Multimodal phenotypic classification of generalized anxiety and panic using structural MRI data and psychosocial factors: Machine learning results from the German National Cohort (NAKO) study

Multimodal phenotypic classification of generalized anxiety and panic using structural MRI data and psychosocial factors: Machine learning results from the German National Cohort (NAKO) study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Multimodal phenotypic classification of generalized anxiety and panic using structural MRI data and psychosocial factors: Machine learning results from the German National Cohort (NAKO) study Julian Gutzeit, Martin Weiß, Tierney Kuhn, Johanna Klinger-König, and 29 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6627834/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Anxiety disorders (ANX) are common and impairing mental health conditions. This study aimed to classify self-reported symptoms of generalized anxiety disorder (GAD) and panic attacks as two psychopathological manifestations of ANX by applying machine learning to a cross-sectional dataset of 26,378 adults from the German National Cohort Study (NAKO). We first explored linear relationships between preselected neuroimaging correlates in MRI scans and anxiety phenotypes. Overall, sex-stratified correlation coefficients - while partly highly significant - were extremely low with r ≤ .04 for panic attacks and r ≤ .06 for GAD symptoms after correction for confounding variables like childhood trauma and depression. We then examined the combined classifying value of whole-brain imaging data of 246 ROIs in addition to psychosocial variables such as self-reported depression symptoms, stress, and childhood trauma, using four machine learning algorithms (support-vector machines with linear and radial kernels, elastic-net regression, and random forest). Neuroimaging data, particularly gray-matter volumes in regions such as the amygdala and superior parietal lobule, contributed to classification, but performance was substantially better when psychosocial variables were added. For both GAD symptoms and panic attacks, depression, stress and childhood trauma were the clearest indicators the classification would show the condition was present. Random forest models based on psychosocial variables alone achieved the highest discrimination performance for GAD symptoms (area under the receiver operating characteristic curve, AUROC = 0.973) and panic attacks (AUROC = 0.933). Combining neuroimaging and psychosocial variables in elastic-net regressions further improved specificity. These results support multimodal approaches to diagnose and investigate ANX that integrate structural brain abnormalities and psychosocial measures to capture the complexity of GAD and panic attacks, enabling the creation of individual risk profiles based on multiple biomarkers. These profiles may guide tailored therapeutic and preventive interventions. Health sciences/Diseases/Psychiatric disorders Biological sciences/Psychology Health sciences/Biomarkers/Diagnostic markers Figures Figure 1 Figure 2 Introduction Anxiety disorders (ANX) are common and seriously impairing disorders, with an estimated lifetime prevalence up to 20% (Penninx et al., 2021; Wittchen et al., 2011). Currently, diagnosis of these disorders is based primarily on clinicians’ assessment (Linden, 2012). This diagnostic challenge underscores the need for objective biomarkers that can complement clinical evaluations and improve the precision of ANX diagnosis. Identifying reliable biomarkers for ANX could potentially complement clinical evaluations and treatment, by identifying patient groups early in disease progression and using the underlying biological causes of their symptoms to inform the choice of treatment (Boeke et al., 2020; Chavanne et al., 2023; Linden, 2012). Generalized anxiety disorder (GAD) occurs with an estimated prevalence of 4-6% and is characterized by excessive, persistent and uncontrollable anxiety and worrying that is associated with nervousness, feelings of threatening uncertainty, and somatic complaints like muscular tensions and physiological hyperarousal (Ruscio et al., 2017; Wittchen et al., 2011). Panic attacks (PA) represent a core symptom of panic disorder (PD) according to diagnostic criteria; however, they occur frequently in all ANX, and are strongly linked to general psychopathology as a separate dimension across mental conditions (Asselmann et al., 2014). PA are defined as sudden and brief episodes of extreme anxiety as well as somatic stress symptoms, which in the case of mental disorders are inappropriate or unrealistically exaggerated compared to the target situation. Affected individuals tend to develop a fear of reexperiencing a PA that is associated with avoidance behavior and further negative behavioral changes, high distress and consequently individual burden (Craske et al., 2017; de Jonge et al., 2016). Prospective data suggests that PA constitutes a risk factor not just for the future development of any ANX, but also mood and substance use disorders (Goodwin et al., 2004). Individuals experiencing PA are at greater risk for increased persistence of mental disorders and impaired functioning, which underscores the importance of preventive treatment and early diagnosis of PA to improve long-term outcomes (Batelaan et al., 2012). Within the complex phenotypic composition of ANX, PA represent characteristic and well-defined symptoms which can be interrogated using validated scales and attributed to biological anxiety circuits (Guan & Cao, 2024). Both GAD and PA are highly comorbid with other psychiatric conditions, most notably with depression (Penninx et al., 2021). This high comorbidity can be partly explained by shared neurobiological mechanisms, e.g. by overlapping environmental and genetic risk factors, as demonstrated in recent cross-disorder genome-wide association studies for pathological anxiety and depression (Kalisch et al., 2024; Strom et al., 2024). In addition, childhood adversity is one of the environmental factors potentially influencing brain morphology and significantly increasing the prevalence of ANX and depression (Grummitt et al., 2024; Teicher et al., 2016). Therefore, it is crucial to incorporate these factors into our models to capture the complexity of anxiety-related phenotypes and improve the accuracy of classification analyses. Several previous studies have investigated neuroimaging data as a promising candidate for biomarker identification. Lower gray matter volumes in bilateral orbitofrontal cortex and ventrolateral prefrontal cortex have been found to correlate with the General Distress dimension of the Tri-level Model (representing transdiagnostic depression and anxiety symptoms, Cichocki et al., 2024). Furthermore, higher gray matter volume in the amygdala has repeatedly been associated with GAD (Etkin et al., 2009; Makovac et al., 2015; Schienle et al., 2011). Functional imaging studies also consistently provide evidence that patients suffering from ANX display increased reactivity in the amygdala in response to negative emotional stimuli, along with insufficient prefrontal control (Abi-Dargham et al., 2023). Unfortunately, none of the biomarkers identified to date has demonstrated a sufficiently reliable predictive value to be used clinically (Abi-Dargham et al., 2023; Boeke et al., 2020). This may be partly due to limitations in previous research, including small sample sizes, heterogeneous analytical and clinical approaches, and relatively simple statistical methods that may not detect subtle associations in the large, high-dimensional data produced by neuroimaging. Machine learning algorithms, however, are particularly effective in analyzing high-dimensional data, (Bzdok & Meyer-Lindenberg, 2018; Janssen et al., 2018), such as neuroimaging data, to identify complex patterns associated with conditions like GAD and PA. Classifiers trained on the gray matter volume of anxiety-associated regions of interest (ROIs) in adolescent subjects achieved moderate predictive [1] value for anxiety-disorder diagnosis in early adulthood (Chavanne et al., 2023). A similar approach achieved robust (albeit modest) performance when classifying PD vs. healthy controls based on subcortical volumes and cortical thickness and surface area (Bruin et al., 2024). Previous MRI-based machine learning studies of adult and adolescent anxiety have only used moderately large samples with imaging data from several hundred individuals (Boeke et al., 2020; Bruin et al., 2024; Chavanne et al., 2023). Increasing the sample size by an order of magnitude can substantially increase the robustness of the findings by reducing the risk of overfitting (Janssen et al., 2018), which may have contributed to the failure of certain models to replicate their predictive validity when evaluated on different data sets not used during training (Boeke et al., 2020). Therefore, the current paper builds on previous research by applying machine learning techniques to classify GAD symptoms and PA, in a very large dataset that includes neuroimaging data from 26,378 adults taken from the German National Cohort Study (NAKO). To highlight the advantages of data-driven machine learning methods over traditional analyses, we first conducted conventional correlational analyses on theory-driven, preselected neuroimaging variables, accounting for known confounders. Subsequently, we performed multiple machine learning analyses (support vector machines with either a radial or linear kernel, random forests, elastic net regression), using the full set of whole-brain gray matter volume data without any regional preselection. This approach enables a more comprehensive, unbiased exploration of the brain, allowing the model to identify potential patterns that might be overlooked in region-specific analyses. Methods Dataset and study population Demographic, psychometric and brain imaging data were taken from NAKO. This population-based prospective cohort study started in 2014 with the goal of investigating risk factors for a wide range of physical and mental chronic conditions, including depression, stress and anxiety symptoms. Baseline assessment was conducted between 2014 and 2019, with the goal of investigating risk factors for a wide range of physical and mental chronic conditions, including depression, stress and anxiety symptoms. NAKO collected biomedical and questionnaire data from 205,415 persons living in Germany aged 19–74 who were chosen at random from compulsory registries of residents in 16 regions across Germany. This study analyzed the subsample of 30,927 participants who also later completed whole-body 3T magnetic resonance imaging (Peters et al., 2022 ). All participants gave written informed consent, and the data transfer was approved by the Use and Access Committee of NAKO. All NAKO study documents have been approved by all responsible local ethical committees and are revised regularly and adapted as needed. The two outcomes of interest in this study were clinically meaningful GAD symptoms and lifetime PAs. GAD symptoms were assessed using the established GAD-7 scale, which consists of seven items (yielding a maximum sum score of 21) to measure symptom load over the past four weeks. Clinically meaningful GAD symptoms were defined as a GAD-7 score of ≥ 10 - indicating at least moderate anxiety - with a sensitivity and specificity of 89% and 82%, respectively, in detecting GAD (Löwe et al., 2008). Lifetime PA was defined as having experienced one or more panic attacks in the past four weeks in addition to at least one previous lifetime panic attack (Löwe et al., 2003). Because only the first part of the PHQ-Panic scale was administered, we could determine the presence of current and prior PAs, the mode of occurrence, and associated disability, but not a full diagnosis of PD. The demographic and psychometric predictors in our study included eight psychosocial variables: age, sex, number of lifetime cigarettes smoked, symptoms of GAD (GAD-7 score; only for classifying PA), PA (PHQ-Panic; only for classifying GAD symptoms), depression (current PHQ-9 score; Kroenke et al., 2001 ), stress (current PHQ-Stress score; Spitzer et al., 1999 ), and childhood trauma (childhood trauma screener, CT-S; Bernstein et al., 2003 ). All psychosocial variables were collected by self-administered touchscreen questionnaires during an in-person baseline examination in the study centers (Peters et al., 2022 ). The detailed description of psychometric and emotional scales within the NAKO is available elsewhere (for GAD-7, panic and stress: Erhardt et al., 2023 ; for childhood trauma: Klinger-König et al., 2023 ; and for depression: Streit et al., 2023 ). After excluding participants with missing values for any of the predictors or outcomes, 26,378 participants were included in the study. MRI acquisition and preprocessing Structural MRI data for all participants was obtained using 3T scanners (Magnetom Skyra; Siemens Healthcare, Erlangen, Germany) across five NAKO study centers (Essen, Neubrandenburg, Berlin, Augsburg, and Heidelberg/Mannheim). Images were acquired using a T1-weighted 3D MPRAGE sequence (1.0 × 1.0 × 1.0 mm (isotropic) voxel; sagittal orientation; repetition time msec/echo time msec/inversion time msec, 2300/2.98/900; 9° flip angle; Bamberg et al., 2015). The T1 images were segmented, normalized, and smoothed. An automated quality-control pipeline was used to assess each image’s sharpness, global and local signal-to-noise ratio, maximum and average estimates for structured image noise and Nyquist ghosting levels, and geometric ratio between foreground and background. Subsequently, board-certified radiologists performed a visual rating using a three-point Likert scale that considered anatomical coverage, minimum differentiable structures, and the presence of artifacts, and excluded any images rated as ‘Poor’ (Schuppert et al., 2023 ). The gray matter volumes of the 246 brain areas defined by the Julich-Brain Cytoarchitectonic Atlas ( https://atlases.ebrains.eu/ ; Amunts et al., 2020 ) using the Software Computation Anatomy Toolbox (CAT12v8; https://neuro-jena.github.io/cat/ ) were extracted from those T1-weighted images that passed quality control. All resulted 246 ROIs, encompassing volumes of both cortical areas and subcortical structures including the limbic system, surface areas and mean cortical thickness values were included as variables in the neuroimaging and combined variable sets described