x-axis) and without (y-axis) across individuals are highly correlated (Pearson correlation, r = 0.92). Fig. 8C: All functional networks mapped a single representative individual with and without MTGR applied. Collectively, these results indicate that MGTR has little effect on functional brain network topography. Fig. 8D, 8E: Functional connectivity matrices for a representative healthy control and individual with depression with (Fig. 8D) and without (Fig. 8E) MGTR. Fig. 8F: Illustration of how the FC matrix is thresholded prior to Infomap. Specifically, a relative (as opposed to absolute) threshold is applied, such that only the top FC values (those at or above the specified percentile) for each node are retained. MGTR shifts weaker FC values (purple circle) more than stronger FC values (green circle), and the former is not retained after thresholding FC matrices before Infomap.[002.1] Figs. 9A-9B illustrate salience network expansion in depression is not explained by differences in brain anatomy or structure, according to various illustrative embodiments of the disclosure. Figs. 9A, 9B: Group differences in surface area assessed at each cortical vertex using independent sample t-tests (Fig. 9A, unthresholded on the left and Fig. 9B, thresholded at P = 0.05, uncorrected). Group differences in surface area were spatially diffuse (Spin permutation tests indicated that they do not localize to any network after correction for multiple comparisons, other than the Auditory network), and are predominantly in the opposite direction (HC > MDD) of what would drive salience networkexpansion in depression. There was no significant difference in the total cortical surface area between the healthy controls and depression samples (independent sample t-test, T= 1.57, P = 0.11), and no significant correlation between salience network size (or the size of any other functional network, after correction for multiple comparisons) and absolute total surface area. All claims regarding equivalence are based on an absence of evidence.

[0022] Figs. 10A-10D provide additional demographic and clinical information, according to various illustrative embodiments of the disclosure. Figs. 10A, 10B: The depression sample was predominantly (56.7%) female and primarily young- or middle-aged adults (40.71 ± 13.82 years old). Fig. 10C: Most (69%) of individuals with Major Depression in our study sample had a comorbid mood or anxiety disorder diagnosis (Generalized Anxiety Disorder: 48%, Social Anxiety Disorder: 22%, Post-Traumatic Stress Disorder: 9%, Obsessive Compulsive Disorder: 7%). Fig. 10D: With respect to medication status, most individuals were on an antidepressant (SSRI; 70%), about a third were on Benzodiazepine (BZD; 30%) or an antiepileptic drug (AED; 37%), and a relatively small number were taking an antipsychotic (AP; 13%) or Lithium as a mood stabilizer (4%).

[0023] Figs. 11A-11B illustrate salience network expansion replicated across study samples when correcting for site effects, according to various illustrative embodiments of the disclosure. The data in the present study is drawn from multiple different sites, which use different imaging sequences and MRI machines. To evaluate if effects of site or scanner impact contribute to the observed salience network expansion in individuals with depression, we applied ComBat harmonization to functional network topography (size) and repeated our analysis. The original (unadjusted) values were used for two of the healthy controls (MyConnectome, Cast-induced plasticity) because they were the only individuals from their respective sites. Fig. 11 A: The original results reproduced from Fig. 3C (with no ComBat harmonization). Data are plotted as mean values + / - SD. Fig. 11B: Discovery samples with ComBat harmonization (Serial Imaging of Major Depression, MDD-SIMD, permutation test, *P = 0.001, Bonferroni correction, Z-score = 6.04). Replication samples with ComBat harmonization (two-tailed independent sample Ltests, n = 48 from Weill Cornell Medicine, MDD-WCM1 : T= 3.14, *P = 0.045, Bonferroni correction, Cohen’s d= 0.65; a second sample of n = 45 from Weill Cornell Medicine, MDD-WCM2: T= 3.72, *P = 0.007, Bonferroni correction, Cohen’s d= 0.77; n = 42 from Stanford University, MDD-SU: T = 3.46, **P = 0.02, Bonferroni correction, Cohen’s d= 0.73). Data are plotted as mean values + / - SD.

[0024] Fig. 12 provides a visual representation of a cross-validation routine, according to various illustrative embodiments of the disclosure. Repeated (lOOx) nested split-half (2 -fold) cross-validation with a grid search optimization strategy for tuning box constraint and kernel size hyperparameters (range of possible values = 0.1, 1, 10, 100).

[0025] Figs. 13A-13C illustrate encroaching parts of the salience network are not localized to brain regions susceptible to noise, according to various illustrative embodiments of the disclosure. Fig. 13A: A heat map quantifying where Salience network expansion was most frequent in individuals with depression. Figs. 13B, 13C: Temporal signal -to-noise (tSNR) and test-retest reliability of whole-brain functional connectivity (FC) at each cortical vertex (as operationalized in). Good tSNR and FC test-retest reliability was achieved throughout cortex, including regions where salience network expansion occurred most frequently in individuals with depression, such as ventral medial prefrontal cortex (MPFC). Spin permutation tests indicated that the tSNR and FC reliability values associated with encroaching Salience network vertices were not significantly different from chance (both ’s > 0.46).

[0026] Fig. 14 illustrates no change in salience network size over a two year period between study time points in ACBD sample, according to various illustrative embodiments of the disclosure. Salience network size in the two ABCD samples from our study (n = 57 no symptom onset / ABCD-HC, n = 57 future symptom onset / ABCD-MDD) at baseline (Tl, age 10) and at the two-year-follow-up study visit (T2, age 12). Shaded error bands represent + / - SD. Significant differences in salience network size remained between the two groups at both Tl (two-tailed independent sample t-test, T = 3.38, P < 0.001, Cohen’s d= 0.60) and T2 (two-tailed independent sample t-test, T = 3.39, P < 0.001, Cohen’s d= 0.60). However, there was no change in salience network size over the two years between Tl and T2 in either the ABCD-HC (two-tailed paired sample t-test, T = 0.55, P = 0.58, Cohen’s d = 0.03) or ABCD-MDD samples (two-tailed, paired sample t-test, T = 1.03, P = 0.31, Cohen’s d= 0.05). All data are plotted as mean values + / - SD.

[0027] Fig. 15 illustrates constructing composite measures of anhedonia and anxiety from a large battery of clinical scales, according to various illustrative embodiments of the disclosure. The composite measures of anhedonia and anxiety we derived by instructing three separate clinicians to quantify (on a scale of 0-3; 0 = not at all, 1 = somewhat, 2 = largely, 3 = very strongly) the extent to each item taps into anhedonia or anxiety. Itemsassigned a score of 1 or greater by all three clinicians were included in the composite measure.

[0028] Figs. 16A-16B illustrate salience network connectivity strength with spatially distributed cortical nodes of multiple functional networks relates to severity of anhedonia and anxiety symptoms within highly-sampled individuals with depression, according to various illustrative embodiments of the disclosure. Columns represent different nodes of the salience network, and the rows are different cortical vertices (ordered by functional network, and subsampled for computational feasibility) in each individual (Fig. 16A, MDD04 and Fig. 16B, MDD06). Correlations not exceeding chance (based on null distribution of correlation coefficients obtained using rotated clinical data) were set to zero.

[0029] Fig. 17 illustrates salience network expansion is also found in two highly- sampled individuals with Bipolar II disorder, according to various illustrative embodiments of the disclosure. To begin to understand whether salience network expansion is present in individuals with primary clinical diagnoses other than major depression, we obtained data from individuals with bipolar disorder (n=2, 5.30 and 28.83 hours of multi-echo fMRI data per-subject), autism spectrum disorder (ASD, without comorbid MDD; n=6, 2.94 ± 0.89 hours of multi-echo fMRI data per-subject), and obsessive compulsive disorder (OCD, without comorbid MDD; n=9, 1 hour of multi-echo fMRI data per-subject) and performed precision functional mapping. This preliminary analysis indicated that the salience network is larger than normal in the two individuals with Bipolar II disorder (statistical significance confirmed using a two-tailed permutation test, P < 0.002), but not in the ASD (P = 0.09) or OCD (P = 0.84) samples. Dx = diagnosis. MDD = Major depressive disorder, BP = bipolar II disorder, ASD = Autism spectrum disorder, OCD = obsessive compulsive disorder. Error bars represent standard deviation.

[0030] Figs. 18A-18D illustrate both data quantity and quality constrain ability to reliably detect salience network expansion in individuals with depression, according to various illustrative embodiments of the disclosure. Fig. 18A: To begin to understand how much data is needed to reliably observe large group differences in salience network size, we iteratively performed precision functional mapping and calculated salience network size in the most highly-sampled individuals from our study (n = 37 healthy controls and n = 6 individuals with depression) using increasing amounts (from 15 minutes to 240 minutes, in 15 minute steps) of their motion-censored and concatenated resting-state fMRI dataset. At each step, we compared the observed effect size (Cohen’s d) associated with the groupdifference in means relative to the “true” effect size ( [observed Cohen’s d I true Cohen’s d\ * 100). The full true effect was not consistently observed until approximately 120 minutes of data. Fig. 18B: Similarity (Dice similarity coefficient) of salience network topography relative to the salience network mapped using each individual’s entire dataset also reached asymptote at approximately 90 to 120 minutes. Collectively, these data indicate that approximately 1.5 to 2 hours of high-quality fMRI data needed per-subject for reliable mapping of salience network, and consistent detection of expansion in individuals with depression. Figs. 18C, 18D: Beyond data quantity, another important consideration for mapping the salience network is that many “traditional” single-echo fMRI datasets tend to have pronounced signal dropout in the ventral medial prefrontal cortex (MPFC), one of the brain regions where we found that the salience network expansion most often encroaches into from the neighboring anterior cingulate cortex (ACC) in depressed individuals in our study. Note the difference between the “traditional” and the multi-echo “precision style” fMRI data with respect to signal dropout (green arrow; due to susceptibility artifact) in the vMPFC (red box). This problem can be exacerbated if field maps are not available (which is often the case in traditional fMRI datasets) to correct for the accompanying geometric spatial distortions. Fine-grained estimates of ACC functional connectivity (a core node of salience network) in the example traditional fMRI dataset are contaminated as a consequence — note how the ACC FC map has little spatial structure, and the random “speckling” pattern that is present throughout the cortex (classic hallmark of an artifact), but this is not the case for the posterior cingulate cortex (PCC) FC map.

[0031] Fig. 19 depicts salience network nomenclature, according to various illustrative embodiments of the disclosure. The brain network that is herein referred to as the salience network is sometimes called other names (“Control C” in) or combined with the parietal memory network, whereas the brain network referred to as cingulo-opercular or action-mode network is sometimes called the “Salience / Ventral Attention” network.

[0032] Figs. 20A-20C illustrate algorithmic assignment of network identities to Infomap communities, according to various illustrative embodiments of the disclosure. Fig. 20A: Each Infomap community was algorithmically assigned to one of 20 possible functional network identities (Default-Parietal, Default- Anterolateral, Default-Dorsolateral, Default- Retrosplenial, Visual-Lateral, Visual -Stream, Visual-Vl, Visual-V5, Frontoparietal, Dorsal Attention, Premotor / Dorsal Attention II, Language, Salience, Cingulo-opercular / Actionmode, Parietal memory, Auditory, Somatomotor-Hand, Somatomotor-Face, Somatomotor-Foot, Auditory, or Somato-Cognitive- Action) according their functional connectivity and spatial locations relative to a specified set of priors. An example of how this procedure works is shown above. In this example (Community 22), there is a strong spatial correlation between this community’s FC and both the Default-Parietal (r = 0.78) and Default- Anterolateral (r = 0.68) subnetworks. The spatial location of the community is far more consistent with the Default-Parietal (0.53 versus 0.13). The winning network assignment is based on the product of FC Similarity and Spatial Probability. All algorithmic assignments are manually reviewed and adjusted in the case of an ambiguous assignment. For example, when correspondence with all the functional network templates is poor (sometimes is the case for communities located in brain regions where signal quality can be poor, such as inferior temporal cortex or deep within the operculum), or when the relative difference in assignment confidence between the winning and runner-up functional network is low. Figs. 20B, 20C: The end result is that Infomap communities assigned to the same functional network will have similar spatial locations and functional connectivity profiles relative to the specified set of priors.

[0033] Fig. 21 depicts functional connectivity priors used for algorithmic assignment of functional network identities to Infomap communities, according to various illustrative embodiments of the disclosure.{0034] Fig. 22 depicts spatial location probability priors used for algorithmic assignment of functional network identities to Infomap communities, according to various illustrative embodiments of the disclosure.

[0035] Fig. 23 depicts a block diagram of a representative server system and client computer system usable to implement various embodiments of the present disclosure.DETAILED DESCRIPTION

[0036] It is to be appreciated that certain aspects, modes, embodiments, variations and features of the present methods are described below in various levels of detail in order to provide a substantial understanding of the present technology. The present disclosure is not to be limited in terms of the particular embodiments described in this application, which are intended as single illustrations of individual aspects of the disclosure. All the various embodiments of the present disclosure will not be described herein. Many modifications and variations of the disclosure can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. Functionally equivalent methods and apparatuseswithin the scope of the disclosure, in addition to those enumerated herein, will be apparent to those skilled in the art from the foregoing descriptions. Such modifications and variations are intended to fall within the scope of the appended claims. The present disclosure is to be limited only by the terms of the appended claims, along with the full scope of equivalents to which such claims are entitled.