below. Correlation analyses To investigate simple linear relationships between specific neuroimaging variables and GAD as well as PA, we first computed Pearson correlations between a pre-selected set of 93 neuroimaging variables and clinically relevant GAD symptoms and PA. This set of neuroimaging variables included global brain metrics (e.g., total gray matter and white matter volumes), regional gray matter volumes (e.g., insula, amygdala, hippocampus, and cingulate cortex), surface areas, subcortical structures, and mean cortical thickness in anxiety-related brain regions (Craske et al., 2017 ; Harrewijn et al., 2021a ; Pessoa, 2023 ). Additionally, we examined network-specific gray matter volumes, such as those associated with the Default Mode and Salience Networks, as well as limbic regions implicated in emotional regulation (Alves et al., 2019 ; Catani et al., 2013 ). A full list of all variables alongside their correlations can be found in the Extended Data Tables S2 and S3. In order to depict the putative influence of age and highly overlapping symptoms on the target outcome of GAD and PA, we used a step by step correction approach, using sequential linear regressions to regress out the impact of confounders on the neuroimaging variables (Snoek et al., 2019 ). For the correlations with GAD symptoms, the analyses were controlled for age, depressive symptoms (PHQ-9 score), PA and childhood trauma, using linear confounder regression sequentially. For the correlations with PA, the analyses were controlled for age, depressive symptoms (PHQ-9 score), clinically relevant GAD symptoms (GAD-7 ≥ 10) and childhood trauma. All correlations were tested for significance, with Bonferroni corrections applied to control for alpha inflation. To account for known sex differences in the prevalence and neurobiological correlates of GAD and PA (Jalnapurkar et al., 2018 ), all correlation analyses were stratified by sex. Machine-learning classification The classification analysis was conducted in R, version 4.4.3 (Core R Team, 2021 ) using four models from the machine learning package caret , version 6.0–94 (Kuhn, 2008 ): support vector machines with either a radial (SVM-R) or linear kernel (SVM-L), random forests (RF), and elastic net regression (ELNET). Each model was trained in two distinct analyses to make binary predictions regarding: (a) the presence versus absence of clinically relevant GAD symptoms (defined as GAD-7 ≥ 10, labelled as GAD symptoms), and (b) the presence versus absence of combined current and lifetime PA (according to the first to items of the PHQ-Panic scale). These analyses were conducted using three different variable sets: 246 neuroimaging variables (N), 8 psychosocial variables (P; including 7 psychosocial variables sex, age, number of lifetime cigarettes, depression, stress, childhood trauma plus PA for classifying clinically relevant GAD symptoms or the GAD-7 score for classifying PA), and a combined set of 254 neuroimaging and psychosocial variables (P + N; for a similar approach, see Chavanne et al., 2023 ). The participants were randomly divided into a training set with n = 21,102 subjects and a test set with n = 5,276 subjects. A linear confound regression was used to regress out the impact of sex, age, total intracranial volume, lifetime cigarettes smoked, childhood trauma, and scanner site on gray matter volume in the training dataset (Snoek et al., 2019 ). We computed multiple linear regressions with these potential confounding variables for each brain area volume. Only the residuals of this regression analyses were included in the machine-learning models. To prevent data leakage – that is, the unintended use of test data information inflating model performance – the regression weights estimated from the training dataset were also applied to control for confounds in the test dataset (More et al., 2021 ). All models were trained using 5-fold cross-validation, using the area under the receiver operating characteristic curve (AUROC) to score model performance. Hyperparameters for each model were tuned using grid search cross-validation with a tune length of 10 for each hyperparameter. For SVMs, we adjusted the regularization parameter C, as well as the scaling parameter sigma when using a radial kernel. Random Forest tuning focused on minimum node size, number of trees, and splitting rule. For ELNET, we optimized the mixing parameter alpha and regularization parameter lambda. All other hyperparameters were left at the default values provided by caret. The best-fitting model was then retrained using the selected hyperparameter configuration on the full training dataset and subsequently evaluated on the test set. Consistent with population epidemiology, only a minority of participants had GAD-7 score ≥ 10 or PA (see Results section). To prevent the imbalance of the data from skewing the classification, a subset of control participants without the target diagnosis was chosen through random undersampling (Ganganwar, 2012 ). This balancing of class sizes was performed independently in each fold of cross-validation, randomly selecting a new subsample of the majority class for each fold. This approach ensured that nearly all participants contributed to the model training, despite the undersampling of the majority class (for a similar approach, see Bruin et al., 2024 ). All variables were standardized to z -scores. After training, the model with the best performance (when using optimal hyperparameters) was identified for each combination of prediction target and variable set. We computed several performance measures on the confusion matrices of the best models for each data set (N, P, N + P). Additionally, we assessed variable importance in the winning N + P models by computing the AUROC for each predictor. This approach tests each predictor individually and computes the AUROC based on predictions made using only that predictor and the model. This method has the advantage of providing the predictive importance of each single predictor for each model and allows for comparisons across multiple classification models (Kuhn, 2012 ). Higher values indicate stronger associations with the outcome and can be interpreted as indicating effect sizes. To further assess the direction of these associations, we computed simpler logistic regression models using the ten most important predictors. In contrast to separate Pearson correlations between individual predictors and various outcomes, these analyses test the effect of all ten predictors simultaneously. Results Of all participants, 4.49% (n = 1,321) reported a GAD-7 score ≥ 10, indicating clinically meaningful anxiety symptoms. Additionally, 4.01% (n = 1,182) reported experiencing PA. For an overview of all participant characteristics, see Table 1 . Table 1 Demographic characteristics of analyzed study participants with valid data (n = 26,378) Variable GAD symptoms Panic attacks Total no yes no yes Participants n 25,217 1,161 25,376 1,002 26,378 Sex Men 14,530 (57.6%) 502 (43.2%) 14,619 (57.6%) 413 (41.2%) 15,032 (57.0%) Women 10,687 (42.4%) 659 (56.8%) 10,757 (42.4%) 589 (58.8%) 11,346 (43.0%) Age mean (SD) 47.5 (12.3) 45.8 (11.8) 47.5 (12.2) 46.8 (12.4) 47.4 (12.2) 18–29 2,855 (11.3%) 148 (12.7%) 2,877 (11.3%) 126 (12.6%) 3,003 (11.4%) 30–39 3,163 (12.5%) 179 (15.4%) 3,211 (12.7%) 131 (13.1%) 3,342 (12.7%) 40–49 7,672 (30.4%) 353 (30.4%) 7,728 (30.5%) 297 (29.6%) 8,025 (30.4%) 50–59 6,864 (27.2%) 324 (27.9%) 6,910 (27.2%) 278 (27.7%) 7,188 (27.2%) 60–74 4,663 (18.5%) 157 (13.5%) 4,650 (18.3%) 170 (17.0%) 4,820 (18.3%) GAD-7 mean (SD) 2.6 (2.2) 12.7 (2.7) 2.8 (2.8) 8.6 (4.6) 3.0 (3.1) GAD-7 ≥ 10 healthy 25,217 (100.0%) 0 (0%) 24,580 (96.9%) 637 (63.6%) 25,217 (95.6%) anxious 0 (0%) 1,161 (100.0%) 796 (3.1%) 365 (36.4%) 1,161 (4.4%) Lifetime Panic Attacks no 24,580 (97.5%) 796 (68.6%) 25376 (100.0%) 0 (0%) 25,376 (96.2%) yes 637 (2.5%) 365 (31.4%) 0 (0%) 1,002 (100.0%) 1,002 (3.8%) Lifetime Cigarettes mean (SD) 6.0 (192.2) 10.6 (260.8) 6.0 (191.6) 11.6 (280.7) 6.2 (195.7) Depressive Symptoms (PHQ-9) mean (SD) 3.3 (2.9) 12.8 (5.2) 3.5 (3.3) 9.4 (5.8) 3.7 (3.6) Stress mean (SD) 3.1 (2.7) 9.2 (3.6) 3.2 (2.8) 7.5 (4.1) 3.4 (3.0) Childhood Trauma Sum Score (CT-S) mean (SD) 7.0 (2.4) 8.8 (3.7) 7.1 (2.5) 8.5 (3.5) 7.1 (2.5) Bonferroni-corrected Pearson correlations between pre-selected neuroimaging variables and GAD symptoms and PA stratified by sex revealed some significant but very weak associations ( p -values reported as p adj ). Overall, the sex-stratified correlation values of any of the neuroimaging variables with target anxiety phenotypes were extremely low, with significant correlations of .032 ≤ |r| ≤ .098 for GAD symptoms after correction for age, depressive symptoms, PA, and childhood trauma. These associations were predominantly present in the female group (see Table 2 ). In females, categorical GAD symptoms had the highest significant correlation with total grey matter and cerebral white matter volumes, cortical gray matter volume and hemispheric surface area, as well as, more specifically, ventral diencephalon volumes (after correction for depression). Furthermore, GAD symptoms were most consistently correlated with hemispheric surface area bilaterally for both sexes. Table 2 Top correlations between clinically relevant GAD symptoms (GAD-7 ≥10) and various neuroimaging variables, corrected for age, depressive symptoms (PHQ-9), panic attacks (PHQ-Panic), and childhood trauma (dimensional CT-S), stratified by sex. Correlation with GAD-7 (cutoff) Neuroimaging variables female male r p_adj r p_adj Total volume Total cerebral white matter volume 0.051 5.6E-06 0.034 0.0035 Total gray matter volume 0.061 5.9E-09 0.039 0.00018 Left hemispheric cerebral white matter volume 0.051 6.7E-06 0.033 0.0043 Subcortical brain region volume Left ventral diencephalon 0.052 2.6E-06 0.026 ns Right ventral diencephalon 0.048 2.2E-05 0.026 ns Brainstem 0.041 9.5E-04 0.021 ns Right amygdala 0.047 4.7E-05 0.026 ns Left amygdala 0.035 0.021 0.032 0.0063 Right thalamus 0.044 3.3E-04 0.032 0.010 Left thalamus 0.040 0.002 0.025 ns Left hippocampus 0.042 5.7E-04 0.028 0.045 Right hippocampus 0.029 ns 0.033 0.0040 Right accumbens area 0.015 ns 0.032 0.0082 Left accumbens area 0.018 ns 0.029 0.041 Cortical/subcortical gray matter volume Right cortical gray matter volume 0.057 8.8E-08 0.035 0.00142 Left cortical gray matter volume 0.058 6.1E-08 0.036 0.00076 Subcortical gray matter volume 0.051 4.5E-06 0.038 0.00023 Left medial orbitofrontal cortex 0.049 1.5E-05 0.032 0.0083 Left rostral anterior cingulate cortex 0.042 6.1E-04 0.019 ns Right rostral anterior cingulate cortex 0.032 ns 0.020 ns Left insula 0.033 0.049 0.034 0.0033 Right insula 0.039 0.004 0.043 1.6E-05 Left isthmus cingulate cortex 0.032 ns 0.016 ns Right isthmus cingulate cortex 0.035 0.017 0.021 ns Surface area Left hemisphere 0.050 8.9E-06 0.041 5.9E-05 Right hemisphere 0.050 1.1E-05 0.040 7.6E-05 Left insula 0.029 ns 0.039 0.0002 Right insula 0.029 ns 0.039 0.0002 Left isthmus cingulate cortex 0.028 ns 0.032 0.0081 Right isthmus cingulate cortex 0.028 ns 0.032 0.0084 Left caudal anterior cingulate cortex 0.007 ns 0.029 0.043 Right caudal anterior cingulate cortex 0.006 ns 0.029 0.043 Left posterior cingulate cortex 0.014 ns 0.031 0.014 Right posterior cingulate cortex 0.014 ns 0.031 0.015 Note. All significant correlations between clinically relevant GAD symptoms and neuroimaging variables with increasing numbers of controlled confounds can be found in the Data Table S1. Significant correlations between panic attacks and several neuroimaging variables – adjusted for different numbers of regressed-out confounders – are reported in the Extended Data Table S1. PA were statistically significantly correlated with neuroimaging variables only in the female group after controlling for age and depression. Volumes of the total grey matter, cerebral white matter, ventral diencephalon, thalamus as well as total cortical, subcortical and the right isthmus cingulate cortex volume were significantly correlated with PA. Interestingly, after correction for childhood trauma score, only the correlations with the diencephalon volume and total gray matter volume remained significant. All significant correlations were very low with .032 ≤ |r| ≤ .040 and p adj. < .047. In males, none of the neuroimaging variables were significantly correlated with PA after adjustment. In summary, the correlation analyses revealed statistically significant correlations of GAD symptoms and PA with a number of different neuroimaging variables. However, the correlation coefficients are extremely small and hard to interpret regarding their importance for predicting GAD and PA. For all correlations, see Extended Data, tables S2 and S3. In the next step, we used machine learning to identify the brain structures and psychosocial variables that are most important for predicting these clinically relevant outcomes. As can be seen in Table 3 , the random forest classifier outperformed other P-models (i.e., models using only psychosocial variables as input data) for both GAD symptoms (AUROC = 0.976) and PA (AUROC = 0.933) in the test dataset. The random forest classifier performed best on the datasets containing only neuroimaging data (N-models) but still showed relatively poor performances for classifying GAD symptoms (AUROC = 0.557) and PA (AUROC = 0.557). ELNET showed the best performance using the combination of neuroimaging data and psychosocial variables (N + P) to classify GAD symptoms (AUROC = 0.961) and PA (AUROC = 0.876), but did not outperform random forest P-models. AUROCs for the winning models for each dataset (P, N, P + N) are depicted in Fig. 1 . Table 3 Area under receiver operating characteristic curve (AUROC) values for each model and each algorithm. GAD Symptoms Panic Attacks model SVM_L SVM_R RF ELNET SVM_L SVM_R RF ELNET N 0.517 0.553 0.557 0.528 0.546 0.520 0.557 0.522 P 0.962 0.950 0.973 0.962 0.875 0.866 0.933 0.877 P + N 0.949 0.954 0.957 0.961 0.852 0.831 0.874 0.876 Note. N: neuroimaging imaging variables; P: psychosocial and psychometric variables; SVM_L = linear support vector machine; SVM_R = radial support vector machine, RF = random forest, ELNET = elastic net regression. The algorithms resulting in the highest AUROC for each model are printed in italic type and the highest value for each outcome additionally in bold type. Table 4 Confusion matrices for the models with the highest AUROC. GAD Symptoms Panic Attacks Metric N (RF) P (RF) P + N (ELNET) N (RF) P (RF) P + N (ELNET) Accuracy 0.499 0.896 0.905 0.544 0.802 0.846 Kappa 0.011 0.383 0.398 0.011 0.214 0.221 AccuracyLower 0.485 0.887 0.897 0.530 0.791 0.836 AccuracyUpper 0.512 0.904 0.913 0.557 0.813 0.856 AccuracyNull 0.958 0.958 0.958 0.961 0.961 0.961 AccuracyPValue 1.000 1.000 1.000 1.000 1.000 1.000 McnemarPValue 0.000 0.000 0.000 0.000 0.000 0.000 Sensitivity 0.575 0.910 0.873 0.522 0.918 0.729 Specificity 0.496 0.895 0.907 0.544 0.797 0.851 Pos Pred Value 0.047 0.275 0.290 0.045 0.156 0.166 Neg Pred Value 0.964 0.996 0.994 0.965 0.996 0.987 Precision 0.047 0.275 0.290 0.045 0.156 0.166 Recall 0.575 0.910 0.873 0.522 0.918 0.729 F1 0.088 0.422 0.436 0.082 0.267 0.271 Prevalence 0.042 0.042 0.042 0.039 0.039 0.039 Detection Rate 0.024 0.038 0.037 0.020 0.036 0.029 Detection Prevalence 0.507 0.139 0.126 0.458 0.231 0.172 Balanced Accuracy 0.535 0.902 0.890 0.533 0.858 0.790 AUROC 0.557 0.973 0.959 0.557 0.933 0.876 Note. RF = random forest, ELNET = elastic net regression. To investigate which variables had the highest predictive importance for the outcome variables, we computed the AUROC importance for the P + N models (Kuhn, 2012 ). Variables with higher AUROC importance scores are considered more predictive. We depicted the top 10 variables (4% of all variables; for similar approach, see Weiß et al., 2024 ) separately for GAD (Fig. 2 A) and PA (Fig. 2 B). The importance scores for all variables of all winning models are provided in the Extended Data as follows: the winning P models are detailed in S4 (GAD) and S5 (PA); the winning N models in S6 (GAD) and S7 (PA); and the winning P + N models in S8 (GAD) and S9 (PA). [FIGURE 1 here] In addition to AUROC, we computed performance metrics derived from the confusion matrices for the top-performing models (P, N, P + N), as summarized in Table 4 . Interestingly, while the elastic net P + N models did not achieve higher AUROCs compared to the random forest P models, both P + N models for GAD and PA exhibited higher unbalanced accuracy and positive predictive value (i.e., the likelihood that a positive classification is a true positive, also referred to as precision) as well as higher specificity values compared to random forest P models. [FIGURE 2 here] The most important psychosocial variables (P) for classifying individuals with clinically relevant symptoms of GAD were depressive symptoms (PHQ-9 score), psychological distress (dimensional PHQ-Stress), childhood trauma (dimensional CT-S), PA, female sex, and age (Fig. 2 A). The most important neuroimaging (N) variables were gray matter volume in the regions of the left internal intermediate fiber masses (IF) of the amygdala, the left area OP8 in the frontal operculum, the right superficial fiber masses (SF) of the amygdala, and the left area 7M in the superior parietal lobule. For the classification of panic attacks, the most important psychosocial variables (P) were generalized anxiety disorder symptoms (GAD-7 score), depressive symptoms (PHQ-9 score), psychological distress (dimensional PHQ-Stress), childhood trauma (dimensional CT-S), female sex, and lifetime cigarettes (Fig. 2 B). The most important neuroimaging (N) variables were gray matter volume in the regions of the left internal intermediate fiber masses of the amygdala, the left area OP8 in the frontal operculum, the right internal medial fiber masses of the amygdala, and the right area 7A in the superior parietal lobule. To evaluate the direction of these associations, the results of the simple logistic regression models for outcomes and the predictor variables are depicted in Table 5 . Table 5 Coefficients and Odds-Ratios of the logistic regression on GAD symptoms and panic attacks with top 10 variables of the winning P + N model. GAD symptoms Independent variable Estimate SE z p OR [95%CI] Depression (PHQ) 1.34 0.04 33.44 < .001 3.84 [3.55 , 4.15] Stress 0.64 0.04 16.03 < .001 1.90 [1.75, 2.05] Childhood Trauma 0.01 0.03 0.24 0.813 1.01 [0.94, 1.08] Panic Attacks 1.08 0.11 9.71 < .001 2.93 [2.36, 3.64] Female 0.02 0.05 0.40 0.686 1.02 [0.93, 1.12] Left IF (Amygdala) 0.07 0.04 1.75 0.080 1.08 [0.99, 1.17] Age -0.06 0.05 -1.25 0.212 0.94 [0.86, 1.04] Left Area OP8 (Frontal Operculum) -0.07 0.05 -1.22 0.221 0.94 [0.84, 1.04] Right SF (Amygdala) -0.11 0.05 -1.98 0.048 0.9 [0.81, 1.00] Left Area 7M (SPL) 0.16 0.05 3.05 0.002 1.18 [1.06, 1.31] Panic Attacks Independent variable Estimate SE z p OR [95%CI] GAD 0.76 0.04 18.87 < .001 2.15 [1.98, 2.33] Depression (PHQ) 0.13 0.04 3.25 0.001 1.14 [1.05, 1.23] Stress 0.24 0.04 6.21 < .001 1.27 [1.18, 1.36] Childhood Trauma 0.12 0.03 4.01 < .001 1.12 [1.06, 1.19] Female 0.10 0.04 2.56 0.010 1.11 [1.03, 1.20] Left IF (Amygdala) -0.02 0.04 -0.47 0.635 0.98 [0.91, 1.06] Left Area OP8 (Frontal Operculum) -0.08 0.04 -1.95 0.051 0.92 [0.85, 1.00] Right MF (Amygdala) 0.08 0.04 2.06 0.040 1.08 [1.00, 1.65] Smoking (Lifetime Cigarettes) -0.02 0.03 -0.61 0.542 0.98 [0.92, 1.03] Right Area 7A (SPL) -0.08 0.04 -1.77 0.076 0.93 [0.85, 1.01] Note. Significant predictors in bold. Gray matter volume data are confound regressed for sex, age, total intracranial volume, lifetime cigarettes smoked, childhood trauma, and scanner site. SPL = superior parietal lobule, SE = standard error of the mean, OR = odds ratio, CI = confidence interval. Discussion In the present study, we employed multiple machine learning models to model clinically relevant symptoms of GAD and PA in a sample of nearly 26,378 individuals. To classify individuals with or without these symptoms, we utilized regional gray matter volume of a whole-brain parcellation, psychosocial variables (age, sex, number of lifetime cigarettes, symptoms of GAD, PA, depression, stress, and childhood trauma), or a combination of both. We found that a random forest model using only psychosocial variables achieved the highest AUROC for classifying both individuals with GAD and PA. ELNET models combining brain volume data and psychosocial variables demonstrated the highest unbalanced accuracy. Depressive symptoms, childhood trauma, psychological distress, and female sex were the psychosocial variables with the highest predictive importance for classifying clinically relevant GAD symptoms and PA. Current and lifetime PA were among the most important variables for classifying GAD symptoms, while GAD symptoms were the most important statistical predictor for classifying PA. Among brain regions, amygdala volume, areas in the frontal operculum, and the superior parietal lobule showed the highest importance for both symptom categories. Generally, models using only psychosocial variables (such as other psychiatric symptoms, psychological stressors, and demographic variables) demonstrated the best-balanced performance in classifying clinically relevant GAD symptoms (Random Forest AUROC = 0.973) and PA (Random Forest AUROC = 0.933). Gray matter volume data did not substantially improve predictive quality and even led to slightly worse performances (ELNET AUROC = 0.959 for GAD symptoms and 0.876 for PA, respectively). These results align with a recent study by Chavanne et al. ( 2023 ), who predicted the onset of ANX in adolescents using gray matter volume and psychometric data. In their study, psychometric variables alone had sufficient predictive power, and performance sometimes declined when brain volume data was included. Furthermore, our gray matter volume data alone showed only poor classification performance of GAD (Random Forest AUROC = 0.557) or PA (Random Forest AUROC = 0.557), a finding in line with previous other machine learning studies (Bruin et al., 2024 ; Chavanne et al., 2023 ; Harrewijn et al., 2021b ; Hill et al., 2025 ). However, adding brain matter volume to psychosocial variables increased the unbalanced accuracy and specificity. This indicates that the ELNET model with a combination of gray matter volume and psychosocial variables might be more reliable in detecting negative cases, i.e. individuals without GAD or PA, than the model without gray matter volume. Thus, our findings align with those recently reported in the ENIGMA mega-analyses. In one study, Hilbert et al. ( 2024 ) highlighted distinct cortical and subcortical structural changes in specific phobia, with pronounced differences based on subtype. Similarly, our results underscore the importance of subcortical volumes in anxiety-related psychopathology. Furthermore, Groenewold et al. ( 2023 ) emphasize the role of subcortical structures, such as the putamen and pallidum, in ANX, consistent with the involvement of neural circuits important for fear and emotional regulation observed in our data. Of note, our simple correlation analyses indicated significant associations between pre-selected neuroimaging variables and GAD symptoms and PA; however, the correlation values were extremely low and thus not meaningful in terms of clinical application at this stage. Together, these findings reinforce the necessity of integrating neuroanatomical and psychosocial dimensions for a comprehensive understanding of ANX. In our case, variable importance analyses revealed that psychosocial characteristics generally held greater predictive power than variations in gray matter volume. Depressive symptoms were identified as the most important predictor for clinically relevant GAD symptoms and the second most important predictor for PA. This finding aligns with existing research showing that anxiety disorders are frequently followed by depression (Solmi et al., 2022 ; Ter Meulen et al., 2021 ). Moreover, there is substantial comorbidity between GAD and depression (with 45–98% of GAD patients suffering from both conditions, Noyes, 2001 ), and between PD – defined as the disorder with PA as a core symptom – and major depression (with about 50% of PD patients suffering from a depressive episode, Gorman & Coplan, 1996 ). Additionally, symptom overlap between GAD and depression as assessed with GAD-7 and PHQ-9 might contribute to relevant predictive power of depression for GAD symptoms. The most important predictors for classifying PA were symptoms of GAD. Conversely, lifetime PA ranked as the fourth most important variable for classifying symptoms of GAD. This relationship can, again, be explained by the high degree of comorbidity between these syndromes and shared neurobiological factors (Noyes, 2001 ). Interestingly, childhood trauma (i.e., experiences of abuse and neglect) emerged as a relatively important factor for classifying symptoms of both GAD and PA. This aligns with previous research demonstrating that childhood trauma is a substantial risk factor for both syndromes (Hovens et al., 2012 , 2015 ; Klinger-König et al., 2024 ; Kuzminskaite et al., 2021 ). It should be noted, however, that while the relatively complex ELNET classifier trained on the combination of psychosocial and neuroimaging data found associations with both GAD and PA, the simple logistic regression models only demonstrated a significant association with PA. Additionally, female sex proved to be a relatively important variable for machine-learning model performance for both conditions. This finding is consistent with a substantial body of research showing that females are affected by anxiety and depressive disorders more frequently than males, and that machine-learning analyses account for sex-related differences better than simple regression (Jalnapurkar et al., 2018 ; Kessler, 2003 ; Leach et al., 2008 ). Lower age showed relatively modest but notable importance for classifying GAD symptoms in the elastic net regression – although not statistically significant, the effect was descriptively in the same direction as in the logistic regressions. This finding could be explained by the typical age at onset of GAD, which is usually during middle/late adulthood (McGrath et al., 2023 ; Solmi et al., 2022 ), and there is evidence that the intensity and number of symptoms might decrease with increasing age (Le Roux et al., 2005 ; Miloyan et al., 2014 ). Additionally, the lifetime number of cigarettes smoked showed a subtle relationship with PA; however, this relationship was not significant in the simpler logistic regression. While studies suggest that smoking might be a risk factor for developing panic disorder and ANX in general (Breslau & Klein, 1999 ; Ter Meulen et al., 2021 ; Zvolensky et al., 2005 ), it is also possible that smoking develops as a coping mechanism for PA (Lavallee et al., 2021 ). In line with this, one recent systematic review on the prospective bidirectional association of smoking and anxiety pathology showed inconsistent results (Fluharty et al., 2017 ). The amygdala displays a complex anatomical and cellular organization with several subregions having distinct functions and brain connections (Janak & Tye, 2015 ). The individual internal fiber masses consist of medial (MF) and intermediate (IF) fiber bundles and separate the amygdaloid substructures (Kedo et al., 2018 ). The IF and MF are attached to the ventromedial part of the basomedial nucleus (BMA) of the laterobasal complex of the amygdala. The BMA has been shown to be the major target of ventral medial prefrontal cortex projections (mPFC), which constitute one top-down regulation pathway to control anxiety states (Adhikari et al., 2015 ). Additionally, previous work implies the existence of a BMA microcircuit involved in fear acquisition, generalization, recall, and extinction in a distinct and sex-dependent manner (Rajbhandari et al., 2021 ). In our study, higher gray matter volume in the left amygdala IF was associated with clinically relevant GAD symptoms (although this only was detected by the machine-learning models, not the logistic regressions), while lower gray matter volume was linked to PA. These differences may reflect distinct clinical profiles between GAD and PA, with potentially higher mPFC involvement in GAD (Via et al., 2018 ). However, as the specific functional implications of the IF have not been evaluated so far, further studies are needed to clarify its utility as a phenotypic marker for different ANX. The superficial amygdala (SF) is located close to the hippocampal formation. Lower grey matter volume of the SF was associated with GAD symptoms. Studies investigating reactivity to emotional faces and activation of SF suggest its implication in social relevance processing and also fear-related memory (Sedwick VM et al., 2022; Zhu Xiao et al., 2023). In general, previous research has yielded mixed findings regarding the relationship between amygdala gray matter volume