[0037] With respect to depression, discussed herein in connection with various example embodiments of the disclosure, decades of neuroimaging studies have revealed modest differences in brain structure and connectivity in depression, hindering mechanistic insights or identifying risk factors for disease onset. Furthermore, while depression is episodic, few longitudinal neuroimaging studies exist, limiting understanding of mechanisms that drive mood state transitions. The emerging field of precision functional mapping has used densely-sampled longitudinal neuroimaging data to reveal behaviorally meaningful differences in brain network topography and connectivity between and within healthy individuals, but this approach has not been applied in, for example, depression.[0038| Example embodiments of the disclosed approach use precision functional mapping to determine whether a brain network has sufficiently changed one or more characteristics or features such as size (e.g., whether the frontostriatal salience network has expanded more than a threshold amount in the cortex of a subject). In example embodiments, this effect is replicable in multiple samples, caused primarily by network border shifts, with three distinct modes of encroachment occurring in different individuals. Salience network expansion is stable over time and thus not significantly affected by mood state, and is detectable in children before the onset of depression later in adolescence. Longitudinal analyses of individuals scanned up to 62 times over 1.5 years identified connectivity changes in frontostriatal circuits that tracked fluctuations in specific symptoms and predicted future anhedonia symptoms. Together, these findings identify a trait-like brain network topology that can confer risk for depression and mood-state dependent connectivity changes in frontostriatal circuits that predict the emergence and remission of depressive symptoms over time.[0039[ Depression, for example, is a heterogeneous and episodic neuropsychiatric syndrome associated with synapse loss and connectivity alterations in frontostriatal networks, and a leading cause of disability worldwide. The neurobiological mechanisms that give rise to specific depressive symptom domains or to changes in mood over time are not well understood, especially at the neural systems level. To date, most functional magneticresonance imaging (fMRI) studies have tested for differences in functional connectivity in cross-sectional comparisons between groups of depressed individuals and healthy, never- depressed controls using group-average (“one-size-fits-all”) parcellations to define functional brain areas and networks. More recently, pioneering work in systems neuroscience has given rise to the field of precision functional mapping, which refers to a suite of new approaches for delineating functional networks entirely within individuals. Precision mapping studies have shown that the topology (size, shape, spatial location) of functional areas and networks in individuals deviates markedly from group-average descriptions, and that individual differences in network topology are stable, heritable, and associated with cognitive abilities and behavior. These tools have not yet been widely applied in clinical populations, including depression. Thus, whether functional network topology differs in individuals with depression has been unknown.

[0040] Depression is a fundamentally episodic neuropsychiatric condition defined by discrete periods of low mood interposed between periods of euthymia, but our understanding of the mechanisms that mediate mood transitions over time is limited. This is due in part to the fact that most studies to date have been cross-sectional, involving data acquired at a single time point, or in some cases, two or three scans acquired before and after an intervention — an approach that is not designed for meaningful statistical inferences at the individual level. Understanding the neurobiological mechanisms that mediate transitions in and out of depressive mood states may require dense-sampling of individual patients over many months. Indeed, densely sampled n-of-1 studies involving intracranial electroencephalogram (EEG) recordings and other assessments have begun to reveal mechanisms that regulate mood state transitions in individual patients receiving deep brain stimulation for depression, but these approaches have not yet been deployed at scale in fMRI studies. Absent such datasets, it is unknown whether changes in brain network connectivity predict the emergence of anhedonia, anxiety, and dysfunction in other depressive symptom domains, or the subsequent remission of these symptoms after a recovery from an episode. In the same way, it is unclear whether atypical network topology measures fluctuate with mood state in individuals with depression or remain stable over time — key questions for understanding cause-and-effect relationships in clinical neuroimaging, for defining potential therapeutic targets in neuromodulation interventions, or identifying at-risk individuals. Until recently, technical limitations have posed significant obstacles to performing precision functional mapping and longitudinal neuroimaging in clinical samples, including depression. Conventional fMRI measurements atthe single subject level are often noisy and have limited reliability, in part because they are sensitive to a variety of imaging artifacts. Acquiring large quantities of data in each subject and / or by using multi-echo fMRI, embodiments of the disclosed approach can generate highly reliable functional connectivity measures and network maps at the level of individual subjects, an important step towards developing and deploying fMRI for clinical translational purposes.

[0041] Various embodiments of the disclosed approach use precision functional mapping tools to delineate topology of functional brain networks in individuals with depression or other psychological conditions, leveraging multiple resting-state fMRI datasets of deeply-sampled individuals. As used herein, “psychiatric condition” is a subset of “psychological condition.” Psychological conditions include psychiatric conditions (e.g., various types of depression) as well as non-psychiatric psychological conditions such as sub- clinical anxiety that may not rise to the level of a clinical psychiatric issue. The frontostriatal salience network was found to be expanded by nearly two-fold in most individuals with depression — an effect replicated thrice using independent samples of repeatedly-sampled individuals with depression (total n=135) and in large-scale group-average data (n=299 individuals with depression, n=932 healthy controls), with three distinct types of encroachment displacing neighboring functional systems occurring across individuals. Salience network expansion was stable over time, and unaffected by changes in mood state. It was also present in children scanned before the onset of depression symptoms that emerged later in adolescence. Longitudinal analyses of densely-sampled individuals revealed mood state-dependent changes in striatal connectivity with anterior cingulate and anterior insular nodes of the salience network that tracked fluctuations in anhedonia and anxiety, respectively, and predicted the subsequent emergence of anhedonic symptoms at future study visits.

[0042] Referring to FIG. 1, in various embodiments, a system 100 may include a computing system 110 (which may be or may include one or more computing devices, colocated or remote to each other), a condition detection system 160 (comprising detectors and / or sensors capable of, e.g., performing scans to generate imaging data on a subject, such as a human patient), an information system 170 (such as a health information system / medical information system) that may record and / or provide health information, and one or more user computing devices 180 that allow users to access one or more of computing system 110, condition detection system 160, and / or information system 170). The computing system110 (e.g., one or more computing devices) may be used to control and / or exchange signals and / or other data with condition detection system 160, information system 170, and / or user computing devices 180, directly (e.g., through wireless and / or wired communication) or indirectly via another component of system 100 (e.g., via any combination of wireless and / or wired communication). The computing system 110 may include one or more processors and one or more volatile and / or non-volatile memories for storing computing code and data that are captured, acquired, recorded, and / or generated.

[0043] The computing system 110 may include a controller 112 that is configured to exchange control signals with condition detection system 160, information system 170, user computing devices 180, and / or any components thereof, allowing the computing system 110 to be used to control, for example, capture of images, acquisition of signals by sensors, positioning or repositioning of subjects or devices, recording or obtaining other information, etc.

[0044] A transceiver 114 allows the computing system 110 to exchange readings, control commands, and / or other data or signals, wirelessly or via wires, directly or indirectly via networking protocols, with, for example, condition detection system 160, information system 170, and / or user computing devices 180, or components thereof, as well as one or more user computing devices. One or more user interfaces 116 allow the computing device 110 to receive user inputs (e.g., via keyboard, touchscreen, microphone, camera, motion detection, biometric scan, etc.) and provide outputs (e.g., via display screens, audio speakers, light emitters, AR / VR / MR headsets etc.) with users. The computing device 110 may additionally include one or more databases 118 for storing, for example, data acquired from one or more systems or devices, signals acquired via one or more sensors, images, biomarker signatures, etc. In some implementations, database 118 (or portions thereof) may alternatively or additionally be part of another computing device that is co-located or remote (e.g., via “cloud computing”) and in communication with computing device 110, condition detection system 160, information system 170, and / or user computing devices 180 or components thereof.

[0045] Condition detection system 160 may include one or more imagers 162, which may be or may include, for example, any system or device that is involved in, for example, capturing images or otherwise obtaining imaging data prior to, during, following, or in consideration of a procedure. Imagers 162 may employ any suitable imaging modality, such as magnetic resonance imaging, fluoroscopy, ultrasound, or digital imagery from a camera.Imagers 162 may include detectors for visible light and / or light in any frequencies of interest, such as (but not limited to) the spectrum from infrared to ultraviolet. Imagers may employ any suitable optical components (e.g., lenses, mirrors, filters, beam splitters, prisms, diffusers, diffraction gratings, etc.), digital components (e.g., charge-coupled devices (CCDs), as well as an area for placement of subjects, computing components (e.g., one or more processors, such as digital signal processors) to process and / or pre-process images, etc. Such imagers may be incorporated into endoscopic devices. Imagers may be, or may employ, any devices, tools, and / or techniques that will provide the desired imaging data, such as an ultrasound transducer. The imager 162 may have the capability of receiving control signals from a computing system (e.g., computing system 110) and / or a computing device (e.g., user computing devices 180) to, for example, initiate or cease image capture, and / or return images or other imaging data or status signals to the computing system (e.g., computing system 110 and / or information system 170) and / or the user device (e.g., user computing devices 180). Sensors 164 may detect, for example, other aspects of the subject, instrument, and / or environment, such as position, motion, temperature, humidity, etc. In certain embodiments, sensors 164 may include devices for detecting electrical properties (e.g., devices for impedance tracking). Condition detection system 160 may include one or more visualization devices 166 (e.g., imaging data processing capabilities and displays or other output devices for visualizing or otherwise presenting the imaging or other data acquired using the imager 162 and / or sensors 164, or renderings or analyses thereof). In various embodiments, visualization devices 166 may include any combination of display devices, headsets, and / or other devices that enable visual presentation of information. The visualization devices 166 may, for example, receive visualizations generated by or via image Tenderer 128.

[0046] User computing devices 180 may include any devices used by practitioners, technicians, or other users to, for example, request scans, control components of condition detection system 160, receive data from computing system 110, condition detection system 160, and / or information system 170, or analyses of data, for use in planning for and / or implementing any medical procedures. User computing devices 180 may include any number of software applications 182 that can serve as an interface between the user and one or more components of system 100. User interfaces 184 include any input devices and / or output devices, such as ones discussed with respect to user interfaces 116, to allow the user to engage with the user computing device 180 and / or with the one or more software applications182. In various embodiments, user interfaces 184 may comprise one or more visualization devices, discussed with respect to visualization devices 166.100471 In various implementations, components of system 100 may be rearranged or integrated in other configurations. For example, computing system 110 (or components thereof) may be integrated with one or more of the condition detection system 160, user computing devices 180, and / or components thereof. It is also noted that not all components of system 100 are required to implement the disclosed approach, and in various embodiments, only a subset of the components of system 100 may be employed. For example, in various embodiments, computing system 110 may obtain and process data that was obtained via an another system that is, or is not, in direct communication with the computing system 110.

[0048] An image analyzer 124 may retrieve or otherwise receive images (e.g., from or via condition detection system 160) and analyze or otherwise process imaging data to extract relevant information. Network detector 126 may use raw or processed data (e.g., raw imaging data or imaging data obtained from or via image analyzer 124) to define, delineate, and / or characterize brain networks or other structures in the imaging data or in images.Image Tenderer 128 may render or otherwise generate images, elements of images, and / or other visualizations, which may be provided to, for example, information system 170 and / or user computing devices 180 for presentation or access to a user via, for example.

[0049] Machine learning platform 130 may be configured to train and update machine learning models, as further discussed herein. Machine learning platform 130 may, for example, employ certain machine learning techniques and algorithms to train and update predictive models or other classifiers. Machine learning platform 130 may include a training data generator 132 which may, for example, generate or otherwise obtain training data, such as images or imaging data and / or labels that identify a characteristic of components of the images (e.g., labels identifying brain networks or characteristics or features thereof). The modeler 134 may use training data to generate models that may be used for, for example, characterizing structures in the brain of the subject and / or for making predictions regarding one or more psychological conditions.

[0050] Referring to FIG. 2A, an example process 200A is illustrated, according to various example embodiments. Various elements of process 200A may be implemented by or via system 100 or components thereof. On the left side is a model generation, training,and / or updating subprocess, and on the right side is a model implementation / inference subprocess. In various embodiments, a process 200A that includes blocks 250A, 255A, and 260A may be performed, without performing blocks 210A, 215B, and 220C. In various other embodiments, a process 200 A that includes blocks 210A, 215B, and 220B may be performed, without performing blocks 250A, 255A, and 260A. In some embodiments, a process 200A that includes blocks 210A, 215A, 220A, 250A, 255A, and 260A may be performed. Process 200 A may begin at block 205 A and either proceed to blocks 210A, 215 A, 220 A, and 295 A (e.g., if a model is to be trained or updated), or begin at block 205A and proceed to blocks 250A, 255A, 260A, and 295A (e.g., if an available model is to be implemented).

[0051] In various embodiments, process 200 A may begin at block 210A with obtaining images and / or other data to be used for training, updating, and / or otherwise generating a machine learning model. The data obtained at block 210A may, at block 215 A, be processed to generate training data suitable for generation of the model. For example, data related to single images or combinations of images may be labeled or otherwise processed to obtain a suitable structure for the training data. At block 220A, the training data may be used to generate one or more models. Process 200A may then end by proceeding to block 295A, or continue to block 250A for model implementation.