and GAD. Some studies report increased amygdala volumes, while others show decreased volumes compared to healthy controls (Schienle et al., 2011 ; for a review, see Madonna et al., 2019 ). A recent study suggests that anomalies in amygdala volume among GAD patients may be driven by comorbid ANX, as individuals with GAD without comorbidities may not exhibit amygdala volume differences compared to healthy controls (Suor et al., 2020 ). Thus, the relationship between amygdala gray matter volume anomalies and GAD may be complex and warrants further investigation in future research. Previous research has shown that smaller amygdala volume was associated with panic disorder (Hayano et al., 2009 ; for a review, see Wang et al., 2021 ), which is in line with our findings of smaller gray matter volume in the left amygdala being an important classification variable for PA. However, we also found higher gray matter volume of the right MF amygdala to be associated with PA. These findings are in contrast to previous research and need further investigation. Given they have different locations within the amygdala, the IF and MF might have individual implications in anxiety-related circuits, which need to be evaluated in further studies. So far, the evidence on volumetric associations with panic (and GAD) is mostly based on total amygdala grey matter volume measurements; however, gray matter volume of subnuclei and fiber masses might more precisely reflect underlying anxiety-related pathological processes, e.g. functional adaptations in response to specific patterns of neural activity inside the amygdala subdivisions or in connection with other brain regions such as the mPFC. Using a detailed microstructural anatomical atlas defining relevant brain areas, like the Julich-Brain Atlas used in this study, could assist in further specifying the differential roles of specific substructures in GAD and panic attacks. In terms of laterality, sex might play a role, as a recent study reported clear sex-differences in the left and right amygdala with differential sex-specific genetic correlation only in the left amygdala (Gui et al., 2025 ). We also found that reduced gray matter volume in the frontal operculum was associated with both GAD and PA, although these relationships were only found with the complex machine-learning analyses and were not significant in the simple logistic regressions. Area OP8 is located in the frontal part of the frontal operculum, the latter being part of a network controlling activity in other brain areas during performance of cognitive tasks in the sense of cognitive control (Higo et al., 2011 ). Previous research has shown that the operculum is associated with negative cognition and worrying (Makovac et al., 2017 ). Excessive worry is the core symptom of GAD and worrying about future PA is an important symptom of panic disorder ( Diagnostic and Statistical Manual of Mental Disorders , 2013). Thus, structural anomalies in the frontal operculum might be an underlying factor or a consequence of anxious ruminations within ANX. Last, we found that larger gray matter volume of the left area of the superior parietal lobule (SPL) was associated with GAD symptoms and smaller gray matter volume of the right area of the SPL was associated with PA. Again, these relationships only emerged in the machine-learning model and were not significant in the logistic regression models. There has been some research showing that ANX might be associated with anomalies and activity changes in superior parietal regions (Wang et al., 2021 , Jin et al., 2020 ). In summary, though there was limited contribution of the neuroimaging variables to the total prediction of GAD symptoms and PA, the analyses detected brain areas clearly involved in anxiety-related circuits, suggesting that neuroanatomical structures could potentially have an additive value in biomarker panels as an imaging-based objective measurement and indication of a present pathological anxiety phenotype. Finally, we would like to point out some limitations and potential directions for future research. First, although the sample size was significantly larger than many previous studies, the population was still limited to an epidemiological cohort consisting of residents of Germany, which may affect the generalizability of the results to other geographical populations and to clinical populations with diagnosed ANX displaying higher symptom severity and potentially higher degrees of neuroanatomical changes. Consequently, we might underestimate the contribution of structural imaging features to GAD and PA classification, and a higher grade of anxiety symptomatology might improve the overall performance of neuroanatomical variables in complex models. Moreover, future studies should aim to replicate these findings in more ethnically, clinically and geographically diverse samples to better understand cross-ethnical differences in the neurobiological underpinnings of GAD and PA. Furthermore, the low prevalence of GAD and PA may have limited the predictive validity in subgroup analyses. Second, the study was based on cross-sectional data, which limits the ability to draw causal conclusions and to make temporal predictions. Longitudinal studies that track brain structure and symptom development over time are needed to assess how structural changes in the brain might influence or be influenced by the progression of ANX. Exploring these additional factors in future research could lead to a more comprehensive understanding of the neurological correlates of GAD and panic-related pathology. Third, despite the use of machine learning algorithms, the predictive power of neuroimaging data alone was relatively low compared to psychosocial variables. This suggests that structural MRI using the imaging protocol, gray/white matter contrast and spatial resolution used in the NAKO study may not fully capture the complexity of ANX. Functional neuroimaging or multimodal approaches, including genetics, may provide additional layers of information that could improve predictive accuracy. Along these lines, we must address the fact that some of the most important features identified by the machine learning analyses did not show significant relationships with the outcome variables when examined using a conventional logistic regression analysis. While this makes interpretation of the results somewhat more challenging, it also highlights the advantages of machine learning approaches in detecting associations that would be difficult to uncover through conventional methods. The applied machine learning approaches can accommodate numerous multicollinear predictors, a known limitation of traditional analyses. We therefore interpret this as a clear strength of our methodology (Altelbany, 2021 ). Conclusion We applied machine learning techniques to a large cohort dataset to classify the presence vs. absence of GAD and PA. The results indicate that psychosocial variables, including symptoms of depression, childhood trauma, and psychological distress are more predictive of GAD and PA than neuroimaging measures of gray matter volume. While the integration of neuroimaging data improved specificity for GAD and PA, it did not significantly enhance overall predictive performance. These findings underscore the importance of considering psychological and demographic factors when developing predictive models for ANX, while also highlighting the potential limitations of relying solely on state-of-the-art structural MRI data. We advocate multimodal approaches and longitudinal designs to better understand the dynamic interplay between neurobiological factors and clinical symptoms, ultimately advancing the identification of objective and measurable biomarkers for ANX. Declarations Data Availability The data that support the findings of this study are part of the German National Cohort (NAKO Gesundheitsstudie). NAKO data are subject to the EU / EEA General Data Protection Regulation and to the NAKO Terms of Use. They can therefore not be deposited in a public repository. Qualified researchers affiliated to EU or EEA institutions may apply for access through the NAKO TransferHub (https://transfer.nako.de). Applications are evaluated by the NAKO Use & Access Committee; successful applicants must sign a Data-Use Agreement and cover associated handling fees. Extended Data Extended Data is available at https://osf.io/wt9yf/?view_only=95484d1d3363458eb7d3cd127cb35d78 All significant correlations between the preselected neural variables and GAD or panic attacks (PA) are presented in Extended Data Table S1. Table S2 provides all correlations between neural variables and GAD, while Table S3 lists those for PA. Variable importance scores from models using only psychosocial predictors are reported in Table S4 (for GAD) and Table S5 (for PA). For models using only neuroimaging variables, importance scores are shown in Table S6 (GAD) and Table S7 (PA). Finally, importance scores for models combining psychosocial and neural variables are presented in Table S8 for GAD and Table S9 for PA. Analysis code is available at https://osf.io/wt9yf/?view_only=95484d1d3363458eb7d3cd127cb35d78 Acknowledgments This project was conducted with data (Application No. NAKO-689) from the German National Cohort (NAKO) (www.nako.de). The NAKO is funded by the Federal Ministry of Education and Research (BMBF) [project funding reference numbers: 01ER1301A/B/C, 01ER1511D, 01ER1801A/B/C/D and 01ER2301A/B/C], DZPG (German Centre for Mental Health Research) and by the BMBF (German Ministry of Education and Research) grant 01EE2303E, federal states of Germany and the Helmholtz Association, the participating universities and the institutes of the Leibniz Association. We thank all participants who took part in the NAKO study. We also thank the staff at the NAKO study centres, the data management and integration centre, and the NAKO head office who enabled the study completion and made the collection of all data possible. Statement of Interest J. Gutzeit, M. Weiß, T. Kuhn, J. Deckert, G. Hein, A. Erhardt-Lehmann, L. Krist, C. Jockwitz, B. Brandes, M.N. Wright, C.M. Friedrich, M. Woeckel, R. Mikolajczyk, T. Keil, S. Castell, P. Betker, M. Rietschel, K. Berger, S. Caspers, M.F. Leitzmann, P. Bohmann, K. Wirkner, Y. Wang, J. Klinger-König, F. Streit, M. Rietschel, and M. Weiß declare that they have no conflicts of interest. H. J. Grabe has received travel grants and speaker honoraria from Neuraxpharm, Servier, Indorsia, and Janssen Cilag. F. Bamberg: Speaker Bureau and unrestricted research grants Siemens Healthineers. C. L. Schlett: Speaker Bureau Siemens Healthineers and Bayer Healthcare; Research Grants Siemens Healthineers. References Abi-Dargham, A., Moeller, S. J., Ali, F., DeLorenzo, C., Domschke, K., Horga, G., Jutla, A., Kotov, R., Paulus, M. P., Rubio, J. M., Sanacora, G., Veenstra-VanderWeele, J., & Krystal, J. H. (2023). Candidate biomarkers in psychiatric disorders: State of the field. World Psychiatry : Official Journal of the World Psychiatric Association (WPA , 22 (2), 236–262. https://doi.org/10.1002/wps.21078 Adhikari, A., Lerner, T. N., Finkelstein, J., Pak, S., Jennings, J. H., Davidson, T. J., Ferenczi, E., Gunaydin, L. A., Mirzabekov, J. J., Ye, L., Kim, S.-Y., Lei, A., & Deisseroth, K. (2015). Basomedial amygdala mediates top-down control of anxiety and fear. Nature , 527 (7577), 179–185. https://doi.org/10.1038/nature15698 Altelbany, S. (2021). Evaluation of Ridge, Elastic Net and Lasso Regression Methods in Precedence of Multicollinearity Problem: A Simulation Study. Journal of Applied Economics and Business Studies , 5 (1), Article 1. https://doi.org/10.34260/jaebs.517 Alves, P. N., Foulon, C., Karolis, V., Bzdok, D., Margulies, D. S., Volle, E., & Thiebaut de Schotten, M. (2019). An improved neuroanatomical model of the default-mode network reconciles previous neuroimaging and neuropathological findings. Communications Biology , 2 (1), 1–14. https://doi.org/10.1038/s42003-019-0611-3 Amunts, K., Mohlberg, H., Bludau, S., & Zilles, K. (2020). Julich-Brain: A 3D probabilistic atlas of the human brain’s cytoarchitecture. Science , 369 (6506), 988–992. https://doi.org/10.1126/science.abb4588 Asselmann, E., Wittchen, H.-U., Lieb, R., Höfler, M., & Beesdo-Baum, K. (2014). Associations of fearful spells and panic attacks with incident anxiety, depressive, and substance use disorders: A 10-year prospective-longitudinal community study of adolescents and young adults. Journal of Psychiatric Research , 55 , 8–14. https://doi.org/10.1016/j.jpsychires.2014.04.001 Bamberg, F., Kauczor, H. U., Weckbach, S., Schlett, C. L., Forsting, M., Ladd, S. C., Greiser, K. H., Weber, M. A., Schulz-Menger, J., Niendorf, T., Pischon, T., Caspers, S., Amunts, K., Berger, K., Bülow, R., Hosten, N., Hegenscheid, K., Kröncke, T., Linseisen, J., … ... (2015). German National Cohort MRI Study Investigators. Radiology , 277 (1), 206–220. https://doi.org/10.1148/radiol.2015142272 Batelaan, N. M., Rhebergen, D., de Graaf, R., Spijker, J., Beekman, A. T. F., & Penninx, B. W. J. H. (2012). Panic attacks as a dimension of psychopathology: Evidence for associations with onset and course of mental disorders and level of functioning. The Journal of Clinical Psychiatry , 73 (9), 1195–1202. https://doi.org/10.4088/JCP.12m07743 Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., Stokes, J., Handelsman, L., Medrano, M., Desmond, D., & Zule, W. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse & Neglect , 27 (2), 169–190. https://doi.org/10.1016/s0145-2134(02)00541-0 Boeke, E. A., Holmes, A. J., & Phelps, E. A. (2020). Toward Robust Anxiety Biomarkers: A Machine Learning Approach in a Large-Scale Sample. Biological Psychiatry. Cognitive Neuroscience and Neuroimaging , 5 (8), 799–807. https://doi.org/10.1016/j.bpsc.2019.05.018 Breslau, N., & Klein, D. F. (1999). Smoking and Panic Attacks: An Epidemiologic Investigation. Archives of General Psychiatry , 56 (12), 1141–1147. https://doi.org/10.1001/archpsyc.56.12.1141 Bruin, W. B., Zhutovsky, P., van Wingen, G. A., Bas-Hoogendam, J. M., Groenewold, N. A., Hilbert, K., Winkler, A. M., Zugman, A., Agosta, F., Åhs, F., Andreescu, C., Antonacci, C., Asami, T., Assaf, M., Barber, J. P., Bauer, J., Bavdekar, S. Y., Beesdo-Baum, K., Benedetti, F., … Aghajani, M. (2024). Brain-based classification of youth with anxiety disorders: Transdiagnostic examinations within the ENIGMA-Anxiety database using machine learning. Nature Mental Health , 2 (1), 104–118. https://doi.org/10.1038/s44220-023-00173-2 Bzdok, D., & Meyer-Lindenberg, A. (2018). Machine Learning for Precision Psychiatry: Opportunities and Challenges. Biological Psychiatry. Cognitive Neuroscience and Neuroimaging , 3 (3), 223–230. https://doi.org/10.1016/j.bpsc.2017.11.007 Catani, M., Dell’Acqua, F., & Thiebaut de Schotten, M. (2013). A revised limbic system model for memory, emotion and behaviour. Neuroscience & Biobehavioral Reviews , 37 (8), 1724–1737. https://doi.org/10.1016/j.neubiorev.2013.07.001 Chavanne, A. V., Paillère Martinot, M. L., Penttilä, J., Grimmer, Y., Conrod, P., Stringaris, A., van Noort, B., Isensee, C., Becker, A., Banaschewski, T., Bokde, A. L. W., Desrivières, S., Flor, H., Grigis, A., Garavan, H., Gowland, P., Heinz, A., Brühl, R., Nees, F., … Artiges, E. (2023). Anxiety onset in adolescents: A machine-learning prediction. Molecular Psychiatry , 28 (2), 639–646. https://doi.org/10.1038/s41380-022-01840-z Cichocki, A. C., Zinbarg, R. E., Craske, M. G., Chat, I. K., Young, K. S., Bookheimer, S. Y., & Nusslock, R. (2024). Transdiagnostic symptom of depression and anxiety associated with reduced gray matter volume in prefrontal cortex. Psychiatry Research. Neuroimaging , 339 , 111791. https://doi.org/10.1016/j.pscychresns.2024.111791 Core R Team. (2021). A Language and Environment for Statistical Computing. In R Foundation for Statistical Computing (Bd. 2, S. https://www.R--project.org) [Software]. R Foundation for Statistical Computing. http://www.r-project.org Craske, M. G., Stein, M. B., Eley, T. C., Milad, M. R., Holmes, A., Rapee, R. M., & Wittchen, H.-U. (2017). Anxiety disorders. Nature Reviews Disease Primers , 3 (1), 1–19. https://doi.org/10.1038/nrdp.2017.24 de Jonge, P., Roest, A. M., Lim, C. C. W., Florescu, S. E., Bromet, E. J., Stein, D. J., Harris, M., Nakov, V., Caldas-de-Almeida, J. M., Levinson, D., Al-Hamzawi, A. O., Haro, J. M., Viana, M. C., Borges, G., O’Neill, S., de Girolamo, G., Demyttenaere, K., Gureje, O., Iwata, N., … Scott, K. M. (2016). Cross-national epidemiology of panic disorder and panic attacks in the world mental health surveys. Depression and Anxiety , 33 (12), 1155–1177. https://doi.org/10.1002/da.22572 Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed). (2013). American psychiatric association. Erhardt, A., Gelbrich, G., Klinger-König, J., Streit, F., Kleineidam, L., Riedel-Heller, S. G., NAKO Investigators, Schmidt, B., Schmiedek, F., Wagner, M., Grabe, H. J., Rietschel, M., Berger, K., & Deckert, J. (2023). Generalised anxiety and panic symptoms in the German National Cohort (NAKO). The World Journal of Biological Psychiatry: The Official Journal of the World Federation of Societies of Biological Psychiatry , 24 (10), 881–896. https://doi.org/10.1080/15622975.2021.2011409 Etkin, A., Prater, K. E., Schatzberg, A. F., Menon, V., & Greicius, M. D. (2009). Disrupted amygdalar subregion functional connectivity and evidence of a compensatory network in generalized anxiety disorder. Archives of General Psychiatry , 66 (12), 1361–1372. https://doi.org/10.1001/archgenpsychiatry.2009.104 Fluharty, M., Taylor, A. E., Grabski, M., & Munafò, M. R. (2017). The Association of Cigarette Smoking With Depression and Anxiety: A Systematic Review. Nicotine & Tobacco Research , 19 (1), 3–13. https://doi.org/10.1093/ntr/ntw140 Ganganwar, V. (2012). An overview of classification algorithms for imbalanced datasets . 2 (4). Goodwin, R. D., Lieb, R., Hoefler, M., Pfister, H., Bittner, A., Beesdo, K., & Wittchen, H.-U. (2004). Panic attack as a risk factor for severe psychopathology. The American Journal of Psychiatry , 161 (12), 2207–2214. https://doi.org/10.1176/appi.ajp.161.12.2207 Gorman, J. M., & Coplan, J. D. (1996). Comorbidity of depression and panic disorder. The Journal of Clinical Psychiatry , 57 Suppl 10 , 34–41; discussion 42-43. Groenewold, N. A., Bas-Hoogendam, J. M., Amod, A. R., Laansma, M. A., Van Velzen, L. S., Aghajani, M., Hilbert, K., Oh, H., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, J. R., Blair, K. S., Hirsch, J., Pantazatos, S. P., Schneier, F. R., Talati, A., Roelofs, K., … Van Der Wee, N. J. A. (2023). Volume of subcortical brain regions in social anxiety disorder: Mega-analytic results from 37 samples in the ENIGMA-Anxiety Working Group. Molecular Psychiatry , 28 (3), 1079–1089. https://doi.org/10.1038/s41380-022-01933-9 Grummitt, L., Baldwin, J. R., Lafoa’i, J., Keyes, K. M., & Barrett, E. L. (2024). Burden of Mental Disorders and Suicide Attributable to Childhood Maltreatment. JAMA Psychiatry , 81 (8), 782–788. https://doi.org/10.1001/jamapsychiatry.2024.0804 Guan, X., & Cao, P. (2024). Brain Mechanisms Underlying Panic Attack and Panic Disorder. Neuroscience Bulletin , 40 (6), 795–814. https://doi.org/10.1007/s12264-023-01088-9 Gui, Y., Zhou, G., Cui, S., Li, H., Lu, H., & Zhao, H. (2025). The left amygdala is genetically sexually-dimorphic: Multi-omics analysis of structural MRI volumes. Translational Psychiatry , 15 (1), 1–7. https://doi.org/10.1038/s41398-025-03223-8 Harrewijn, A., Cardinale, E. M., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Hilbert, K., Cardoner, N., Porta-Casteràs, D., Gosnell, S., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, K. S., Blair, J. R., Hammoud, M. Z., Milad, M. R., Burkhouse, K. L., Phan, K. L., … Pine, D. S. (2021a). Cortical and subcortical brain structure in generalized anxiety disorder: Findings from 28 research sites in the ENIGMA-Anxiety Working Group. Translational Psychiatry , 11 (1), 1–15. https://doi.org/10.1038/s41398-021-01622-1 Harrewijn, A., Cardinale, E. M., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Hilbert, K., Cardoner, N., Porta-Casteràs, D., Gosnell, S., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, K. S., Blair, J. R., Hammoud, M. Z., Milad, M. R., Burkhouse, K. L., Phan, K. L., … Pine, D. S. (2021b). Cortical and subcortical brain structure in generalized anxiety disorder: Findings from 28 research sites in the ENIGMA-Anxiety Working Group. Translational Psychiatry , 11 (1), Article 1. https://doi.org/10.1038/s41398-021-01622-1 Hayano, F., Nakamura, M., Asami, T., Uehara, K., Yoshida, T., Roppongi, T., Otsuka, T., Inoue, T., & Hirayasu, Y. (2009). Smaller amygdala is associated with anxiety in patients with panic disorder. Psychiatry and Clinical Neurosciences , 63 (3), 266–276. https://doi.org/10.1111/j.1440-1819.2009.01960.x Higo, T., Mars, R. B., Boorman, E. D., Buch, E. R., & Rushworth, M. F. S. (2011). Distributed and causal influence of frontal operculum in task control. Proceedings of the National Academy of Sciences , 108 (10), 4230–4235. https://doi.org/10.1073/pnas.1013361108 Hilbert, K., Boeken, O. J., Langhammer, T., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Zugman, A., Åhs, F., Arolt, V., Beesdo-Baum, K., Björkstrand, J., Blackford, J. U., Blanco-Hinojo, L., Böhnlein, J., Bülow, R., Cano, M., Cardoner, N., Caseras, X., Dannlowski, U., … Lueken, U. (2024). Cortical and Subcortical Brain Alterations in Specific Phobia and Its Animal and Blood-Injection-Injury Subtypes: A Mega-Analysis From the ENIGMA Anxiety Working Group. American Journal of Psychiatry , 181 (8), 728–740. https://doi.org/10.1176/appi.ajp.20230032 Hill, E. D., Kashyap, P., Raffanello, E., Wang, Y., Moffitt, T. E., Caspi, A., Engelhard, M., & Posner, J. (2025). Prediction of mental health risk in adolescents. Nature Medicine . https://doi.org/10.1038/s41591-025-03560-7 Hovens, J. G. F. M., Giltay, E. J., Spinhoven, P., Hemert, A. M. van, & Penninx, B. W. J. H. (2015). Impact of Childhood Life Events and Childhood Trauma on the Onset and Recurrence of Depressive and Anxiety Disorders. The Journal of Clinical Psychiatry , 76 (7), 8559. https://doi.org/10.4088/JCP.14m09135 Hovens, J. G. F. M., Giltay, E. J., Wiersma, J. E., Spinhoven, P., Penninx, B. W. J. H., & Zitman, F. G. (2012). Impact of childhood life events and trauma on the course of depressive and anxiety disorders. Acta Psychiatrica Scandinavica , 126 (3), 198–207. https://doi.org/10.1111/j.1600-0447.2011.01828.x Jalnapurkar, I., Allen, M., & Pigott, T. (2018). Sex differences in anxiety disorders: A review. J Psychiatry Depress Anxiety , 4 (12), 3–16. Janak, P. H., & Tye, K. M. (2015). From circuits to behaviour in the amygdala. Nature , 517 (7534), 284–292. https://doi.org/10.1038/nature14188 Janssen, R. J., Mourão-Miranda, J., & Schnack, H. G. (2018). Making Individual Prognoses in Psychiatry Using Neuroimaging and Machine Learning. Biological Psychiatry. Cognitive Neuroscience and Neuroimaging , 3 (9), 798–808. https://doi.org/10.1016/j.bpsc.2018.04.004 Jin, H., Zhang, B., Cui, H., Li, W., Li, H., Hu, Q., Wang, J., & Li, C. (2020). Altered Function of Superior Parietal Lobule Associated with Perceptive Awareness in First-Episode Drug-Naïve Panic Disorders: A Preliminary fMRI Study. Neuropsychiatric Disease and Treatment , 16 , 1653–1659. https://doi.org/10.2147/NDT.S248453 Kalisch, R., Russo, S. J., & Müller, M. B. (2024). Neurobiology and systems biology of stress resilience. Physiological Reviews , 104 (3), 1205–1263. https://doi.org/10.1152/physrev.00042.2023 Kedo, O., Zilles, K., Palomero-Gallagher, N., Schleicher, A., Mohlberg, H., Bludau, S., & Amunts, K. (2018). Receptor-driven, multimodal mapping of the human amygdala. Brain Structure and Function , 223 (4), 1637–1666. https://doi.org/10.1007/s00429-017-1577-x Kessler, R. C. (2003). Epidemiology of women and depression. Journal of Affective Disorders , 74 (1), 5–13. https://doi.org/10.1016/S0165-0327(02)00426-3 Klinger-König, J., Erhardt, A., Streit, F., Völker, M. P., Schulze, M. B., Keil, T., Fricke, J., Castell, S., Klett-Tammen, C. J., Pischon, T., Karch, A., Teismann, H., Michels, K. B., Greiser, K. H., Becher, H., Karrasch, S., Ahrens, W., Meinke-Franze, C., Schipf, S., … Grabe, H. J. (2024). Childhood Trauma and Somatic and Mental Illness in Adulthood. Deutsches Ärzteblatt International , 121 (1), 1–8. https://doi.org/10.3238/arztebl.m2023.0225 Klinger-König, J., Streit, F., Erhardt, A., Kleineidam, L., Schmiedek, F., Schmidt, B., Wagner, M., Deckert, J., Rietschel, M., Berger, K., & Grabe, H. J. (2023). The assessment of childhood maltreatment and its associations with affective symptoms in adulthood: Results of the German National Cohort (NAKO). The World Journal of Biological Psychiatry , 24 (10), 897–908. https://doi.org/10.1080/15622975.2021.2011406 Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine , 16 (9), 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x Kuhn, M. (2008). Building Predictive Models in R Using the caret Package. Journal of Statistical Software , 28 , 1–26. https://doi.org/10.18637/jss.v028.i05 Kuhn, M. (2012). Variable importance using the caret package. Journal of Statistical Software , 6 , 1–6. Kuzminskaite, E., Penninx, B. W. J. H., van Harmelen, A.-L., Elzinga, B. M., Hovens, J. G. F. M., & Vinkers, C. H. (2021). Childhood Trauma in Adult Depressive and Anxiety Disorders: An Integrated Review on Psychological and Biological Mechanisms in the NESDA Cohort. Journal of Affective Disorders , 283 , 179–191. https://doi.org/10.1016/j.jad.2021.01.054 Lavallee, K. L., Zhang, X. C., Schneider, S., & Margraf, J. (2021). A longitudinal examination of the relationship between smoking and panic, anxiety, and depression in Chinese and German students. Addictive Behaviors Reports , 14 , 100347. https://doi.org/10.1016/j.abrep.2021.100347 Le Roux, H., Gatz, M., & Wetherell, J. L. (2005). Age at Onset of Generalized Anxiety Disorder in Older Adults. The American Journal of Geriatric Psychiatry , 13 (1), 23–30. https://doi.org/10.1097/00019442-200501000-00005 Leach, L. S., Christensen, H., Mackinnon, A. J., Windsor, T. D., & Butterworth, P. (2008). Gender differences in depression and anxiety across the adult lifespan: The role of psychosocial mediators. Social Psychiatry and Psychiatric Epidemiology , 43 (12), 983–998. https://doi.org/10.1007/s00127-008-0388-z Linden, D. E. J. (2012). The challenges and promise of neuroimaging in psychiatry. Neuron , 73 (1), 8–22. https://doi.org/10.1016/j.neuron.2011.12.014 Madonna, D., Delvecchio, G., Soares, J. C., & Brambilla, P. (2019). Structural and functional neuroimaging studies in generalized anxiety disorder: A systematic review. Brazilian Journal of Psychiatry , 41 , 336–362. https://doi.org/10.1590/1516-4446-2018-0108 Makovac, E., Meeten, F., Watson, D. R., Garfinkel, S. N., Critchley, H. D., & Ottaviani, C. (2015). Neurostructural abnormalities associated with axes of emotion dysregulation in generalized anxiety. Neuroimage Clin , 10 , 172–181. https://doi.org/10.1016/j.nicl.2015.11.022. Makovac, E., Smallwood, J., Watson, D. R., Meeten, F., Critchley, H. D., & Ottaviani, C. (2017). The verbal nature of worry in generalized anxiety: Insights from the brain. NeuroImage : Clinical , 17 , 882–892. https://doi.org/10.1016/j.nicl.2017.12.014 McGrath, J. J., Al-Hamzawi, A., Alonso, J., Altwaijri, Y., Andrade, L. H., Bromet, E. J., Bruffaerts, R., Almeida, J. M. C. de, Chardoul, S., Chiu, W. T., Degenhardt, L., Demler, O. V., Ferry, F., Gureje, O., Haro, J. M., Karam, E. G., Karam, G., Khaled, S. M., Kovess-Masfety, V., … Zaslavsky, A. M. (2023). Age of onset and cumulative risk of mental disorders: A cross-national analysis of population surveys from 29 countries. The Lancet Psychiatry , 10 (9), 668–681. https://doi.org/10.1016/S2215-0366(23)00193-1 Miloyan, B., Byrne, G. J., & Pachana, N. A. (2014). Age-related changes in generalized anxiety disorder symptoms. International Psychogeriatrics , 26 (4), 565–572. https://doi.org/10.1017/S1041610213002470 More, S., Eickhoff, S. B., Caspers, J., & Patil, K. R. (2021). Confound Removal and Normalization in Practice: A Neuroimaging Based Sex Prediction Case Study. Machine Learning and Knowledge Discovery in Databases. Applied Data Science and Demo Track , 12461 , 3–18. https://doi.org/10.1007/978-3-030-67670-4_1 Noyes, R. (2001). Comorbidity in Generalized Anxiety Disorder. Psychiatric Clinics of North America , 24 (1), 41–55. https://doi.org/10.1016/S0193-953X(05)70205-7 Penninx, B. W., Pine, D. S., Holmes, E. A., & Reif, A. (2021). Anxiety disorders. Lancet (London, England) , 397 (10277), 914–927. https://doi.org/10.1016/S0140-6736(21)00359-7 Pessoa, L. (2023). How many brain regions are needed to elucidate the neural bases of fear and anxiety? Neuroscience & Biobehavioral Reviews , 146 , 105039. https://doi.org/10.1016/j.neubiorev.2023.105039 