[0052] In various embodiments, process 200A may begin at block 250A (or proceed to block 250A from block 220A) with acquisition of data (e.g., imaging data) that includes or otherwise corresponds to a subject. The data may include, for example, one or more images obtained during a scan of the subject. At block 255 A, process 200 A includes analyzing the imaging data (e.g., determining a functional connectivity model and / or generating a metric indicative of relative occupancy of one or more functional networks in the brain). At block 260A, process 200A includes applying a machine learning model (e.g., a classifier) to the acquired data for generation of a prediction. The one or more predictions, or information derived from or based on one or more predictions, may be provided to one or more users.Providing the prediction or information may include, for example, transmitting the prediction or the information to a user computing device, displaying the prediction or the information on a display device or printer, and / or storing the prediction or the information in a computer memory of a networked computing system (such as information system 170) that is accessible to one or more user devices through a network (e.g., the Internet or other network). Process 200A may then end by proceeding to block 295 A. In some embodiments, blocks210A, 215 A, and 220 A may be repeated to update the model based on, for example, updated or otherwise different training data or varied model architecture.100531 Referring to FIG. 2B, an example process 200B is illustrated, according to various example embodiments. Various elements of process 200B may be implemented by or via system 100 or components thereof. At block 210B, imaging data from one or more scans of a subject may be obtained. The imaging data may be from fMRI scans of the subject. The scans may last a minimum time, such as at least 10 minutes, at least 15 minutes, at least 30 minutes, at least 45 minutes, at least 60 minutes, or longer. The scan may be performed using, for example, imager 162 and / or other components of system 100. The scan may be initiated by or via, for example, computing system 110 and / or a user computing device 180. Imaging data may also be acquired from databases 118 and / or from information system 170.|0054| At block 215B, a functional connectivity model may be generated based on the imaging data. In various embodiments, the functionality connectivity model may be a matrix. The functional connectivity model may be based on, may include, or may be, correlations among, or between, cortical vertices and subcortical voxels in the imaging data. The functional connectivity model may be generated by, for example, image analyzer 124.|0055| At block 220B, the functional connectivity model may be used to generate a metric. The metric may relate to a feature or characteristic of one or more functional networks in the brain. For example, the metric may indicate a relative occupancy of one or more functional networks in the brain. In some embodiments, the metric may be generated by network detector 126.

[0056] At block 225B, the metric or derivation thereof may be provided to a machine learning classifier. The classifier may provide a prediction or other output that is related to a prediction concerning a psychological condition of the subject who was scanned. The classifier may be used to generate a prediction by the machine learning platform 130. At block 230B, the output of the classifier, the prediction, or information related to the output and / or the prediction, may be provided to one or more users, such as by transmitting (e.g., using transceiver 114) the prediction or information to one or more user devices 180, by storing the prediction or information in a memory of information system 170 and / or in database 118, and / or by displaying the prediction or information on a display device (e.g., of user interfaces 116, of visualization devices 166, and / or of user interfaces 184). One or moreusers can use the prediction or information to, for example, diagnose a psychological condition, determine a treatment regimen for the condition or its symptoms, characterize severity of a condition or its symptoms, to detect or predict changes to a condition or its symptoms, to evaluate or predict the effectiveness of a treatment, etc.

[0057] In an illustrative example, a multi-echo resting state fMRI scan may be acquired, for at least 15 minutes or, for example, about 30 minutes. In other examples, a single-echo scan may be obtained for a longer duration of, for example, 60 to 120 minutes. The duration of the scan can help obtain stable and clinically useful data. Based on the multiecho fMRI scan, a functional connectivity matrix that summarizes or otherwise is indicative of the correlations between, for example, the BOLD signal timeseries of cortical vertices and / or subcortical voxels (in some embodiments, between all cortical vertices and subcortical voxels). The correlations between nodes less than a threshold distance such as 30 millimeters (mm) apart (geodesic and Euclidean space used for cortico-cortical and sub corti cal -corti cal distance, respectively) to zero. In other implementations, this threshold distance may be greater than 30 mm or less than 30 mm. Correlations between voxels belonging to subcortical structures are also set to zero. The functional connectivity matrix may be thresholded to retain the strongest percent correlations (e.g., 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, 2, and 5%) to each vertex and voxel. Using a community detection algorithm, a graph describing functional connectivity between each vertex and voxel may be generated, grouped into communities. The optimal scale for further analysis is defined as the graph threshold producing the best size-weighted average homogeneity relative to the median of the size- weighted average homogeneity calculated from randomly rotated networks. The communities at this threshold are reviewed and functional network identities assigned on the basis of community topology and connectivity. To calculate the size of each functional network, the process may include first measuring the surface area (in square mm) that each vertex in the individual’s mid-thickness surface is responsible for. Next, the process may include calculating the relative contribution of each functional network to the total cortical surface area by taking the total surface area of all network vertices in relation to the total cortical surface area. This provides a percent occupancy measure for the salience network and all other networks. Depression (and three depression subtypes) can be diagnosed independent of current mood state (i.e. whether or not the individual is currently depressed) by applying a machine learning classifier (e.g., support vector machine) trained to assign a diagnosis (e.g., depression or healthy) or otherwise provide a prediction by weighting thepercentage occupancy of specific brain networks to generate a numerical prediction. Antidepressant response to transcranial magnetic stimulation (TMS) or other treatment can be predicted in a similar way by applying a machine learning classifier (e.g., support vector machine) trained to predict antidepressant response to the treatment (e.g., percent improvement in depressive symptoms) by weighting the percent occupancy of specific brain networks to generate a numerical prediction.

[0058] Example use cases include: diagnostic biomarkers for diagnosing depression in individual patients, as well as diagnosing other related psychological disorders; for identifying individuals at risk for developing depressive symptoms; for identifying the optimal stimulation targets for non-invasive and invasive brain stimulation interventions (e.g., TMS, deep brain stimulation, focused ultrasound); for predicting response to TMS treatments and thus stratifying individual patients to the treatments most likely to benefit them; and / or for objectively quantifying and tracking the severity of mood symptoms in specific domains, which would have applications in clinical trials for example.(0059] Salience network expansion in depression

[0060] Numerous neuroimaging studies involving large cohorts of patients with depression have identified differences in functional connectivity and brain structure, often involving the anterior cingulate cortex, orbitofrontal cortex, insular cortex, and subgenual cingulate cortex — a therapeutic target for deep brain stimulation — but the effect sizes in large-scale meta- analyses are modest (e.g. Cohen’s d= 0.1-0.15 for structural measures and d = 0.13-0.26 for functional connectivity measures). Whether the topological features of large-scale functional brain networks — their shape, spatial location, and size — are altered in depression is unknown.

[0061] Various embodiments of the disclosed approach use precision functional mapping to delineate the topology of functional brain networks in subjects. In an illustrative study, six highly sampled individuals with unipolar major depression underwent on average 621.5 minutes of multi-echo fMRI scanning (range: 58-1,792 minutes) across 22 sessions (range: 2- 62 sessions). We refer to this dataset as the Serial Imaging of Major Depression (SIMD) dataset. To contextualize the severity of depressive symptoms in these individuals, the mean 17-item Hamilton Depression Rating Scale (HDRS17) score (averaged across study visits, excluding those when these individuals were in remission) was 15.7 ± 3.7 (range: 10.5-22.2), indicating a range of severity levels from mild to severe. The same precisionmapping procedures were applied to 37 highly-sampled healthy controls with an average of 327.49 minutes of fMRI data per subject (range: 43.36-841.2 minutes) across 12 sessions (range: 2-84 sessions). The healthy controls did not receive any intervention or treatment. See the Methods section below for additional illustrative details for various example embodiments.

[0062] It was discernable upon visual inspection that the salience network, which is involved in reward processing and conscious integration of autonomic feedback and responses with internal goals and environmental demands, was markedly larger in these individuals with depression (Figs. 3A, 3B). In four of the six individuals, the salience network was expanded more than two-fold, outside the range observed in all 37 healthy controls (Fig. 3C, left). On average, the salience network occupied 73% more of the cortical surface relative to the mean in healthy controls (5.49% ± 0.76% of cortex in SIMD versus 3.17% ± 0.85% of cortex in healthy controls), giving rise to a large group-level effect (Cohen’s d= 1.99). This effect was replicated using an alternative network parcellation algorithm (Fig. 7) and without use of global signal regression (Fig. 8), indicating that it is robust to methodological variation, and was not explained by group differences in brain anatomy or structure (see Fig. 9) or head motion (independent sample t-test comparing mean framewise displacement, T = 0.73, P = 0.47, claims about equivalence are based on an absence of evidence).

[0063] To further validate this finding, this procedure was repeated in three samples (n = 48 and n = 45 from Weill Cornell Medicine, and n = 42 from Stanford University) of individuals with depression. Detailed imaging, demographic, and clinical information for these samples is available in Table 1 and Fig. 10. The effect was replicated thrice (Fig. 3C, right), again with medium to large effect sizes (Cohen’s d= 0.77 - 0.84), remained statistically significant when controlling for the sex ratio imbalance in our samples (56.7% of individuals with depression were female, versus 31% of the healthy controls, see Fig. 10), and with or without correction for potential site- or scanner-induced biases (see Fig. 11). We also evaluated if representation of the salience network was similarly increased in the striatum, which is thought to relate to anatomically well defined, interconnected loops in which the cortex projects to the striatum, and the striatum projects back to cortex indirectly via the thalamus, but found that the difference in group means was not statistically significant (Fig. 3D)

[0064] Salience network expansion in depression was also evident in density maps (Fig. 3E), which convey the percentage of individuals with salience network representation at each cortical vertex or striatal voxel. These maps confirmed a similar overall pattern of cortical and subcortical representation in both groups, consistent with descriptions in previous reports, but also revealed that the borders of the salience network frequently extended further outwards from their centroids in each cortical zone in depressed individuals. For example, in the anterior cingulate cortex, network borders shifted more anteriorly into the pregenual cortex, and in lateral prefrontal cortex, network borders shifted more anteriorly towards the frontal pole (see red boxes in Fig. 3E). Accordingly, expansion of the salience network in cortex was accompanied by contraction of neighboring functional systems in the SIMD sample. However, the specific patterns of contraction did not replicate in all datasets — a finding we return to in the following section. Otherwise, consistent and reproducible group differences in network size were specific to the salience network (no significant differences in the size of any other network after correcting for multiple comparisons).

[0065] To better understand whether this effect was also detectable in large, previously published samples involving conventional single-echo fMRI data, we identified the salience network in group-average functional connectivity data from two large datasets involving n = 812 and n = 120 healthy controls, respectively, and in a third dataset involving n = 299 individuals with treatment-resistant depression scanned in association with a neuromodulation intervention study. The cortical representation of the salience network was >70% larger in the 299-subject depression sample compared to two healthy control samples. Furthermore, highly similar patterns of salience network topology and functional connectivity were produced in split-half analyses of each SIMD dataset, indicating that salience network expansion was a robust and reproducible feature of these highly-sampled individual’s brains.

[0066] Given the magnitude of the effect reported in Fig. 3C, we went on to test whether individuals with depression could be distinguished algorithmically from healthy controls using only the size of each functional network as predictive features. Thus, we trained a linear support vector machine classifier to differentiate individuals with depression from healthy control individuals based on the size of all 20 functional networks, pooling data from n = 37 healthy controls acquired from 5 different scanners and n = 141 individuals with depression acquired from two different scanners / manufacturers (i.e. all the data in Fig. 3C). See the Methods and Fig. 12 for details regarding classifier training. Overall, support vector machine classifiers correctly differentiated depression cases from healthy controls with78.4% accuracy (permutation test, P = 0.001; Fig. 3F), correctly identifying 82.5% of depression cases, for a positive predictive value of 89.5%. Feature importance was evaluated by examining the linear predictor coefficients and calculating the change in accuracy after exclusion (Figs. 3G, 3H). As expected, salience network size was the most distinguishing feature. Together, these analyses indicate that the salience network is markedly expanded in most individuals with depression, with large effect sizes that are reproducible in multiple samples involving different data acquisition and analysis procedures, and sufficient in magnitude to support individual classifications with high accuracy rates.[0067.1 Three salience network expansion modes[0068| Individual differences in functional brain organization occur in two forms: ectopic intrusions, in which isolated pieces of a functional network are observed in an atypical location, and border shifts, in which the boundary of a network expands (or contracts) and encroaches on its neighbors. Border shifts are heritable and associated with known mechanisms of cortical expansion controlled by genetic programs that refine boundaries between functional areas during development and with experience or in response to environmental influences. Furthermore, macroscale networks in both humans and nonhuman primates are organized in a hierarchy associated with cortical gradients in gene expression and functional properties, with unimodal sensorimotor areas at the base and heteromodal association areas such as the default mode network at the apex. Thus, as a first step toward understanding the mechanisms that give rise to cortical expansion of the salience network in depression, we tested whether it was driven primarily by border shifts or ectopic intrusions, and whether it tended to affect lower-level, unimodal sensorimotor networks or heteromodal association areas positioned higher in this hierarchy. To this end, we first generated a central tendency functional network map for the 37 healthy controls (Fig. 4A). Second, we identified parts of the salience network in each of the 141 individuals with depression that did not overlap with the salience network in the group-average map for healthy controls and classified them as either ectopic intrusions or border shifts (Figs. 4B, 4C). Next, we calculated an encroachment profile for each subject, by quantifying the degree of encroachment on every other functional network, defined as the relative contribution of each functional network to the total surface area of the encroaching portion of the salience network (Fig. 4C).