Peters, A., Consortium, G. N. C., Peters, A., Greiser, K. H., Göttlicher, S., Ahrens, W., Albrecht, M., Bamberg, F., Bärnighausen, T., Becher, H., Berger, K., Beule, A., Boeing, H., Bohn, B., Bohnert, K., Braun, B., Brenner, H., Bülow, R., Castell, S., … Zschocke, J. (2022). Framework and baseline examination of the German National Cohort (NAKO. European journal of epidemiology , 37 (10), 1107–1124. https://doi.org/10.1007/s10654-022-00890-5 Rajbhandari, A. K., Octeau, C. J., Gonzalez, S., Pennington, Z. T., Mohamed, F., Trott, J., Chavez, J., Ngyuen, E., Keces, N., Hong, W. Z., Neve, R. L., Waschek, J., Khakh, B. S., & Fanselow, M. S. (2021). A Basomedial Amygdala to Intercalated Cells Microcircuit Expressing PACAP and Its Receptor PAC1 Regulates Contextual Fear. Journal of Neuroscience , 41 (15), 3446–3461. https://doi.org/10.1523/JNEUROSCI.2564-20.2021 Ruscio, A. M., Hallion, L. S., Lim, C. C. W., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., Andrade, L. H., Borges, G., Bromet, E. J., Bunting, B., Caldas de Almeida, J. M., Demyttenaere, K., Florescu, S., de Girolamo, G., Gureje, O., Haro, J. M., He, Y., Hinkov, H., Hu, C., … Scott, K. M. (2017). Cross-sectional Comparison of the Epidemiology of DSM-5 Generalized Anxiety Disorder Across the Globe. JAMA Psychiatry , 74 (5), 465–475. https://doi.org/10.1001/jamapsychiatry.2017.0056 Schienle, A., Ebner, F., & Schäfer, A. (2011). Localized gray matter volume abnormalities in generalized anxiety disorder. European Archives of Psychiatry and Clinical Neuroscience , 261 (4), 303–307. https://doi.org/10.1007/s00406-010-0147-5 Schuppert, C., Rospleszcz, S., Hirsch, J. G., Hoinkiss, D. C., Köhn, A., Krüchten, R., Russe, M. F., Keil, T., Krist, L., Schmidt, B., Michels, K. B., Schipf, S., Brenner, H., Kröncke, T. J., Pischon, T., Niendorf, T., Schulz-Menger, J., Forsting, M., Völzke, H., … Schlett, C. L. (2023). Automated image quality assessment for selecting among multiple magnetic resonance image acquisitions in the German National Cohort study. Scientific Reports , 13 (1), 22745. https://doi.org/10.1038/s41598-023-49569-1 Snoek, L., Miletić, S., & Scholte, H. S. (2019). How to control for confounds in decoding analyses of neuroimaging data. NeuroImage , 184 , 741–760. https://doi.org/10.1016/j.neuroimage.2018.09.074 Solmi, M., Radua, J., Olivola, M., Croce, E., Soardo, L., Salazar de Pablo, G., Il Shin, J., Kirkbride, J. B., Jones, P., Kim, J. H., Kim, J. Y., Carvalho, A. F., Seeman, M. V., Correll, C. U., & Fusar-Poli, P. (2022). Age at onset of mental disorders worldwide: Large-scale meta-analysis of 192 epidemiological studies. Molecular Psychiatry , 27 (1), 281–295. https://doi.org/10.1038/s41380-021-01161-7 Spitzer, R. L., Kroenke, K., & Williams, J. B. (1999). Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA , 282 (18), 1737–1744. https://doi.org/10.1001/jama.282.18.1737 Streit, F., Zillich, L., Frank, J., Kleineidam, L., Wagner, M., Baune, B. T., Klinger-König, J., Grabe, H. J., Pabst, A., Riedel-Heller, S. G., Schmiedek, F., Schmidt, B., Erhardt, A., Deckert, J., NAKO Investigators, Rietschel, M., & Berger, K. (2023). Lifetime and current depression in the German National Cohort (NAKO). The World Journal of Biological Psychiatry: The Official Journal of the World Federation of Societies of Biological Psychiatry , 24 (10), 865–880. https://doi.org/10.1080/15622975.2021.2014152 Strom, N. I., Verhulst, B., Bacanu, S.-A., Cheesman, R., Purves, K. L., Gedik, H., Mitchell, B. L., Kwong, A. S., Faucon, A. B., Singh, K., Medland, S., Colodro-Conde, L., Krebs, K., Hoffmann, P., Herms, S., Gehlen, J., Ripke, S., Awasthi, S., Palviainen, T., … Hettema, J. M. (2024). Genome-wide association study of major anxiety disorders in 122,341 European-ancestry cases identifies 58 loci and highlights GABAergic signaling. medRxiv: The Preprint Server for Health Sciences , 2024.07.03.24309466. https://doi.org/10.1101/2024.07.03.24309466 Suor, J. H., Jimmy, J., Monk, C. S., Phan, K. L., & Burkhouse, K. L. (2020). Parsing differences in amygdala volume among individuals with and without social and generalized anxiety disorders across the lifespan. Journal of Psychiatric Research , 128 , 83–89. https://doi.org/10.1016/j.jpsychires.2020.05.027 Teicher, M. H., Samson, J. A., Anderson, C. M., & Ohashi, K. (2016). The effects of childhood maltreatment on brain structure, function and connectivity. Nature Reviews Neuroscience , 17 (10), 652–666. https://doi.org/10.1038/nrn.2016.111 Ter Meulen, W. G., Draisma, S., van Hemert, A. M., Schoevers, R. A., Kupka, R. W., Beekman, A. T. F., & Penninx, B. W. J. H. (2021). Depressive and anxiety disorders in concert-A synthesis of findings on comorbidity in the NESDA study. Journal of Affective Disorders , 284 , 85–97. https://doi.org/10.1016/j.jad.2021.02.004 ter Meulen, W. G., Draisma, S., van Hemert, A. M., Schoevers, R. A., Kupka, R. W., Beekman, A. T. F., & Penninx, B. W. J. H. (2021). Depressive and anxiety disorders in concert–A synthesis of findings on comorbidity in the NESDA study. Journal of Affective Disorders , 284 , 85–97. https://doi.org/10.1016/j.jad.2021.02.004 Via, E., Fullana, M. A., Goldberg, X., Tinoco-González, D., Martínez-Zalacaín, I., Soriano-Mas, C., Davey, C. G., Menchón, J. M., Straube, B., Kircher, T., Pujol, J., Cardoner, N., & Harrison, B. J. (2018). Ventromedial prefrontal cortex activity and pathological worry in generalised anxiety disorder. The British Journal of Psychiatry: The Journal of Mental Science , 213 (1), 437–443. https://doi.org/10.1192/bjp.2018.65 Wang, X., Cheng, B., Wang, S., Lu, F., Luo, Y., Long, X., & Kong, D. (2021). Distinct grey matter volume alterations in adult patients with panic disorder and social anxiety disorder: A systematic review and voxel-based morphometry meta-analysis. Journal of Affective Disorders , 281 , 805–823. https://doi.org/10.1016/j.jad.2020.11.057 Weiß, M., Gutzeit, J., Appel, K. S., Bahmer, T., Beutel, M., Deckert, J., Fricke, J., Hanß, S., Hettich-Damm, N., Heuschmann, P. U., Horn, A., Jauch-Chara, K., Kohls, M., Krist, L., Lorenz-Depiereux, B., Otte, C., Pape, D., Reese, J.-P., Schreiber, S., … Hein, G. (2024). Depression and fatigue six months post-COVID-19 disease are associated with overlapping symptom constellations: A prospective, multi-center, population-based cohort study. Journal of Affective Disorders , 352 , 296–305. https://doi.org/10.1016/j.jad.2024.02.041 Wittchen, H. U., Jacobi, F., Rehm, J., Gustavsson, A., Svensson, M., Jönsson, B., Olesen, J., Allgulander, C., Alonso, J., Faravelli, C., Fratiglioni, L., Jennum, P., Lieb, R., Maercker, A., van Os, J., Preisig, M., Salvador-Carulla, L., Simon, R., & Steinhausen, H.-C. (2011). The size and burden of mental disorders and other disorders of the brain in Europe 2010. European Neuropsychopharmacology: The Journal of the European College of Neuropsychopharmacology , 21 (9), 655–679. https://doi.org/10.1016/j.euroneuro.2011.07.018 Zvolensky, M. J., Feldner, M. T., Leen-Feldner, E. W., & McLeish, A. C. (2005). Smoking and panic attacks, panic disorder, and agoraphobia: A review of the empirical literature. Clinical Psychology Review , 25 (6), 761–789. https://doi.org/10.1016/j.cpr.2005.05.001 Footnotes In this context, “predict” does not imply a temporal or causal forecast. Instead, it follows the common usage in machine learning and statistics, referring to the explanatory value of a model. The term “predict” is used throughout this paper in that sense, unless explicitly indicated otherwise. Additional Declarations Yes H. J. Grabe has received travel grants and speaker honoraria from Neuraxpharm, Servier, In-dorsia, and Janssen Cilag. F. Bamberg: Speaker Bureau and unrestricted research grants Siemens Healthineers. C. L. Schlett: Speaker Bureau Siemens Healthineers and Bayer Healthcare; Research Grants Siemens Healthineers. Cite Share Download PDF Status: Under Review Version 1 posted Unknown event 02 Oct, 2025 Editorial decision: Reject after peer review 26 Sep, 2025 Review # 2 received at journal 21 Sep, 2025 Review # 1 received at journal 13 Sep, 2025 Reviewer # 2 agreed at journal 09 Sep, 2025 Reviewer # 1 agreed at journal 08 Sep, 2025 Reviewers invited by journal 08 Sep, 2025 Editor assigned by journal 12 May, 2025 Submission checks completed at journal 12 May, 2025 First submitted to journal 09 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6627834","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":511674167,"identity":"0b37dbb5-c47f-455a-b394-bdc83237abb7","order_by":0,"name":"Julian Gutzeit","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYNACAwYGxmbmA4wNQLqBsHJmkBYDBsZ2tgRStICs4ecxIE6LfET+4Q8fCv4wMDfzfJOc2cYg209Ii+GNZAbDGSCHNfNuk9zYxmA8k5A1hjOSGZJ5YFoebmNI3HCACC2H/4C18DwDa9lPSIu8RDJjMwNEC5vkRpAthPxiwPPYmLHHwBiohc3YcuY/CeMZBG1pT3z84ccfOQbD/sMPb/acsZHtbyBkC9TM+o0QlRKEnAW0BWamPGG1o2AUjIJRMFIBAKQxPbCKZkKcAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-1434-1645","institution":"University Hospital Würzburg, Würzburg, Germany","correspondingAuthor":true,"prefix":"","firstName":"Julian","middleName":"","lastName":"Gutzeit","suffix":""},{"id":511674168,"identity":"c5497c42-c126-4467-8c7c-65c1e978dd5d","order_by":1,"name":"Martin Weiß","email":"","orcid":"https://orcid.org/0000-0002-0569-0907","institution":"University of Würzburg, Institute of Psychology","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Weiß","suffix":""},{"id":511674169,"identity":"18f50d5c-6e48-47c0-b58a-4237124aa268","order_by":2,"name":"Tierney Kuhn","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tierney","middleName":"","lastName":"Kuhn","suffix":""},{"id":511674170,"identity":"7676b240-2d41-4da3-9a3d-799c7071a66b","order_by":3,"name":"Johanna Klinger-König","email":"","orcid":"https://orcid.org/0000-0003-2287-7914","institution":"University Medicine Greifswald","correspondingAuthor":false,"prefix":"","firstName":"Johanna","middleName":"","lastName":"Klinger-König","suffix":""},{"id":511674172,"identity":"b3c5a4f6-aa98-4bf6-8d88-62fda00db3ec","order_by":4,"name":"Fabian Streit","email":"","orcid":"https://orcid.org/0000-0003-1080-4339","institution":"Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Fabian","middleName":"","lastName":"Streit","suffix":""},{"id":511674173,"identity":"19dbc7ba-5459-48be-b3fc-c55822e95858","order_by":5,"name":"Christiane Jockwitz","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Christiane","middleName":"","lastName":"Jockwitz","suffix":""},{"id":511674175,"identity":"e67b8c9d-86d8-43b9-98a3-ab52256c8f90","order_by":6,"name":"Berit Brandes","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Berit","middleName":"","lastName":"Brandes","suffix":""},{"id":511674176,"identity":"4013cce3-0b7f-4a18-b3f6-c96ae394b80b","order_by":7,"name":"Marvin Wright","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Marvin","middleName":"","lastName":"Wright","suffix":""},{"id":511674177,"identity":"688f78e7-8f86-4c56-a65a-b363ded7d7dd","order_by":8,"name":"Christoph Friedrich","email":"","orcid":"https://orcid.org/0000-0001-7906-0038","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Christoph","middleName":"","lastName":"Friedrich","suffix":""},{"id":511674178,"identity":"62958de3-4f4f-430c-9faa-0f75071e1b12","order_by":9,"name":"Margarethe Woeckel","email":"","orcid":"https://orcid.org/0000-0003-0219-8906","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Margarethe","middleName":"","lastName":"Woeckel","suffix":""},{"id":511674179,"identity":"91428ecd-ad9d-4f55-954a-16de88230690","order_by":10,"name":"Rafael Mikolajczyk","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Rafael","middleName":"","lastName":"Mikolajczyk","suffix":""},{"id":511674180,"identity":"18b93621-00fb-4523-910b-54da28b8a993","order_by":11,"name":"Thomas Keil","email":"","orcid":"https://orcid.org/0000-0002-9108-3360","institution":"Institute of Social Medicine, Epidemiology and Health Economics, Charité - Universitätsmedizin Berlin","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Keil","suffix":""},{"id":511674181,"identity":"e17e3c5c-1ccf-43d9-95bc-bade1b46afa9","order_by":12,"name":"Stefanie Castell","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Stefanie","middleName":"","lastName":"Castell","suffix":""},{"id":511674182,"identity":"a17bc866-8576-4af2-848e-49d48ae9bfe1","order_by":13,"name":"Phlinie Betker","email":"","orcid":"https://orcid.org/0009-0003-5534-0010","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Phlinie","middleName":"","lastName":"Betker","suffix":""},{"id":511674183,"identity":"a7692105-6ebc-4ed6-8a71-75e7331e358a","order_by":14,"name":"Christopher Schlett","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Schlett","suffix":""},{"id":511674184,"identity":"8bff7a3d-30c9-47a3-97d2-29f7c5433acd","order_by":15,"name":"Till Bärnighausen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Till","middleName":"","lastName":"Bärnighausen","suffix":""},{"id":511674185,"identity":"d4e64b3d-ad48-4618-bc2b-027ad4479837","order_by":16,"name":"Fabian Bamberg","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Fabian","middleName":"","lastName":"Bamberg","suffix":""},{"id":511674186,"identity":"7fed3256-d757-432a-9010-99b22172373c","order_by":17,"name":"Matthias Günther","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Matthias","middleName":"","lastName":"Günther","suffix":""},{"id":511674187,"identity":"14e832a0-fe29-4d47-a846-8d461443e98f","order_by":18,"name":"Jochen Hirsch","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jochen","middleName":"","lastName":"Hirsch","suffix":""},{"id":511674188,"identity":"7cc22e61-afff-4c2f-a17b-5a30b949dd73","order_by":19,"name":"Tobias Pischon","email":"","orcid":"https://orcid.org/0000-0003-1568-767X","institution":"Max Delbrück Center","correspondingAuthor":false,"prefix":"","firstName":"Tobias","middleName":"","lastName":"Pischon","suffix":""},{"id":511674189,"identity":"b4f33e3c-c291-4930-b942-89824e9c5b16","order_by":20,"name":"Thoralf Niendorf","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Thoralf","middleName":"","lastName":"Niendorf","suffix":""},{"id":511674190,"identity":"3939bbdf-1822-42b7-b1e2-b7256d38f43c","order_by":21,"name":"Michael Leitzmann","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Leitzmann","suffix":""},{"id":511674191,"identity":"f939ee93-d804-4671-a916-991956b2bc49","order_by":22,"name":"Patricia Bohmann","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Patricia","middleName":"","lastName":"Bohmann","suffix":""},{"id":511674192,"identity":"5893b79f-9046-4cb4-9eda-2c858bb61443","order_by":23,"name":"Kerstin Wirkner","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Kerstin","middleName":"","lastName":"Wirkner","suffix":""},{"id":511674193,"identity":"f9415a4b-5d6f-4cf7-96af-408e62808220","order_by":24,"name":"Lilian Krist","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lilian","middleName":"","lastName":"Krist","suffix":""},{"id":511674194,"identity":"18350e5e-eb04-4637-8d0a-89dea3aefc3f","order_by":25,"name":"Yanding Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yanding","middleName":"","lastName":"Wang","suffix":""},{"id":511674195,"identity":"0613d6cc-4582-4c8a-9fd5-1b56cc1ea37f","order_by":26,"name":"Klaus Berger","email":"","orcid":"","institution":"University of Muenster","correspondingAuthor":false,"prefix":"","firstName":"Klaus","middleName":"","lastName":"Berger","suffix":""},{"id":511674196,"identity":"53a5ee6c-64fa-480b-89a2-580e6334f77f","order_by":27,"name":"Sebastian Walther","email":"","orcid":"https://orcid.org/0000-0003-4026-3561","institution":"University Hospital Würzburg","correspondingAuthor":false,"prefix":"","firstName":"Sebastian","middleName":"","lastName":"Walther","suffix":""},{"id":511674197,"identity":"17d2d752-9cd2-4935-8fca-646598013acf","order_by":28,"name":"Hans Grabe","email":"","orcid":"","institution":"University Medicine Greifswald","correspondingAuthor":false,"prefix":"","firstName":"Hans","middleName":"","lastName":"Grabe","suffix":""},{"id":511674198,"identity":"f9a5e331-6804-483e-9ae1-fc17595ccc5d","order_by":29,"name":"Jürgen Deckert","email":"","orcid":"","institution":"University Hospital of Würzburg","correspondingAuthor":false,"prefix":"","firstName":"Jürgen","middleName":"","lastName":"Deckert","suffix":""},{"id":511674199,"identity":"1b7b63f3-3914-46a4-9a0a-349d3d3ac51d","order_by":30,"name":"Svenja Caspers","email":"","orcid":"","institution":"Research Centre Jülich","correspondingAuthor":false,"prefix":"","firstName":"Svenja","middleName":"","lastName":"Caspers","suffix":""},{"id":511674200,"identity":"bc877b26-9545-4658-822f-98adc9ba4303","order_by":31,"name":"Grit Hein","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Grit","middleName":"","lastName":"Hein","suffix":""},{"id":511674201,"identity":"0372fc8d-e6fb-4502-94cd-f40a8da60896","order_by":32,"name":"Angelika Ehrardt-Lehmann","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Angelika","middleName":"","lastName":"Ehrardt-Lehmann","suffix":""}],"badges":[],"createdAt":"2025-05-09 10:56:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6627834/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6627834/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91306631,"identity":"5713d159-b391-452b-a1c0-b0b4b3a1554b","added_by":"auto","created_at":"2025-09-15 06:33:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":616927,"visible":true,"origin":"","legend":"\u003cp\u003eAUROC curves for each winning model are shown for the following datasets: 246 neuroimaging variables (N), consisting of whole-brain imaging variables from the Juelich Atlas (Amunts et al., 2020); eight psychosocial variables (P) – including sex, age, lifetime cigarettes smoked, depression, stress, childhood trauma, and either panic attacks (PA) for classifying clinically relevant GAD symptoms or the GAD‑7 score for classifying PA; and the combination of both datasets (N+P; total of 254 variables). Panel A displays results for GAD symptoms, and Panel B for PA. RF: random forest, ELNET: elastic net regression.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-6627834/v1/e35ccd64a95ea49d252a1e56.png"},{"id":91307801,"identity":"b9826f3b-1f44-4b64-bae1-f6b7ccaa79ce","added_by":"auto","created_at":"2025-09-15 06:41:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":599041,"visible":true,"origin":"","legend":"\u003cp\u003eTop 10 important variables (4%) for the elastic net classification of GAD (Panel A) and panic attacks (Panel B) symptoms. SPL: superior parietal lobule of parietal lobe; IF: internal intermediate fiber masses of amygdala; MF: internal medial fiber masses of amygdala; SF: superficial amygdala; OP8: frontal operculum 8.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-6627834/v1/77683109cf0195d8714d1ae4.png"},{"id":92277050,"identity":"1459c941-85e7-4243-ad5f-d41607a063a5","added_by":"auto","created_at":"2025-09-26 15:34:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2837443,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6627834/v1/3329918b-28f1-4e1c-8e57-9ae9c2a355f7.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e\nH. J. Grabe has received travel grants and speaker honoraria from Neuraxpharm, Servier, In-dorsia, and Janssen Cilag. F. Bamberg: Speaker Bureau and unrestricted research grants Siemens Healthineers. C. L. Schlett: Speaker Bureau Siemens Healthineers and Bayer Healthcare; Research Grants Siemens Healthineers.","formattedTitle":"Multimodal phenotypic classification of generalized anxiety and panic using structural MRI data and psychosocial factors: Machine learning results from the German National Cohort (NAKO) study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAnxiety disorders (ANX) are common and seriously impairing disorders, with an estimated lifetime prevalence up to 20% (Penninx et al., 2021; Wittchen et al., 2011). Currently, diagnosis of these disorders is based primarily on clinicians\u0026rsquo; assessment (Linden, 2012). This diagnostic challenge underscores the need for objective biomarkers that can complement clinical evaluations and improve the precision of ANX diagnosis. Identifying reliable biomarkers for ANX could potentially complement clinical evaluations and treatment, by identifying patient groups early in disease progression and using the underlying biological causes of their symptoms to inform the choice of treatment (Boeke et al., 2020; Chavanne et al., 2023; Linden, 2012).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGeneralized anxiety disorder (GAD) occurs with an estimated prevalence of 4-6% and is characterized by excessive, persistent and uncontrollable anxiety and worrying that is associated with nervousness, feelings of threatening uncertainty, and somatic complaints like muscular tensions and physiological hyperarousal (Ruscio et al., 2017; Wittchen et al., 2011).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePanic attacks (PA) represent a core symptom of panic disorder (PD) according to diagnostic criteria; however, they occur frequently in all ANX, and are strongly linked to general psychopathology as a separate dimension across mental conditions (Asselmann et al., 2014). PA are defined as sudden and brief episodes of extreme anxiety as well as somatic stress symptoms, which in the case of mental disorders are inappropriate or unrealistically exaggerated compared to the target situation. Affected individuals tend to develop a fear of reexperiencing a PA that is associated with avoidance behavior and further negative behavioral changes, high distress and consequently individual burden (Craske et al., 2017; de Jonge et al., 2016). Prospective data suggests that PA constitutes a risk factor not just for the future development of any ANX, but also mood and substance use disorders (Goodwin et al., 2004). Individuals experiencing PA are at greater risk for increased persistence of mental disorders and impaired functioning, which underscores the importance of preventive treatment and early diagnosis of PA to improve long-term outcomes (Batelaan et al., 2012). Within the complex phenotypic composition of ANX, PA represent characteristic and well-defined symptoms which can be interrogated using validated scales and attributed to biological anxiety circuits (Guan \u0026amp; Cao, 2024). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBoth GAD and PA are highly comorbid with other psychiatric conditions, most notably with depression (Penninx et al., 2021). This high comorbidity can be partly explained by shared neurobiological mechanisms, e.g. by overlapping environmental and genetic risk factors, as demonstrated in recent cross-disorder genome-wide association studies for pathological anxiety and depression (Kalisch et al., 2024; Strom et al., 2024). In addition, childhood adversity is one of the environmental factors potentially influencing brain morphology and significantly increasing the prevalence of ANX and depression (Grummitt et al., 2024; Teicher et al., 2016). Therefore, it is crucial to incorporate these factors into our models to capture the complexity of anxiety-related phenotypes and improve the accuracy of classification analyses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSeveral previous studies have investigated neuroimaging data as a promising candidate for biomarker identification. Lower gray matter volumes in bilateral orbitofrontal cortex and ventrolateral prefrontal cortex have been found to correlate with the General Distress dimension of the Tri-level Model (representing transdiagnostic depression and anxiety symptoms, Cichocki et al., 2024). Furthermore, higher gray matter volume in the amygdala has repeatedly been associated with GAD (Etkin et al., 2009; Makovac et al., 2015; Schienle et al., 2011). Functional imaging studies also consistently provide evidence that patients suffering from ANX display increased reactivity in the amygdala in response to negative emotional stimuli, along with insufficient prefrontal control (Abi-Dargham et al., 2023). Unfortunately, none of the biomarkers identified to date has demonstrated a sufficiently reliable predictive value to be used clinically (Abi-Dargham et al., 2023; Boeke et al., 2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis may be partly due to limitations in previous research, including small sample sizes, heterogeneous analytical and clinical approaches, and relatively simple statistical methods that may not detect subtle associations in the large, high-dimensional data produced by neuroimaging. Machine learning algorithms, however, are particularly effective in analyzing high-dimensional data, (Bzdok \u0026amp; Meyer-Lindenberg, 2018; Janssen et al., 2018), such as neuroimaging data, to identify complex patterns associated with conditions like GAD and PA. Classifiers trained on the gray matter volume of anxiety-associated regions of interest (ROIs) in adolescent subjects achieved moderate predictive\u003csup\u003e[1]\u0026nbsp;\u003c/sup\u003evalue for anxiety-disorder diagnosis in early adulthood (Chavanne et al., 2023). A similar approach achieved robust (albeit modest) performance when classifying PD vs. healthy controls based on subcortical volumes and cortical thickness and surface area (Bruin et al., 2024).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePrevious MRI-based machine learning studies of adult and adolescent anxiety have only used moderately large samples with imaging data from several hundred individuals (Boeke et al., 2020; Bruin et al., 2024; Chavanne et al., 2023). Increasing the sample size by an order of magnitude can substantially increase the robustness of the findings by reducing the risk of overfitting (Janssen et al., 2018), which may have contributed to the failure of certain models to replicate their predictive validity when evaluated on different data sets not used during training (Boeke et al., 2020). Therefore, the current paper builds on previous research by applying machine learning techniques to classify GAD symptoms and PA, in a very large dataset that includes neuroimaging data from 26,378 adults taken from the German National Cohort Study (NAKO). To highlight the advantages of data-driven machine learning methods over traditional analyses, we first conducted conventional correlational analyses on theory-driven, preselected neuroimaging variables, accounting for known confounders. Subsequently, we performed multiple machine learning analyses (support vector machines with either a radial or linear kernel, random forests, elastic net regression), using the full set of whole-brain gray matter volume data without any regional preselection. This approach enables a more comprehensive, unbiased exploration of the brain, allowing the model to identify potential patterns that might be overlooked in region-specific analyses.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eDataset and study population\u003c/p\u003e\u003cp\u003eDemographic, psychometric and brain imaging data were taken from NAKO. This population-based prospective cohort study started in 2014 with the goal of investigating risk factors for a wide range of physical and mental chronic conditions, including depression, stress and anxiety symptoms. Baseline assessment was conducted between 2014 and 2019, with the goal of investigating risk factors for a wide range of physical and mental chronic conditions, including depression, stress and anxiety symptoms. NAKO collected biomedical and questionnaire data from 205,415 persons living in Germany aged 19–74 who were chosen at random from compulsory registries of residents in 16 regions across Germany. This study analyzed the subsample of 30,927 participants who also later completed whole-body 3T magnetic resonance imaging (Peters et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). All participants gave written informed consent, and the data transfer was approved by the Use and Access Committee of NAKO. All NAKO study documents have been approved by all responsible local ethical committees and are revised regularly and adapted as needed.\u003c/p\u003e\u003cp\u003eThe two outcomes of interest in this study were clinically meaningful GAD symptoms and lifetime PAs. GAD symptoms were assessed using the established GAD-7 scale, which consists of seven items (yielding a maximum sum score of 21) to measure symptom load over the past four weeks. Clinically meaningful GAD symptoms were defined as a GAD-7 score of ≥ 10 - indicating at least moderate anxiety - with a sensitivity and specificity of 89% and 82%, respectively, in detecting GAD (Löwe et al., 2008). Lifetime PA was defined as having experienced one or more panic attacks in the past four weeks in addition to at least one previous lifetime panic attack (Löwe et al., 2003). Because only the first part of the PHQ-Panic scale was administered, we could determine the presence of current and prior PAs, the mode of occurrence, and associated disability, but not a full diagnosis of PD.