[0069] This analysis confirmed that salience network expansion was not randomly distributed — instead, it was due primarily to border shifts affecting three neighboring higher-order functional systems, with three distinct encroachment profiles occurring in different individuals. Although salience network expansion involved both ectopic intrusions and border shifts, the latter were more common (Fig. 4D), and both tended to result in encroachment on the default, frontoparietal, or cingulo-opercular networks (Fig. 4E), not unimodal sensorimotor networks. Comparison to 73 independent molecular, microstructural, electrophysiological, developmental, and functional brain maps from neuromaps toolbox revealed that salience network expansion frequently occurred in brain regions with less intracortical myelin, and thus greater capacity for synaptic plasticity, and where individual differences in functional connectivity and the concentration of particular neurotransmitter receptors (p-opioid, histamine H3 receptors) are most pronounced. Additional comparisons to maps of FC test-retest reliability and temporal signal-to-noise confirmed that these brain regions were not more susceptible to noise than chance (see Fig. 13). It was also evident that the salience network tended to encroach upon specific functional networks in different cortical zones (Fig. 4F). For example, in the lateral prefrontal cortex, the salience network expanded rostrally and tended to displace the frontoparietal network. In contrast, in the anterior cingulate and anterior insular cortex, the default mode and cingulo-opercular networks were disproportionately affected, respectively. Clustering individuals by their encroachment profiles revealed three distinct modes (Fig. 4G), involving predominantly the default mode network, the frontoparietal network, or a combination of the frontoparietal and cingulo-opercular networks. This heterogeneity may partly explain our observation that the salience network was consistently expanded in all three datasets, but corresponding contractions of other functional networks were more variable.

[0070] The results above indicate that salience network expansion is driven primarily by encroachment upon the frontoparietal, cingulo-opercular, and default mode networks, and suggest that cortical space at the boundary between networks may be allocated to different functional systems in individuals with depression. To test this, and to further validate our findings, we compared the strength of functional connectivity between encroaching nodes of the salience network (dark gray vertices in the left part of Fig. 4C), and the functional networks that typically occupy that space in healthy controls. This analysis was performed using split-halves of each individual’s resting-state fMRI dataset to assess the stability of the salience network assignment associated with the encroaching vertices relative to the runner- up assignments (most often either the default mode, frontoparietal, or cingulo-opercular network). As expected, the functional connectivity of encroaching salience network nodeswith the rest of the salience network was significantly stronger (mean Z( / ) = 0.26) than with the displaced networks (all mean Z(r) < 0.12), consistent with weakened connectivity between encroaching nodes and the functional networks that typically occupy that space in healthy controls. Together, these results show that frontostriatal salience network expansion is driven primarily by network border shifts that affect three specific higher-order functional systems and spare others, with distinct modes of encroachment occurring in three subgroups of patients.[00711 Salience network topology is trait-like

[0072] Major depressive disorder is a fundamentally episodic condition defined by discrete periods of low mood interposed between periods of euthymia. We evaluated if changes in salience network topology accompany changes in the overall severity of depression symptoms that occur during mood state transitions — a hypothesis that our longitudinal SIMD dataset was well-suited to test. However, and consistent with previous work describing functional network topography in healthy adults as very stable features affected very little by cognitive state or daily variation, we found that salience network topology was stable over time in individuals with and without depression (Fig. 5A). Furthermore, within-subject analyses revealed no significant correlation between fluctuations in depression symptoms (HDRS6, a more sensitive measure of changes on shorter timescales) and changes in salience network size over time in any of the densely-sampled individuals in our SIMD dataset (Fig. 5B). To address the same question, we asked whether salience network size changed after a rapid acting antidepressant treatment, leveraging samples of patients scanned before and after a conventional six-week course of repetitive transcranial magnetic stimulation (rTMS; n = 90) or an accelerated, one-week intensive course of rTMS (n = 45). There was no significant pre-to-post change in salience network size in either sample (Fig. 5D). In addition, neither the severity of symptoms during the current episode (Fig. 5E), nor the total number of depressive episodes individuals reported experiencing during their lifetime (Fig. 5F) explained individual differences in salience network size. Collectively, these findings indicate that salience network topology is a stable feature of individuals with major depressive disorder, but not a marker of depressive episodes, and unrelated to the severity of their symptom severity, or to the chronicity of their illness.

[0073] These observations suggest that instead of driving changes in depressive symptoms over time, salience network expansion is a stable marker of risk for developing depression. To test this hypothesis, we asked whether salience network expansion waspresent earlier in life, before the onset of depressive symptoms in individuals. Using data from the Adolescent Brain Cognitive Development (ABCD) study, we identified n=57 children who did not have significant depressive symptoms when they were scanned at ages 10 and 12, but then went on to develop clinically significant depressive symptoms at either age 13 or 14 (Fig. 5G). An equal number of children from the ABCD study with no depressive symptoms at any time point were also identified as a control sample. Precision functional mapping revealed that on average, the salience network occupied 35.93% more of cortex in children with no current or prior symptoms of depression at the time of their fMRI scans, but who subsequently developed clinically significant symptoms of depression, relative to children with no depressive symptoms at any study time point (Fig. 5G: 3.81% ± 1.58% of cortex in ABCD-MDD versus 2.80% ± 1.48% of cortex in ABCD-HC). There was no significant change in salience network size in the two years between the baseline and two- year follow-up visits in either sample (Fig. 14). A similar effect was observed in adults with late-onset depression. Together, these results show that cortical expansion of the salience network is a trait-like feature of brain network organization that is stable over weeks, months, and years, unaffected by mood state, and detectable in children prior to the onset of depression symptoms in adolescence.

[0074] Connectivity state predicts anhedonia

[0075] The results above indicate that topological features of the salience network, such as its size, shape, and spatial location are stable over time and do not fluctuate with mood state. However, this observation does not preclude the possibility that functional connectivity between specific salience network nodes fluctuate in strength, and that such fluctuations contribute to the emergence of depressive episodes and their subsequent remission. To test this, we first asked whether changes in functional connectivity strength between nodes of the salience network either co-occur with or predict fluctuations in symptom severity over time within individuals, focusing initially on hedonic function, a core feature of depression that is associated with frontostriatal circuits, and is aligned with the putative role of salience network, and accumbens-anterior cingulate circuits more specifically, in reward processing and goal-oriented effortful behavior. Our analyses focused on two of the patients from the SIMD dataset (MDD04 and MDD06) who were repeatedly scanned and assessed by clinicians longitudinally over 8-18 months, providing sufficient data for this analysis. This afforded an opportunity to ask for the first time at the level of single, densely sampled individuals — whose data effectively served as independent, well-powered n-of-1 experiments — how variability in brain network functional connectivity relates to fluctuations in specific symptom domains.10076] We began with MDD04 because this individual was studied over the longest period of time (62 study visits over 1.5 years) and had the most fMRI data (29.96 hours of fMRI data in total), and reserved MDD06 as a replication dataset (57 study visits over 12 months, the initial 39 study visits had fMRI data prior to DBS implantation, 18.85 hours of fMRI data in total). During a period spanning over 1.5 years, we observed significant fluctuations in ten anhedonia-related measures (Fig. 6A), which were derived from five standardized depressive symptom scales and identified by a consensus decision by three clinicians (see Fig. 15), ranging from mild / negligible to severe. We tested whether changes in functional connectivity between nodes of the salience network were correlated with changes in anhedonia in this individual over time, as measured by a principal component analysis of the ten anhedonia-related measures in Fig. 4A and summarized by the first component score. We found that functional connectivity between multiple cortical and striatal salience network nodes was correlated with changes in anhedonia over time (Figs. 6C, 6D), with the strongest effects observed for connectivity between the nucleus accumbens and anterior cingulate cortex. An identical analysis in MDD06, involving 39 study visits with clinical and fMRI data over 8 months, replicated this effect (Figs. 6C, 6D). This finding remained significant in both individuals when including head motion at each study visit as a covariate.

[0077] Next, we asked whether salience network functional connectivity was predictive of symptom severity at future study visits and whether the effect was specific to anhedonia or extended to other symptom domains. A cross-correlation analysis examining correlations with symptoms in past, present, and future study visits showed that functional connectivity between the salience network nodes in the nucleus accumbens and anterior cingulate was not only correlated with current anhedonia symptoms but also predicted the future emergence or remission of anhedonia symptoms in the next study visit in MDD04 (Fig. 6E, top panel), typically with a lag of approximately one week. The significance of this effect was confirmed using permutation tests with circular rotation to preserve temporal autocorrelation, indicating that accumbens- anterior cingulate connectivity at a given visit predicted future anhedonia approximately one week later, even after controlling for correlations in anhedonia measures over time. Of note, salience network connectivity correlations were replicated in MDD06 for current symptoms, but not for future symptoms(Fig. 6E, bottom panel), which may relate to differences in their antidepressant treatments in this observational setting (MDD06 was undergoing maintenance electroconvulsive treatment unrelated to this study).

[0078] To determine whether changes in nucleus accumbens-anterior cingulate functional connectivity not only predicted changes in anhedonia within individual subjects over time, but also explained individual differences in anhedonia at a given point in time, we repeated this analysis cross-sectionally using the entire n = 135 cohort of replication subjects using a standardized self-report measure of anhedonia. However, this analysis did not reveal a significant correlation between individual differences in functional connectivity between the anterior cingulate and nucleus accumbens and anhedonia across individuals (Fig. 6F), underscoring the value of within-subject analyses.

[0079] Finally, to evaluate the specificity of this effect, we asked whether nucleus accumbens-anterior cingulate connectivity was also associated with anxiety, a symptom domain that co-occurs with depression, but is often dissociable from anhedonia. For example, “dysphoric” (sadness, anhedonia) and “anxiosomatic” (anxiety, somatic) symptoms were dissociable from one another in a recent study mapping response to rTMS intervention to different stimulation sites. We did not observe a significant correlation between accumbens- anterior cingulate connectivity and anxiety in either individual (Fig. 6G), indicating a more important role for this circuit in anhedonia. Of note, there are several neuroimaging and circuit physiology studies implicating the insula in the expression of anxiety and the processing of aversive states. Motivated by this work, we performed an analogous analysis asking whether changes in striatal connectivity with the anterior insula area of the salience network were correlated with fluctuations in anxiety symptoms over time within each subject. In accord with our prediction, we found that striatal connectivity with anterior insula was significantly correlated with anxiety symptoms in MDD04 and replicated this effect in MDD06 (Fig. 6H). An exploratory whole-brain analysis evaluating how salience network connectivity strength to the rest of the cortex relates to fluctuations in the severity of anhedonia and anxiety symptoms is summarized in Fig. 16. Collectively, these findings show that although the salience network is stably expanded in individuals with depression, and that this expansion appears to occur early in life, frontostriatal connectivity within this network also fluctuates over time, and changes in striatal connectivity with the anterior cingulate and anterior insula track the emergence and remission of anhedonia and anxiety symptoms, respectively.

[0080] Interpreting differences in topology

[0081] In various embodiments, precision functional mapping in deeply sampled individuals with depression revealed a marked expansion of the salience network that was robust and reproducible in multiple samples, with medium to large effect sizes relative to previously reported neuroimaging abnormalities in depression. This effect was driven primarily by network border shifts that encroached on three specific functional systems — the frontoparietal, cingulo-opercular and default mode networks — with three distinct modes of encroachment in different individuals. This effect was stable over time, not sensitive to mood state or a marker of depressive episodes, and emerged early in life in children who went on to develop depressive symptoms later in adolescence. At the same time, changes in striatal connectivity with anterior cingulate and anterior insula nodes of the salience network tracked the emergence and remission of anhedonia and anxiety, respectively, and predicted future changes in hedonic function in one individual. Of note, our analysis benefited from the use of precision functional mapping in combination with large quantities of high-quality, densely- sampled multi-echo fMRI data, which may be critical for mapping individual differences in network topology precisely (see Fig. 18), and this might in part explain why these findings have not been reported in the literature previously.

[0082] Regarding the mechanisms underlying salience network expansion in depression, key results from this report and other studies point to at least two hypotheses. First, converging evidence from multiple sources indicates that individual differences in network topology are regulated by activity-dependent mechanisms and related to the extent to which a given network is actively used. To date, most studies evaluating variability in the size of functional areas or networks across individual humans or other animals have focused primarily on the motor and visual systems. These studies have shown how different body parts have distinct representations in the primary motor cortex (Ml) that differ in size, and cortical representation is closely related to the dexterity of the corresponding limb, such that the upper limbs occupy more cortical surface area than the lower limbs, as one example. Motor training can increase the representation of the trained muscle or limb in Ml, whereas limb amputation, casting, and congenital limb defects all decrease the representation of the disused limb and increase the representation of other body parts. The total surface area of primary visual cortex (VI) can vary up to three-fold in healthy young adults, and is correlated with individual differences in visual awareness and contrast sensitivity. Likewise, total cortical representation of the frontoparietal network was found to be positively correlatedwith executive function abilities in children. Together, these reports suggest that salience network expansion — accompanied by a corresponding contraction of the frontoparietal, cingulo-opercular, or default mode networks — may reflect a reallocation of cortical territory and information processing priorities in individuals with depression, which could in turn contribute to alterations in salience network functions such as interoceptive awareness, reward learning, autonomic signal processing, and effort valuation.

[0083] Second, converging data indicate that cortical network topology is strongly influenced not only by externally modulated, activity-dependent mechanisms but also by intrinsic genetic programs. Numerous transcription factors regulate cell adhesion molecules, exhibit strong expression gradients across the cortical sheet during development, and covary with aspects of cortical organization, including the size or location of functional areas. Deletion of these patterning factors can result in contraction or expansion of functional areas. Conversely, increased expression of Emx2 increases the size of VI and decreases the size of somatomotor areas. Although our findings do not speak directly to this question, at least three observations are consistent with a role for intrinsic developmental genetic programs as opposed to exclusively activity-dependent mechanisms. First, salience network expansion was highly stable, irrespective of an individual’s current mood state, indicating that acute mood-related changes in network activity did not influence network size. Second, salience network expansion emerged early in life, consistent with a developmentally regulated mechanism. And third, salience network expansion was driven by spatially organized border shifts, which are known to be heritable, and which tended to encroach upon neighboring networks in specific directions, expanding anteriorly and disproportionately targeting higher- order heteromodal association cortex while sparing unimodal sensorimotor areas.]0084[ Trait versus state effects in depression

[0085] The disclosed findings may also open new avenues to addressing two fundamental challenges to using insights from clinical neuroimaging research to rethink our approach to diagnosing and treating depression. First, as noted above, MRI studies spanning two decades have identified anatomical and functional connectivity alterations that are robust and reproducible in large-scale meta-analyses but are highly variable across subjects with modest effect sizes (typically, = 0.10 - 0.35), which complicates effects to leverage these effects for clinical purposes. In contrast, salience network expansion was observed in most individuals with depression in our sample, readily apparent on visual inspection, and associated with medium to large effect sizes ( = 0.77-1.99). This effect was detected withoutcorrections for site- or scanner-induced biases — which can be a significant confound for multi-site neuroimaging data.

[0086] Biomarkers in multiple areas of medicine come in different forms, some of which are sensitive to current symptoms, while others are stable trait-like markers of disease, or a marker of risk for developing symptoms. The characteristics of the brain revealed through embodiments of the disclosed approach can be applied to other forms of psychopathology. For example, a preliminary analysis indicated that the salience network is also larger than normal in two individuals with bipolar II disorder, but not autism spectrum disorder or obsessive compulsive disorder (see Fig. 17), which might reflect common deficits in behavioral domains, such as reward processing, that are also linked to salience network function. However, various embodiments of the disclosed approach can use salience network expansion to help predict susceptibility to depression symptoms, and could have important implications for designing therapeutic neuromodulation interventions, which could have widely varying effects due to individual differences in network topology. In some embodiments, precise and reliable mapping of the salience network and consistent detection of salience network expansion in individuals with depression may require 1.5 to 2 hours of high-quality fMRI data per-subject (see Fig. 18), which may be an obstacle for retrospective analysis of traditional fMRI datasets not optimized for precision functional mapping at the individual -level. It is also noteworthy that the brain network that we and others have referred to as the salience network is sometimes called other names (e.g., “Control C”) or combined with the parietal memory network, whereas the brain network we refer to as cingulo- opercular / action-mode network is sometimes called the “Salience / Ventral Attention” network (see Fig. 19). Developing a standardized functional brain network nomenclature will improve the interpretability of insights gleaned from precision functional mapping like in the present study.|0087] Further, the disclosed approach supports the use of precision functional mapping and deep, longitudinal sampling for understanding cause and effect in clinical neuroimaging studies of depression. The analyses discussed above reveal stable, trait-like differences in salience network topology that are not only associated with depression but also emerge early in life in children with no history of depression and predict the subsequent emergence of depressive symptoms in adolescence. At the same time, they show how changes in functional connectivity strength between specific salience network nodes track the emergence and remission of dysfunction in specific symptom domains within individualsover time, and in at least one individual, predict the future emergence of anhedonia symptoms at least one week before they occur. In this way, they show how dense sampling and longitudinal designs will open new avenues for understanding cause and effect and for designing personalized, prophylactic treatments.

[0088] METHODS FOR EXAMPLE EMBODIMENTS

[0089] Datasets(0090] The datasets used are described briefly below. Overall, the depression sample collectively consisted of 141 individuals (mean age = 40.71 ± 13.82, 56.7% female) with a diagnosis of major depression (based on DSM-IV-TR criteria and confirmed by the MiniInternational Neuropsychiatric Interview administered by a trained clinician) drawn from 5 sources — the Serial Imaging of Major Depression (SIMD, mean age = 29.47 ± 8.28 years, 3F / 3M), Weill Cornell rTMS 1 (conventional 6-week rTMS, mean age = 40.89 ± 12.73 years, 27F / 21M), Weill Cornell rTMS 2 (accelerated, 1-week rTMS, mean age = 40.89 ± 12.73 years, 21F / 24M), Stanford University rTMS (conventional 6-week rTMS, mean age = 38.09 ± 12.77 years, 29F / 13M), and Weill Cornell Late-onset Depression datasets (mean age = 66.60 ± 5.31 years, 5F / 0M). The healthy control sample collectively consisted of 37 healthy adults (mean age = 31.72 ± 7.08 years, 1 IF) drawn from 6 sources — the Weill Cornell Multi-echo (mean age = 33.42 ± 9.10 years, 0F / 7M), MyConnectome (a single 45 year-old male), Midnight Scan Club (MSC; mean age = 29.1 ± 3.3 years, 5F / 5M), Cast-induced Plasticity (a single 27 year-old male), Natural Scenes Dataset (NSD; mean age = 26.50 ± 4.24 years, 6F / 2M), and Eskalibur datasets (mean age = 31.4 ± 5.4 years, 5F / 5M). We note that the MSC, MyConnectome, and Cast Induced Plasticity study were obtained online (https: / / openneuro.org / ) in a preprocessed, fully denoised, and surface-registered format, and no additional preprocessing or denoising was performed for the present study.

[0091] MRI acquisition

[0092] Serial Imaging of Major Depression dataset: Data were acquired on a Siemens Magnetom Prisma 3T scanner at the Citigroup Biomedical Imaging Center of Weill Cornell's medical campus using a Siemens 32-channel head coil. Two multi-echo, multi-band resting-state fMRI scans were collected using a T2*-weighted echo-planar sequence covering the full brain (TR: 1355 ms; TE1 : 13.40 ms, TE2: 31.11 ms, TE3: 48.82 ms, TE4: 66.53 ms, and TE5: 84.24 ms; FOV: 216 mm; flip angle: 68° (the Ernst angle for gray matter assuming a T1 value of 1400 ms); 2.4 mm isotropic voxels; 72 slices; AP phaseencoding direction; in-plane acceleration factor: 2; and multi-band acceleration factor: 6) with 640 volumes acquired per scan for a total acquisition time of 14 minutes and 27 seconds. Spin echo EPI images with opposite phase encoding directions (AP and PA) but identical geometrical parameters and echo spacing were acquired before each resting-state scan. Multi-echo Tl-weighted (TR / TI: 2500 / 1000 ms; TE1 : 1.7 ms, TE2: 3.6 ms, TE3: 5.5 ms, TE4: 7.4 ms ; FOV: 256 mm; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) and T2-weighted anatomical images (TR: 3200 ms; TE: 563 ms; FOV: 256; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) were acquired at the end of each session.

[0093] Weill Cornell rTMS 1 and 2 datasets: MRI data were acquired on a Siemens Magnetom Prisma 3T machine at the Citigroup Biomedical Imaging Center of Weill Cornell's medical campus using a Siemens 32-channel head coil. Two multi-echo, multiband resting-state fMRI scans were collected at each study visit using a T2*-weighted echo- planar sequence covering the full brain (TR: 1300 ms; TE1 : 12.60 ms, TE2: 29.51 ms, TE3: 46.42 ms, and TE4: 63.33 ms; FOV: 216 mm; flip angle: 67° (the Ernst angle for gray matter assuming a T1 value of 1400 ms); 2.5 mm isotropic voxels; 60 slices; AP phase encoding direction; in-plane acceleration factor: 2; and multi-band acceleration factor: 4) with 650 volumes acquired per scan for a total acquisition time of 14 minutes and 5 seconds. Spin echo EPI images with opposite phase encoding directions (AP and PA) but identical geometrical parameters and echo spacing were acquired before each resting-state scan. Multi-echo Tl-weighted (TR / TI: 2500 / 1000 ms; TE1 : 1.7 ms, TE2: 3.6 ms, TE3: 5.5 ms, TE4: 7.4 ms ; FOV: 256; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) and T2-weighted anatomical images (TR: 3200 ms; TE: 563 ms; FOV: 256; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) were acquired at the end of each session.10094] Stanford University rTMS dataset: MRI data were acquired on a GE SIGNA 3T machine at the Center for Neurobiological Imaging on Stanford University’s campus using a Nova Medical 32- channel head coil. Four multi-echo, multi-band resting-state fMRI scans were collected using a T2*-weighted echo-planar sequence covering the full brain (TR: 1330 ms; TE1 : 13.7 ms, TE2: 31.60 ms, TE3: 49.50 ms, and TE4: 67.40 ms; flip angle: 67° (the Ernst angle for gray matter assuming a T1 value of 1400 ms); 3 mm isotropic voxels; 52 slices; AP phase encoding direction; in-plane acceleration factor: 2; and multi-band acceleration factor: 4) with 338 volumes acquired per scan for a total acquisitiontime of 7 minutes and 30 seconds. Spin echo EPI images with opposite phase encoding directions (AP and PA) but identical geometrical parameters and echo spacing were acquired before each resting-state scan. T1 -weighted and T2-weighted anatomical images were acquired at the end of each session.

[0095] Weill Cornell Late-onset Depression dataset: MRI data were acquired on a Siemens Magnetom Prisma 3T machine at the Citigroup Biomedical Imaging Center of Weill Cornell's medical campus using a Siemens 32-channel head coil. Two multi-echo, multi-band resting-state fMRI scans were collected at each study visit using a T2*-weighted echo-planar sequence covering the full brain (TR: 1300 ms; TE1 : 12.60 ms, TE2: 29.51 ms, TE3: 46.42 ms, and TE4: 63.33 ms; FOV: 216 mm; flip angle: 67° (the Ernst angle for gray matter assuming a T1 value of 1400 ms); 2.5 mm isotropic voxels; 60 slices; AP phase encoding direction; in-plane acceleration factor: 2; and multi-band acceleration factor: 4) with 480 volumes acquired per scan for a total acquisition time of 10 minutes and 38 seconds. Spin echo EPI images with opposite phase encoding directions (AP and PA) but identical geometrical parameters and echo spacing were acquired before each resting-state scan. Multi-echo Tl-weighted (TR / TI: 2500 / 1000 ms; TE1 : 1.7 ms, TE2: 3.6 ms, TE3: 5.5 ms, TE4: 7.4 ms ; FOV: 256 mm; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) and T2-weighted anatomical images (TR: 3200 ms; TE: 563 ms; FOV: 256; flip angle: 8°, and 208 sagittal slices with a 0.8 mm slice thickness) were acquired at the end of each session.

[0096] Anatomical preprocessing and cortical surface generation

[0097] Anatomical data were preprocessed and cortical surfaces generated using the Human Connectome Project (HCP) PreFreeSurfer, FreeSurfer, and PostFreeSurfer pipelines (version 4.3).

[0098] Multi-echo fMRI preprocessing

[0099] Preprocessing of multi-echo data minimized spatial interpolation and volumetric smoothing while preserving the alignment of echoes. The single-band reference (SBR) images (one per echo) for each scan were averaged. The resultant average SBR images were aligned, averaged, co- registered to the ACPC aligned Tl-weighted anatomical image, and simultaneously corrected for spatial distortions using FSL’s topup and epi reg programs. Freesurfer’ s bbregister algorithm was used to refine this co-regi strati on. For each scan, echoes were combined at each timepoint and a unique 6 DOF registration (one pervolume) to the average SBR image was estimated using FSL’s MCFLIRT tool, using a 4- stage (sine) optimization. All of these steps (co-registration to the average SBR image, ACPC alignment, and correcting for spatial distortions) were concatenated using FSL’s convertwarp tool and applied as a single spline warp to individual volumes of each echo after correcting for slice time differences using FSL’s slicetimer program. The functional images underwent a brain extraction using the co-registered brain extracted Tl-weighted anatomical image as a mask and corrected for signal intensity inhomogeneities using ANT’s N4BiasFieldCorrection tool. All denoising was performed on preprocessed, ACPC- aligned images.

[0100] Multi-echo fMRI denoising

[0101] Preprocessed multi-echo data were submitted to multi-echo ICA (ME-ICA), which is designed to isolate spatially structured T2*- (neurobiological; “BOLD-like”) and SO-dependent (non-neurobiological; “not BOLD-like”) signals and implemented using the “tedana.py” workflow. In short, the preprocessed, ACPC-aligned echoes were first combined according to the average rate of T2* decay at each voxel across all time points by fitting the monoexponential decay, S(t) = SOe-l'T2* . From these T2* values, an optimally- combined multi-echo (OC-ME) time-series was obtained by combining echoes using a weighted average (WTE = TE * e -TE / T2*). The covariance structure of all voxel timecourses was used to identify major signals in the OC-ME time-series using principal component and independent component analysis. Components were classified as either T2*-dependent (and retained) or SO-dependent (and discarded), primarily according to their decay properties across echoes. Mean gray matter time-series regression was performed to remove spatially diffuse noise. Temporal masks were generated for censoring high motion time-points using a framewise displacement (FD) threshold of 0.3 mm and a backward difference of two TRs, for an effective sampling rate comparable to historical FD measurements (approximately 2 to 4 seconds). Prior to the FD calculation, head realignment parameters were filtered using a stopband Butterworth filter (0.2 - 0.35 Hz) to attenuate the influence of respiration on motion parameters.

[0102] Single-echo fMRI denoising

[0103] The following denoising procedures were applied to the NSD and ABCD datasets. The NSD dataset was obtained in an already preprocessed (but not yet denoised) format. For the ABCD data, Fast Track (unprocessed) neuroimaging data was obtained viaNDA command line utilities (https: / / github.com / NDAR / nda-tools) and subjected to the preprocessing steps used for multi-echo fMRI data (omitting steps involving combination of echoes). Preprocessed single-echo data were then submitted to ICA-AROMA. Mean gray matter time-series regression was performed to remove spatially diffuse noise. Temporal masks were generated for censoring high motion time-points, as done for the multi-echo fMRI datasets.

[0104] Surface processing and CIFTI generation of fMRI data

[0105] The denoised fMRI time-series was mapped to the individual’s fsLR 32k midthickness surfaces with native cortical geometry preserved (using the “-ribbon- constrained” method), combined into the Connectivity Informatics Technology Initiative (CIFTI) format, and spatially smoothed with geodesic (for surface data) and Euclidean (for volumetric data) Gaussian kernels (c = 2.55 mm) using Connectome Workbench command line utilities. This yielded time courses representative of the entire cortical surface, subcortex (accumbens, amygdala, caudate, hippocampus, pallidum, putamen, thalamus, brainstem), and cerebellum, but excluding non-gray matter tissue. Spurious coupling between subcortical voxels and adjacent cortical tissue was mitigated by regressing the average time-series of cortical tissue < 20 mm in Euclidean space from a subcortical voxel.

[0106] Precision mapping of functional brain networks in individuals

[0107] A functional connectivity matrix summarizing the correlation between the time-courses of all cortical vertices and subcortical voxels across all study visits was constructed. Correlations between nodes < 10 mm apart (geodesic and Euclidean space used for cortico-cortical and subcortical-cortical distance, respectively) were set to zero. Correlations between voxels belonging to subcortical structures were set to zero. Functional connectivity matrices were thresholded in such a way that they retained at least the strongest X% correlations (0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, 2, and 5%) to each vertex and voxel and were used as inputs for the InfoMap community detection algorithm, one of the most widely used approaches for delineating functional brain networks and their boundaries in individuals. Free parameters (for example, the number of algorithm repetitions) for the Infomap algorithm were fixed across subjects. The total number of communities identified by Infomap is controlled in part by how many connections are retained in the functional connectivity matrix after thresholding. The optimal scale for further analysis across individuals was defined as the graph threshold producing the best size-weighted averagehomogeneity relative to the median of the size-weighted average homogeneity calculated from randomly rotated networks. Size-weighted average homogeneity was maximized relative to randomly rotated communities at the 0.1% graph density and resulted in 89.13 ± 8.04 communities on average across individuals.

[0108] Each Infomap community was algorithmically assigned to one of 20 possible functional network identities (Default-Parietal, Default-Anterolateral, Default-Dorsolateral, Default-Retrosplenial, Visual -Lateral, Visual-Stream, Visual-Vl, Visual-V5, Frontoparietal, Dorsal Attention, Premotor / Dorsal Attention II, Language, Salience, Cingulo-opercular / Action-mode, Parietal memory, Auditory, Somatomotor-Hand, Somatomotor-Face, Somatomotor-Foot, Auditory, or Somato-Cognitive- Action) primarily according their functional connectivity and spatial locations relative to a specified set of priors. See Fig. 20 for additional details regarding algorithmic assignments, and Figs. 21-22 for the visualizations of the functional network priors used in this study.

[0109] Functional brain networks were also mapped brain-wide using the multiplex version of the InfoMap community detection algorithm. In a multiplex network, physical nodes (brain regions) can exist in multiple layers (study visits). A temporal network (node x node x study visit) summarizing the correlation between the time-courses of all cortical vertices and subcortical voxels across study visits was constructed for each patient. Correlations between nodes less than 10 mm apart (geodesic and Euclidean space used for cortico-cortical and subcortical-cortical distance, respectively) were set to zero.Correlations between voxels belonging to subcortical structures were set to zero. Links between layers were generated automatically using neighborhood flow coupling. The temporal distance between layers was constrained to 1 using the multilayer-relax-limif ’ option to encode the temporal order of study visits. Multiplex functional network parcellations were used for the analyses performed in the “ Salience network topology is trail-like" section and Figs. 5A-5E.[01.10] Calculating functional network size and spatial locations in individuals10111] We first measured the surface area (in mm2) that each vertex in the individual’s midthickness surface is responsible for (“wb command —surface-vertex- areas”). Next, we calculated the relative contribution (size) of each functional network to the total cortical surface area by taking the total surface area of all network vertices in relation to the total cortical surface area. In the striatum, where each voxel represents thesame amount of tissue, the relative contribution of each functional network to the total striatal volume was calculated by taking the total number of network voxels in relation to the total striatal voxels. The statistical significance of group differences in network size were evaluated using permutation tests and independent sample t- tests (the latter implemented using Matlab’s ttest2.m function). Effect size (Cohen’s d) was calculated as the difference in group means divided by pooled standard deviation. Assumptions regarding equal variance were adjusted when appropriate (based on two-sample F-tests performed using Matlab’s vartest2.m function). The relative difference between groups was calculated as the absolute difference divided by network size in healthy controls. Density maps were created by calculating the percentage of individuals with salience network representation at each cortical vertex or striatum voxel. These procedures collectively correspond to the analyses performed in the “Connectivity state predicts anhedonia" section and Figs. 3C-3E.

[0112] Classification analysis10H3] Functional network size (the % of total cortical surface area occupied by each network, 20 networks / features total) were used as predictive features in a support vector machine classifier to distinguish individuals with depression and healthy controls. The model was trained using repeated (100 iterations) nested split-half (2-fold) cross- validation with a grid search optimization strategy for hyperparameter tuning (box constraint, kernel size). The Synthetic Minority Oversampling Technique (SMOTE) was used to prevent classification bias in favor of the majority class, and was performed on training data only to prevent data leakage. Classification accuracy was calculated as the percentage of correct predictions, and statistical significance assessed using permutation tests (shuffled diagnostic labels, 1000 iterations). A confusion matrix was created using Matlab’s confmat.m function. Feature importance was evaluated by iteratively omitting each functional network and evaluating the resulting loss in accuracy. These procedures collectively are related to the analyses performed in the “Salience network expansion in depression" section and Figs. 3F-3I.

[0114] Evaluating how salience network expansion displaces other functional systems

[0115] The parts of each depressed individual’s salience network map that did and did not overlap with the salience network in the group average healthy control map were operationalized as “non- encroaching” and “encroaching”, respectively. The group averagehealthy control map was obtained by calculating the mode assignment across healthy controls at each point in the brain. Encroaching clusters were identified (“wb command - cifti-find-clusters”) and were classified as border shifts if any part of the cluster was within 3.5 mm (in geodesic space) of a salience network vertex in the group average healthy control map, and as ectopic intrusions if they did not. An encroachment profile was calculated as the relative contribution of each functional network to the total surface area of the encroaching portion of the salience network. Individuals were clustered on the basis of the similarity of their encroachment profiles using the Louvain method(“community louvain.m” function from the Brain Connectivity Toolbox). These procedures correspond to the analyses performed in the “ Three salience network expansion modes" section and Figs. 4A-4G.

[0116] Assessing the stability of salience network topography across time10117] The multiplex versions of each individual’s salience network were used to assess the extent to which network topography (size) varied across study time points in highly-sampled individuals with and without depression. Variability in salience network size was correlated with the overall severity of depressive symptoms (Hamilton Depression Rating Scale, HDRS6) using Matlab’s corr.m function. In the replication samples (“Weill Cornell Medicine rTMS 1”, “Weill Cornell Medicine rTMS 2”, “Stanford University rTMS” samples), we assessed pre-to-post change in salience network topography using paired two-tailed paired sample / -tests via Matlab’s ttest.m function. Data were binned according to treatment duration (conventional 6-week rTMS or accelerated 1-week rTMS). The number of depressive episodes in each individual’s lifetime was inferred from their Mini-International Neuropsychiatric Interview. These procedures collectively correspond to the analyses performed in the “Salience network topology is trail-like" section and Fig. 5.[0118[ Evaluating salience network topography early in life prior to symptom onset

[0119] We used the ABCD dataset (release 5.0) to test if atypical salience network topology precedes the onset of depression symptoms. Symptoms of depression in the ABCD study were operationalized using the ASEBA DSM-oriented scale for depression (“mh_p_cbcl.csv”) from the ABCD Parent Child Behavior Checklist (CBCL). After excluding subjects with missing behavioral data or those with MRI data flagged internally ABCD for data quality issues, we identified n = 58 subjects (37F) meeting criteria for onsetof clinical depression symptoms at the 3-year follow-up (t-score > 70 at or after the 3-year follow-up and t-scores < 65 at the previous study visits). One participant’s data was not accessible on Fast Track, resulting in n = 57 total. An equal number of subjects with no clinically significant depression symptoms at any study time point (t-scores < 65 at all study time points) were randomly selected as a control sample. The statistical significance of group differences in salience network size were evaluated using permutation tests and independent sample t-tests (the latter implemented using Matlab’s ttest2.m function). Assumptions regarding equal variance were adjusted when appropriate (based on two- sample F-tests performed using Matlab’s vartest2.m function). These procedures collectively correspond to the analyses performed in the “Salience network topology is traitlike” section and Fig. 5G.

[0120] Longitudinal analyses relating changes in connectivity with symptom severity

[0121] Composite measures of anhedonia and anxiety related symptoms were obtained instructing three clinicians to quantify (on a scale of 0-3; 0 = not at all, 1 = somewhat, 2 = largely, 3 = very strongly) the extent to each item from the battery of clinical scales administered to the SIMD subjects reflects anhedonia or anxiety related symptoms. Items assigned a score of 1 or greater by all three clinicians were included in the composite measures (see Fig. 15). For each subject, separately, the consensus items were min-max normalized, adjusted for valence (so that higher scores reflect more severe symptoms across all items), and then subjected to a principal component analysis to extract a time course (PCI) of anhedonia or anxiety severity across study visits. To validate this approach, we quantified the similarity to validated measures of anhedonia and anxiety using independent data, and observed good correspondence (Pearson correlations > 0.4). Functional connectivity strength between all pairs of cortical (anterior cingulate, lateral prefrontal, anterior insula cortex) and striatal (nucleus accumbens, caudate, putamen) nodes of the salience network was calculated for each study visit, separately, and correlated with the anhedonia or anxiety PCI. This analysis was constrained to the three major cortical and striatal nodes of the salience network in part to reduce the likelihood of false positives. Correlations not exceeding chance (based on null distribution of correlation coefficients obtained using rotated clinical data) were set to zero. Circular permutation tests (using Matlab’s circshift.m function) were used to preserve temporal autocorrelation. Cross-correlation analyses were performed using Matlab’s crosscorr.m function (with “NumLags” set to 2). For the cross-sectional analysis, the totalSnaith-Hamilton Pleasure Scale (SHAPS) score was calculated using baseline clinical data. These procedures collectively correspond to the analyses performed in the “Connectivity state predicts anhedonia" section and Fig. 8.

[0122] Supplemental description of rTMS interventions

[0123] The first rTMS trial is multi-site (data collected at Weill Cornell Medicine in New York, NY and Stanford University in Palo Alto, CA) and is associated with the “Weill Cornell Medicine rTMS 1” and “Stanford University rTMS” study samples. MRI data was collected on average 16 ± 18 days prior to and 9 ± 9 days after treatment. Participants received 30 sessions of intermediate theta burst stimulation (iTBS; triplet 50 Hz bursts, repeated at 5 Hz; 2 seconds on and 8 seconds off; 600 pulses per session; total duration of 3 minutes 9 seconds) once daily on weekdays over 6 to 8 weeks using a MagPro X100 / R30 stimulator. Participants were allocated to receive iTBS to either their left dorsolateral prefrontal cortex (DLPFC, MNI coordinates X = -38, Y = 44, Z = 26) or dorsomedial prefrontal cortex (DMPFC, MNI coordinates X = 0, Y = 60, Z = 60). For left DLPFC treatment, the Cool-B65 coil was positioned at a 45-degree angle relative to the anterior- posterior midline, with treatment delivered at 120% resting motor threshold (RMT). For DMPFC treatment, the C00I-DB8O fluid-cooled coil was oriented at 180 degrees relative to the anterior-posterior midline, with treatment delivered at 120% RMT. The Brainsight 2 Frameless stereotactic system for image guided TMS research (Rogue Research) was used for neuronavigation. This system uses infrared reflectors attached to a headband worn by the subject to coregister the Tlw image with the participant’s head.

[0124] The second rTMS trial is associated with the “Weill Cornell Medicine rTMS 2” study sample. MRI data was collected on average 12 ± 15 days prior to treatment and 14 ± 23 days after treatment. Participants received 50 sessions (10 sessions per-day, one every hour) of iTBS (triplet 50 Hz bursts, repeated at 5 Hz; 2 seconds on and 8 seconds off; 1,800 pulses per session; total duration of 9 minutes 39 seconds) over 5 consecutive weekdays using a MagPro X100 / R30 stimulator. The stimulation target in each individual’s DLPFC was based on their subgenual functional connectivity. The Brainsight 2 Frameless stereotactic system for image guided TMS research (Rogue Research) was used for neuronavigation.TABLE 1: Demographic, imaging, and clinical information for study samples

[0125] Various operations described herein can be implemented on computer systems having various configurations. FIG. 23 shows a simplified block diagram of a representative server system 2300 (e.g., computing system 110 or other components in FIG. 1) and client computer system 2314 (e.g., computing system 110, condition detection system 160, information system 230, and / or guidance system 180) usable to implement variousembodiments of the present disclosure. In various embodiments, server system 2300 or similar systems can implement services or servers described herein or portions thereof. Client computer system 2314 or similar systems can implement clients described herein.

[0126] Server system 2300 can have a modular design that incorporates a number of modules 2302 (e.g., blades in a blade server embodiment); while two modules 2302 are shown, any number can be provided. Each module 2302 can include processing unit(s) 2304 and local storage 2306.

[0127] Processing unit(s) 2304 can include a single processor, which can have one or more cores, or multiple processors. In some embodiments, processing unit(s) 2304 can include a general-purpose primary processor as well as one or more special-purpose coprocessors such as graphics processors, digital signal processors, or the like. In some embodiments, some or all processing units 2304 can be implemented using customized circuits, such as application specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs). In some embodiments, such integrated circuits execute instructions that are stored on the circuit itself. In other embodiments, processing unit(s) 2304 can execute instructions stored in local storage 2306. Any type of processors in any combination can be included in processing unit(s) 2304.

[0128] Local storage 2306 can include volatile storage media (e.g., conventional DRAM, SRAM, SDRAM, or the like) and / or non-volatile storage media (e.g., magnetic or optical disk, flash memory, or the like). Storage media incorporated in local storage 2306 can be fixed, removable or upgradeable as desired. Local storage 2306 can be physically or logically divided into various subunits such as a system memory, a read-only memory (ROM), and a permanent storage device. The system memory can be a read-and-write memory device or a volatile read-and-write memory, such as dynamic random-access memory. The system memory can store some or all of the instructions and data that processing unit(s) 2304 need at runtime. The ROM can store static data and instructions that are needed by processing unit(s) 2304. The permanent storage device can be a non-volatile read-and-write memory device that can store instructions and data even when module 2302 is powered down. The term “storage medium” as used herein includes any medium in which data can be stored indefinitely (subject to overwriting, electrical disturbance, power loss, or the like) and does not include carrier waves and transitory electronic signals propagating wirelessly or over wired connections.

[0129] In some embodiments, local storage 2306 can store one or more software programs to be executed by processing unit(s) 2304, such as an operating system and / or programs implementing various server functions or any system or device described herein.

[0130] Software” refers generally to sequences of instructions that, when executed by processing unit(s) 2304 cause server system 2300 (or portions thereof) to perform various operations, thus defining one or more specific machine embodiments that execute and perform the operations of the software programs. The instructions can be stored as firmware residing in read-only memory and / or program code stored in non-volatile storage media that can be read into volatile working memory for execution by processing unit(s) 2304. Software can be implemented as a single program or a collection of separate programs or program modules that interact as desired. From local storage 2306 (or non-local storage described below), processing unit(s) 2304 can retrieve program instructions to execute and data to process in order to execute various operations described above.

[0131] In some server systems 2300, multiple modules 2302 can be interconnected via a bus or other interconnect 2308, forming a local area network that supports communication between modules 2302 and other components of server system 2300. Interconnect 2308 can be implemented using various technologies including server racks, hubs, routers, etc.

[0132] A wide area network (WAN) interface 2310 can provide data communication capability between the local area network (interconnect 2308) and a larger network, such as the Internet. Conventional or other activities technologies can be used, including wired (e.g., Ethernet, IEEE 802.3 standards) and / or wireless technologies (e.g., Wi-Fi, IEEE 802.11 standards).

[0133] In some embodiments, local storage 2306 is intended to provide working memory for processing unit(s) 2304, providing fast access to programs and / or data to be processed while reducing traffic on interconnect 2308. Storage for larger quantities of data can be provided on the local area network by one or more mass storage subsystems 2312 that can be connected to interconnect 2308. Mass storage subsystem 2312 can be based on magnetic, optical, semiconductor, or other data storage media. Direct attached storage, storage area networks, network-attached storage, and the like can be used. Any data stores or other collections of data described herein as being produced, consumed, or maintained by a service or server can be stored in mass storage subsystem 2312. In some embodiments,additional data storage resources may be accessible via WAN interface 2310 (potentially with increased latency).10134] Server system 2300 can operate in response to requests received via WAN interface 2310. For example, one of modules 2302 can implement a supervisory function and assign discrete tasks to other modules 2302 in response to received requests. Conventional work allocation techniques can be used. As requests are processed, results can be returned to the requester via WAN interface 2310. Such operation can generally be automated. Further, in some embodiments, WAN interface 2310 can connect multiple server systems 2300 to each other, providing scalable systems capable of managing high volumes of activity. Conventional or other techniques for managing server systems and server farms (collections of server systems that cooperate) can be used, including dynamic resource allocation and reallocation.|0135| Server system 2300 can interact with various user-owned or user-operated devices via a wide-area network such as the Internet. An example of a user-operated device is shown in FIG. 23 as client computing system 2314. Client computing system 2314 can be implemented, for example, as a consumer device such as a smartphone, other mobile phone, tablet computer, wearable computing device (e.g., smart watch, eyeglasses), desktop computer, laptop computer, and so on.[0136| Client computing system 2314 can communicate via WAN interface 2310. Client computing system 2314 can include conventional computer components such as processing unit(s) 2316, storage device 2318, network interface 2320, user input device 2322, and user output device 2324. Client computing system 2314 can be a computing device implemented in a variety of form factors, such as a desktop computer, laptop computer, tablet computer, smartphone, other mobile computing device, wearable computing device, or the like.

[0137] Processor 2316 and storage device 2318 can be similar to processing unit(s) 2304 and local storage 2306 described above. Suitable devices can be selected based on the demands to be placed on client computing system 2314; for example, client computing system 2314 can be implemented as a “thin” client with limited processing capability or as a high-powered computing device. Client computing system 2314 can be provisioned with program code executable by processing unit(s) 2316 to enable various interactions with server system 2300 of a message management service such as accessing messages,performing actions on messages, and other interactions described above. Some client computing systems 2314 can also interact with a messaging service independently of the message management service.

[0138] Network interface 2320 can provide a connection to a wide area network (e.g., the Internet) to which WAN interface 2310 of server system 2300 is also connected. In various embodiments, network interface 2320 can include a wired interface (e.g., Ethernet) and / or a wireless interface implementing various RF data communication standards such as Wi-Fi, Bluetooth, or cellular data network standards (e.g., 3G, 4G, 5G, LTE, etc.).

[0139] User input device 2322 can include any device (or devices) via which a user can provide signals to client computing system 2314; client computing system 2314 can interpret the signals as indicative of particular user requests or information. In various embodiments, user input device 2322 can include any or all of a keyboard, touch pad, touch screen, mouse or other pointing device, scroll wheel, click wheel, dial, button, switch, keypad, microphone, and so on.

[0140] User output device 2324 can include any device via which client computing system 2314 can provide information to a user. For example, user output device 2324 can include a display to display images generated by or delivered to client computing system 2314. The display can incorporate various image generation technologies, e.g., a liquid crystal display (LCD), light-emitting diode (LED) including organic light-emitting diodes (OLED), projection system, cathode ray tube (CRT), or the like, together with supporting electronics (e.g., digital-to-analog or analog-to-digital converters, signal processors, or the like). Some embodiments can include a device such as a touchscreen that function as both input and output device. In some embodiments, other user output devices 2324 can be provided in addition to or instead of a display. Examples include indicator lights, speakers, tactile “display” devices, printers, and so on.(0141] Some embodiments include electronic components, such as microprocessors, storage and memory that store computer program instructions in a computer readable storage medium. Many of the features described in this specification can be implemented as processes that are specified as a set of program instructions encoded on a computer readable storage medium. When these program instructions are executed by one or more processing units, they cause the processing unit(s) to perform various operation indicated in the program instructions. Examples of program instructions or computer code include machine code, suchas is produced by a compiler, and files including higher-level code that are executed by a computer, an electronic component, or a microprocessor using an interpreter. Through suitable programming, processing unit(s) 2304 and 2316 can provide various functionality for server system 2300 and client computing system 2314, including any of the functionality described herein as being performed by a server or client, or other functionality associated with message management services.

[0142] It will be appreciated that server system 2300 and client computing system 2314 are illustrative and that variations and modifications are possible. Computer systems used in connection with embodiments of the present disclosure can have other capabilities not specifically described here. Further, while server system 2300 and client computing system 2314 are described with reference to particular blocks, it is to be understood that these blocks are defined for convenience of description and are not intended to imply a particular physical arrangement of component parts. For instance, different blocks can be but need not be located in the same facility, in the same server rack, or on the same motherboard. Further, the blocks need not correspond to physically distinct components. Blocks can be configured to perform various operations, e.g., by programming a processor or providing appropriate control circuitry, and various blocks might or might not be reconfigurable depending on how the initial configuration is obtained. Embodiments of the present disclosure can be realized in a variety of apparatus including electronic devices implemented using any combination of circuitry and software.

[0143] While the disclosure has been described with respect to specific embodiments, one skilled in the art will recognize that numerous modifications are possible. For instance, although specific examples of rules (including triggering conditions and / or resulting actions) and processes for generating suggested rules are described, other rules and processes can be implemented. Embodiments of the disclosure can be realized using a variety of computer systems and communication technologies including but not limited to specific examples described herein.

[0144] Embodiments of the present disclosure can be realized using any combination of dedicated components and / or programmable processors and / or other programmable devices. The various processes described herein can be implemented on the same processor or different processors in any combination. Where components are described as being configured to perform certain operations, such configuration can be accomplished, e.g., by designing electronic circuits to perform the operation, by programming programmableelectronic circuits (such as microprocessors) to perform the operation, or any combination thereof. Further, while the embodiments described above may make reference to specific hardware and software components, those skilled in the art will appreciate that different combinations of hardware and / or software components may also be used and that particular operations described as being implemented in hardware might also be implemented in software or vice versa.

[0145] Computer programs incorporating various features of the present disclosure may be encoded and stored on various computer readable storage media; suitable media include magnetic disk or tape, optical storage media such as compact disk (CD) or DVD (digital versatile disk), flash memory, and other non-transitory media. Computer readable media encoded with the program code may be packaged with a compatible electronic device, or the program code may be provided separately from electronic devices (e.g., via Internet download or as a separately packaged computer-readable storage medium).

[0146] Thus, although the disclosure has been described with respect to specific embodiments, it will be appreciated that the disclosure is intended to cover all modifications and equivalents within the scope of the following claims.

[0147] As utilized herein, the terms “approximately,” “about,” “substantially”, and similar terms are intended to have a broad meaning in harmony with the common and accepted usage by those of ordinary skill in the art to which the subject matter of this disclosure pertains. It should be understood by those of skill in the art who review this disclosure that these terms are intended to allow a description of certain features described and claimed without restricting the scope of these features to the precise numerical ranges provided. Accordingly, these terms should be interpreted as indicating that insubstantial or inconsequential modifications or alterations of the subject matter described and claimed are considered to be within the scope of the disclosure as recited in the appended claims. The terms “approximately,” “about,” “substantially,” and similar terms in reference to a number or value may generally be taken to include numbers or values that fall within a range of 1%, 5%, or 10% in either direction (greater than or less than) of the number or value unless otherwise stated or otherwise evident from the context (except where such number would be less than 0% or greater than 100% of a possible value).

[0148] It should be noted that the terms “exemplary,” “example,” “potential,” and variations thereof, as used herein to describe various embodiments, are intended to indicatethat such embodiments are possible examples, representations, or illustrations of possible embodiments (and such terms are not intended to connote that such embodiments are necessarily extraordinary or superlative examples).

[0149] The term “coupled” and variations thereof, as used herein, means the joining of two members directly or indirectly to one another. Such joining may be stationary (e.g., permanent or fixed) or moveable (e.g., removable or releasable). Such joining may be achieved with the two members coupled directly to each other, with the two members coupled to each other using a separate intervening member and any additional intermediate members coupled with one another, or with the two members coupled to each other using an intervening member that is integrally formed as a single unitary body with one of the two members. If “coupled” or variations thereof are modified by an additional term (e.g., directly coupled), the generic definition of “coupled” provided above is modified by the plain language meaning of the additional term (e.g., “directly coupled” means the joining of two members without any separate intervening member), resulting in a narrower definition than the generic definition of “coupled” provided above. Such coupling may be mechanical, electrical, or fluidic.

[0150] The term “or,” as used herein, is used in its inclusive sense (and not in its exclusive sense) so that when used to connect a list of elements, the term “or” means one, some, or all of the elements in the list. Conjunctive language such as the phrase “at least one of X, Y, and Z,” unless specifically stated otherwise, is understood to convey that an element may be either X, Y, Z; X and Y; X and Z; Y and Z; or X, Y, and Z (i.e., any combination of X, Y, and Z). Thus, such conjunctive language is not generally intended to imply that certain embodiments require at least one of X, at least one of Y, and at least one of Z to each be present, unless otherwise indicated.

[0151] References herein to the positions of elements (e.g., “top,” “bottom,” “above,” “below”) are merely used to describe the orientation of various elements in the Figures. It should be noted that the orientation of various elements may differ according to other exemplary embodiments, and that such variations are intended to be encompassed by the present disclosure.

[0152] The embodiments described herein have been described with reference to drawings. The drawings illustrate certain details of specific embodiments that implement the systems, methods and programs described herein. However, describing the embodimentswith drawings should not be construed as imposing on the disclosure any limitations that may be present in the drawings.101531 It is important to note that the construction and arrangement of the devices, assemblies, and steps as shown in the various exemplary embodiments is illustrative only. Additionally, any element disclosed in one embodiment may be incorporated or utilized with any other embodiment disclosed herein. Although only one example of an element from one embodiment that can be incorporated or utilized in another embodiment has been described above, it should be appreciated that other elements of the various embodiments may be incorporated or utilized with any of the other embodiments disclosed herein.

[0154] The foregoing description of embodiments has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from this disclosure. The embodiments were chosen and described in order to explain the principles of the disclosure and its practical application to enable one skilled in the art to utilize the various embodiments and with various modifications as are suited to the particular use contemplated. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the embodiments without departing from the scope of the present disclosure as expressed in the appended claims.

[0155] Unless defined otherwise, all technical and scientific terms used herein generally have the same meaning as commonly understood by one of ordinary skill in the art to which this technology belongs. As used in this specification and the appended claims, the singular forms “a”, “an” and “the” include plural referents unless the content clearly dictates otherwise. For example, reference to “a cell” includes a combination of two or more cells, and the like. Generally, the nomenclature used herein and the laboratory procedures in cell culture, molecular genetics, organic chemistry, analytical chemistry and nucleic acid chemistry and hybridization described below are those well-known and commonly employed in the art.

[0156] As used herein, the terms “individual”, “patient”, or “subject” are used interchangeably and refer to an individual organism, a vertebrate, a mammal, or a human. In a preferred embodiment, the individual, patient or subject is a human.EXAMPLES

[0157] Some sample embodiments are disclosed below, in order to represent illustrative embodiments, which one skilled in the art will understand may be further modified, combined, constrained, etc. according to the entirety of this disclosure.

[0158] Embodiment Al : A method comprising: acquiring imaging data from a scan of a brain of a subject, the scan lasting a minimum time; generating, based on the imaging data, a functional connectivity model with correlations between or among cortical vertices and subcortical voxels in the imaging data; generating, based on the functional connectivity model, a metric for one or more functional networks in the brain; providing the metric to a machine learning classifier to generate an output from the classifier, the output providing a prediction related to a psychological condition; and providing the prediction, or information generated from the prediction, to one or more users.10159] Embodiment A2: The method of Embodiment Al, wherein providing the prediction or the information comprises at least one of transmitting the prediction or the information to a user device, displaying the prediction or the information on a display device, and / or storing the prediction or the information in a non-transitory computer-readable memory of a networked computing system that is accessible, through the network, to one or more user devices of the one or more users.

[0160] Embodiment A3: The method of Embodiment Al or A2, wherein the scan is a functional magnetic resonance imaging (fMRI) scan of the brain of the subject.

[0161] Embodiment A4: The method of any of Embodiments Al - A3, wherein the minimum time is 10 minutes.

[0162] Embodiment A5: The method of any of Embodiments Al - A3, wherein the minimum time is 15 minutes.

[0163] Embodiment A6: The method of any of Embodiments Al - A5, wherein the scan is a muti-echo resting-state fMRI (ME-rsfMRI) scan of the brain of the subject.{0164] Embodiment A7: The method of any of Embodiments Al - A5, wherein the scan is a single-echo resting-state fMRI scan of the brain of the subject.

[0015] Embodiment A8: The method of Embodiment A7, wherein the minimum time is 45 minutes.

[0166] Embodiment A9: The method of any of Embodiments Al - A8, wherein the correlations of the functional connectivity model are correlations between a blood-oxygenlevel-dependent (BOLD) signal time series of cortical vertices and / or the subcortical voxels.

[0167] Embodiment A10: The method of any of Embodiments Al - A9, wherein the correlations of the functional connectivity model are correlations between cortical vertices, between subcortical voxels, and / or between cortical vertices and subcortical voxels.

[0168] Embodiment Al 1 : The method of any of Embodiments Al - A10, wherein the correlations of the functional connectivity model are correlations between two (pairs of) cortical vertices, or between two (pairs of) subcortical voxels, or between one cortical vertex and one subcortical voxel (cortical vertex and subcortical voxel pairs).

[0169] Embodiment A12: The method of any of Embodiments Al - Al l, wherein the machine learning classifier is a support vector machine (SVM) classifier.

[0170] Embodiment A13: The method of any of Embodiments Al - A12, wherein the machine learning classifier is trained using (a) diagnosis status as class labels and / or (b) functional brain network size as features.

[0171] Embodiment A14: The method of any of Embodiments Al - A13, wherein generating the functional connectivity model comprises setting correlations between nodes that are less than a threshold distance apart to zero, or setting correlations between nodes that are greater than the threshold distance apart to zero.

[0172] Embodiment A15: The method of any of Embodiments Al - A14, further comprising identifying the one or more functional networks in the imaging data using a community detection algorithm.

[0173] Embodiment A16: The method of any of Embodiments Al - A15, wherein generating the metric comprises at least one of determining a surface area for each vertex in a region of the brain, and / or determining a relative contribution of each functional network to a total cortical surface area.

[0174] Embodiment A17: The method of any of Embodiments Al - A16, wherein the metric is an occupancy measure for a network.

[0175] Embodiment A18: The method of any of Embodiments Al - A17, wherein the metric is indicative of an expansion of one or more networks of the subject and / or a contraction of one or more networks of the subject.

[0176] Embodiment A19: The method of any of Embodiments Al - A18, wherein the metric is indicative of an expansion of one or more salience networks of the subject.

[0177] Embodiment A20: The method of any of Embodiments Al - A19, wherein the metric is indicative of an expansion of a frontostriatal salience network of the subject.[01781 Embodiment A21 : The method of any of Embodiments Al - A20, wherein the metric is indicative of a contraction of at least one of a default mode network, a frontoparietal control network, and / or a cingulo-opercular control network.

[0179] Embodiment A22: The method of any of Embodiments Al - A21, wherein the metric is indicative of a contraction of a plurality of networks.

[0180] Embodiment A23: The method of any of Embodiments Al - A22, wherein the metric is indicative of a contraction of a plurality of adjacent networks.

[0181] Embodiment A24: The method of any of Embodiments Al - A23, wherein the metric is indicative of a contraction of three adjacent networks.

[0182] Embodiment A25: The method of any of Embodiments Al - A24, wherein the metric is indicative of a contraction of a default mode network, a frontoparietal control network, and a cingulo-opercular control network.

[0183] Embodiment A26: The method of any of Embodiments Al - A25, further comprising using the prediction or the information to (a) determine a treatment for the psychological condition, (b) predict a response to the treatment, (c) quantify symptoms, and / or (d) predict changes in symptoms.|0184| Embodiment A27: The method of any of Embodiments Al - A26, wherein the metric is a biomarker that is not sensitive to scanner-related artifacts, such that the method does not necessitate adjusting for scanner effects.

[0185] Embodiment A28: The method of any of Embodiments Al - K 1 wherein the metric is stable such that the metric is not dependent on a mood state.

[0186] Embodiment A29: The method of any of Embodiments Al - K 1 wherein the metric is dependent on a mood state.

[0187] Embodiment A30: The method of any of Embodiments Al - A29, comprising using the for tracking changes in symptoms and / or for predicting symptoms before they occur.

[0188] Embodiment A31 : The method of any of Embodiments Al - A30, comprising tracking current anhedonic symptoms based on changes in functional connectivity between or among nodes.

[0189] Embodiment A32: The method of any of Embodiments Al - A31, comprising tracking current anhedonic symptoms based on changes in functional connectivity between anterior cingulate and nucleus accumbens nodes of the salience network.

[0190] Embodiment A33: The method of any of Embodiments Al - A32, comprising predicting future anhedonic symptoms based on changes in functional connectivity between or among nodes.

[0191] Embodiment A34: The method of any of Embodiments Al - A33, comprising predicting future anhedonic symptoms based on changes in functional connectivity between anterior cingulate and nucleus accumbens nodes of the salience network.

[0192] Embodiment A35: The method of any of Embodiments Al - A34, comprising tracking anxiety symptoms based on changes in functional connectivity between or among nodes.

[0193] Embodiment A36: The method of any of Embodiments Al - A35, comprising tracking anxiety symptoms based on changes in functional connectivity between anterior insula and nucleus accumbens nodes of the salience network tracking.

[0194] Embodiment A > . The method of any of Embodiments Al - A36, comprising using metrics as outcome measures in treatments for psychological conditions such as anhedonia or anxiety.

[0195] Embodiment A38: The method of any of Embodiments Al - A37, comprising using metrics for guiding neuromodulation treatments to target one or more circuits in the brain.

[0196] Embodiment A39: The method of any of Embodiments Al - A38, comprising using metrics to predict future symptoms, so as to, for example, intervene earlier in treating patients.

[0197] Embodiment A40: The method of any of Embodiments Al - A39, comprising using changes in networks (e.g., stable salience network expansion) to predict risk for becoming depressed in a subject who has not previously had depression.

[0198] Embodiment Bl : A computing system comprising one or more processors, the computing system configured to perform the method of any of Embodiments Al - A40.

[0199] Embodiment B2: The computing system of Embodiment Bl, configured to communicate with a medical imager to acquire the imaging data.

[0200] Embodiment B3 : The computing system of Embodiment B2, wherein the medical imager is or comprises an MRI scanner.[0201 [ Embodiment B4: The computing system of Embodiment B2 or B3, wherein the medical imager is or comprises an fMRI scanner.|0202[ Embodiment Cl: A system comprising a medical imager and one or more processors configured to perform the method of any of Embodiments Al - A40.

[0203] Embodiment C2: The system of Embodiment Cl, wherein the medical imager is or comprises an MRI scanner.

[0204] Embodiment C3: The system of Embodiment Cl or C2, wherein the medical imager is or comprises an fMRI scanner.EQUIVALENTS

[0205] The present technology is not to be limited in terms of the particular embodiments described in this application, which are intended as single illustrations of individual aspects of the present technology. Many modifications and variations of this present technology can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. Functionally equivalent methods and apparatuses within the scope of the present technology, in addition to those enumerated herein, will be apparent to those skilled in the art from the foregoing descriptions. Such modifications and variations are intended to fall within the scope of the present technology. It is to be understood that this present technology is not limited to particular methods, reagents, compounds compositions or biological systems, which can, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting.

[0206] In addition, where features or aspects of the disclosure are described in terms of Markush groups, those skilled in the art will recognize that the disclosure is also thereby described in terms of any individual member or subgroup of members of the Markush group.

[0207] As will be understood by one skilled in the art, for any and all purposes, particularly in terms of providing a written description, all ranges disclosed herein also encompass any and all possible subranges and combinations of subranges thereof. Any listed range can be easily recognized as sufficiently describing and enabling the same range being broken down into at least equal halves, thirds, quarters, fifths, tenths, etc. As a non-limiting example, each range discussed herein can be readily broken down into a lower third, middle third and upper third, etc. As will also be understood by one skilled in the art all language such as “up to,” “at least,” “greater than,” “less than,” and the like, include the number recited and refer to ranges which can be subsequently broken down into subranges as discussed above. Finally, as will be understood by one skilled in the art, a range includes each individual member. Thus, for example, a group having 1-3 cells refers to groups having 1, 2, or 3 cells. Similarly, a group having 1-5 cells refers to groups having 1, 2, 3, 4, or 5 cells, and so forth.

[0208] All patents, patent applications, provisional applications, and publications referred to or cited herein are incorporated by reference in their entirety, including all figures and tables, to the extent they are not inconsistent with the explicit teachings of this specification.

Claims

CLAIMS:

1. A method comprising: acquiring imaging data from a scan of a brain of a subject, the scan lasting a minimum time; generating, based on the imaging data, a functional connectivity model with correlations between or among cortical vertices and / or subcortical voxels in the imaging data; generating, based on the functional connectivity model, a metric for one or more functional networks in the brain; providing the metric to a machine learning classifier to generate an output from the classifier, the output providing a prediction related to a psychological condition; and providing the prediction, or information generated from the prediction, to one or more users, wherein providing the prediction or the information comprises at least one of transmitting the prediction or the information to a user device, displaying the prediction or the information on a display device, and / or storing the prediction or the information in a non- transitory computer-readable memory of a networked computing system that is accessible, through the network, to one or more user devices of the one or more users.

2. The method of claim 1, wherein the scan is a functional magnetic resonance imaging (fMRI) scan of the brain of the subject, and wherein the minimum time is 10 minutes.

3. The method of claim 1, wherein the scan is a muti-echo resting-state fMRI (ME- rsfMRI) scan of the brain of the subject.

4. The method of claim 1, wherein the correlations of the functional connectivity model are correlations between a blood-oxygen-level-dependent (BOLD) signal time series of the cortical vertices and / or the subcortical voxels.

5. The method of claim 1, wherein the machine learning classifier is a support vector machine (SVM) classifier trained using (a) diagnosis status as class labels and (b) functional brain network size as features.

6. The method of claim 1, wherein generating the functional connectivity model comprises setting correlations between nodes that are less than a threshold distance apart to zero, or setting correlations between nodes that are greater than the threshold distance apart to zero.

7. The method of claim 1, further comprising identifying the one or more functional networks in the imaging data using a community detection algorithm.

8. The method of claim 1, wherein generating the metric comprises at least one of determining a surface area for each vertex in a region of the brain, or determining a relative contribution of each functional network to a total cortical surface area.

9. The method of claim 1, wherein the metric is an occupancy measure for a network.

10. The method of claim 1, wherein the metric is indicative of an expansion of a salience network of the subject or a contraction of a plurality of adjacent networks of the subject.

11. The method of claim 1, further comprising using the prediction or the information to (a) determine a treatment for the psychological condition, (b) predict a response to the treatment, (c) quantify symptoms, and / or (d) predict changes in symptoms.

12. The method of claim 1, wherein the metric is a biomarker that is not sensitive to scanner-related artifacts, such that the method does not necessitate adjusting for scanner effects.

13. The method of claim 1, wherein the metric is stable such that the metric is not dependent on a mood state.

14. A system comprising one or more processors, the system configured to perform the method of claim 1.

15. The system of claim 1, wherein the system comprises an fMRI scanner to acquire the imaging data, or the system comprises a computing system configured to communicate with an fMRI scanner to acquire the imaging data.

Images