\u003c/p\u003e\u003cp\u003eThe demographic and psychometric predictors in our study included eight psychosocial variables: age, sex, number of lifetime cigarettes smoked, symptoms of GAD (GAD-7 score; only for classifying PA), PA (PHQ-Panic; only for classifying GAD symptoms), depression (current PHQ-9 score; Kroenke et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), stress (current PHQ-Stress score; Spitzer et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), and childhood trauma (childhood trauma screener, CT-S; Bernstein et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). All psychosocial variables were collected by self-administered touchscreen questionnaires during an in-person baseline examination in the study centers (Peters et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The detailed description of psychometric and emotional scales within the NAKO is available elsewhere (for GAD-7, panic and stress: Erhardt et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; for childhood trauma: Klinger-König et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; and for depression: Streit et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). After excluding participants with missing values for any of the predictors or outcomes, 26,378 participants were included in the study.\u003c/p\u003e\u003cp\u003eMRI acquisition and preprocessing\u003c/p\u003e\u003cp\u003e Structural MRI data for all participants was obtained using 3T scanners (Magnetom Skyra; Siemens Healthcare, Erlangen, Germany) across five NAKO study centers (Essen, Neubrandenburg, Berlin, Augsburg, and Heidelberg/Mannheim). Images were acquired using a T1-weighted 3D MPRAGE sequence (1.0 × 1.0 × 1.0 mm (isotropic) voxel; sagittal orientation; repetition time msec/echo time msec/inversion time msec, 2300/2.98/900; 9° flip angle; Bamberg et al., 2015). The T1 images were segmented, normalized, and smoothed.\u003c/p\u003e\u003cp\u003eAn automated quality-control pipeline was used to assess each image’s sharpness, global and local signal-to-noise ratio, maximum and average estimates for structured image noise and Nyquist ghosting levels, and geometric ratio between foreground and background. Subsequently, board-certified radiologists performed a visual rating using a three-point Likert scale that considered anatomical coverage, minimum differentiable structures, and the presence of artifacts, and excluded any images rated as ‘Poor’ (Schuppert et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe gray matter volumes of the 246 brain areas defined by the Julich-Brain Cytoarchitectonic Atlas ( \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://atlases.ebrains.eu/\u003c/span\u003e\u003cspan address=\"https://atlases.ebrains.eu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; Amunts et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) using the Software Computation Anatomy Toolbox (CAT12v8; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://neuro-jena.github.io/cat/\u003c/span\u003e\u003cspan address=\"https://neuro-jena.github.io/cat/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were extracted from those T1-weighted images that passed quality control. All resulted 246 ROIs, encompassing volumes of both cortical areas and subcortical structures including the limbic system, surface areas and mean cortical thickness values were included as variables in the neuroimaging and combined variable sets described below.\u003c/p\u003e\u003cp\u003eCorrelation analyses\u003c/p\u003e\u003cp\u003eTo investigate simple linear relationships between specific neuroimaging variables and GAD as well as PA, we first computed Pearson correlations between a pre-selected set of 93 neuroimaging variables and clinically relevant GAD symptoms and PA. This set of neuroimaging variables included global brain metrics (e.g., total gray matter and white matter volumes), regional gray matter volumes (e.g., insula, amygdala, hippocampus, and cingulate cortex), surface areas, subcortical structures, and mean cortical thickness in anxiety-related brain regions (Craske et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Harrewijn et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e; Pessoa, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Additionally, we examined network-specific gray matter volumes, such as those associated with the Default Mode and Salience Networks, as well as limbic regions implicated in emotional regulation (Alves et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Catani et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). A full list of all variables alongside their correlations can be found in the Extended Data Tables S2 and S3.\u003c/p\u003e\u003cp\u003eIn order to depict the putative influence of age and highly overlapping symptoms on the target outcome of GAD and PA, we used a step by step correction approach, using sequential linear regressions to regress out the impact of confounders on the neuroimaging variables (Snoek et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). For the correlations with GAD symptoms, the analyses were controlled for age, depressive symptoms (PHQ-9 score), PA and childhood trauma, using linear confounder regression sequentially. For the correlations with PA, the analyses were controlled for age, depressive symptoms (PHQ-9 score), clinically relevant GAD symptoms (GAD-7 ≥ 10) and childhood trauma. All correlations were tested for significance, with Bonferroni corrections applied to control for alpha inflation.\u003c/p\u003e\u003cp\u003eTo account for known sex differences in the prevalence and neurobiological correlates of GAD and PA (Jalnapurkar et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), all correlation analyses were stratified by sex.\u003c/p\u003e\u003cp\u003eMachine-learning classification\u003c/p\u003e\u003cp\u003eThe classification analysis was conducted in R, version 4.4.3 (Core R Team, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) using four models from the machine learning package \u003cem\u003ecaret\u003c/em\u003e, version 6.0–94 (Kuhn, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2008\u003c/span\u003e): support vector machines with either a radial (SVM-R) or linear kernel (SVM-L), random forests (RF), and elastic net regression (ELNET). Each model was trained in two distinct analyses to make binary predictions regarding: (a) the presence versus absence of clinically relevant GAD symptoms (defined as GAD-7 ≥ 10, labelled as GAD symptoms), and (b) the presence versus absence of combined current and lifetime PA (according to the first to items of the PHQ-Panic scale). These analyses were conducted using three different variable sets: 246 neuroimaging variables (N), 8 psychosocial variables (P; including 7 psychosocial variables sex, age, number of lifetime cigarettes, depression, stress, childhood trauma plus PA for classifying clinically relevant GAD symptoms or the GAD-7 score for classifying PA), and a combined set of 254 neuroimaging and psychosocial variables (P + N; for a similar approach, see Chavanne et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe participants were randomly divided into a training set with n = 21,102 subjects and a test set with n = 5,276 subjects. A linear confound regression was used to regress out the impact of sex, age, total intracranial volume, lifetime cigarettes smoked, childhood trauma, and scanner site on gray matter volume in the training dataset (Snoek et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We computed multiple linear regressions with these potential confounding variables for each brain area volume. Only the residuals of this regression analyses were included in the machine-learning models. To prevent data leakage – that is, the unintended use of test data information inflating model performance – the regression weights estimated from the training dataset were also applied to control for confounds in the test dataset (More et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAll models were trained using 5-fold cross-validation, using the area under the receiver operating characteristic curve (AUROC) to score model performance. Hyperparameters for each model were tuned using grid search cross-validation with a tune length of 10 for each hyperparameter. For SVMs, we adjusted the regularization parameter C, as well as the scaling parameter sigma when using a radial kernel. Random Forest tuning focused on minimum node size, number of trees, and splitting rule. For ELNET, we optimized the mixing parameter alpha and regularization parameter lambda. All other hyperparameters were left at the default values provided by caret. The best-fitting model was then retrained using the selected hyperparameter configuration on the full training dataset and subsequently evaluated on the test set.\u003c/p\u003e\u003cp\u003e Consistent with population epidemiology, only a minority of participants had GAD-7 score ≥ 10 or PA (see Results section). To prevent the imbalance of the data from skewing the classification, a subset of control participants without the target diagnosis was chosen through random undersampling (Ganganwar, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). This balancing of class sizes was performed independently in each fold of cross-validation, randomly selecting a new subsample of the majority class for each fold. This approach ensured that nearly all participants contributed to the model training, despite the undersampling of the majority class (for a similar approach, see Bruin et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). All variables were standardized to \u003cem\u003ez\u003c/em\u003e-scores.\u003c/p\u003e\u003cp\u003eAfter training, the model with the best performance (when using optimal hyperparameters) was identified for each combination of prediction target and variable set. We computed several performance measures on the confusion matrices of the best models for each data set (N, P, N + P). Additionally, we assessed variable importance in the winning N + P models by computing the AUROC for each predictor. This approach tests each predictor individually and computes the AUROC based on predictions made using only that predictor and the model. This method has the advantage of providing the predictive importance of each single predictor for each model and allows for comparisons across multiple classification models (Kuhn, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Higher values indicate stronger associations with the outcome and can be interpreted as indicating effect sizes. To further assess the direction of these associations, we computed simpler logistic regression models using the ten most important predictors. In contrast to separate Pearson correlations between individual predictors and various outcomes, these analyses test the effect of all ten predictors simultaneously.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eOf all participants, 4.49% (n\u0026thinsp;=\u0026thinsp;1,321) reported a GAD-7 score\u0026thinsp;\u0026ge;\u0026thinsp;10, indicating clinically meaningful anxiety symptoms. Additionally, 4.01% (n\u0026thinsp;=\u0026thinsp;1,182) reported experiencing PA. For an overview of all participant characteristics, see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDemographic characteristics of analyzed study participants with valid data (n\u0026thinsp;=\u0026thinsp;26,378)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eGAD symptoms\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003ePanic attacks\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eno\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eyes\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eno\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eyes\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParticipants\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25,217\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1,161\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25,376\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1,002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e26,378\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eSex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMen\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14,530 \u003c/p\u003e\u003cp\u003e(57.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e502 \u003c/p\u003e\u003cp\u003e(43.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14,619 \u003c/p\u003e\u003cp\u003e(57.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e413 \u003c/p\u003e\u003cp\u003e(41.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e15,032 (57.0%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWomen\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10,687 \u003c/p\u003e\u003cp\u003e(42.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e659 \u003c/p\u003e\u003cp\u003e(56.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e10,757 \u003c/p\u003e\u003cp\u003e(42.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e589 \u003c/p\u003e\u003cp\u003e(58.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e11,346 (43.0%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47.5 (12.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e45.8 (11.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e47.5 (12.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e46.8 (12.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e47.4 (12.2)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e18\u0026ndash;29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2,855 \u003c/p\u003e\u003cp\u003e(11.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e148 \u003c/p\u003e\u003cp\u003e(12.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2,877 \u003c/p\u003e\u003cp\u003e(11.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e126 \u003c/p\u003e\u003cp\u003e(12.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3,003 (11.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e30\u0026ndash;39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3,163 \u003c/p\u003e\u003cp\u003e(12.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e179 \u003c/p\u003e\u003cp\u003e(15.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3,211 \u003c/p\u003e\u003cp\u003e(12.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e131 \u003c/p\u003e\u003cp\u003e(13.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3,342 (12.7%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40\u0026ndash;49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7,672 \u003c/p\u003e\u003cp\u003e(30.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e353 \u003c/p\u003e\u003cp\u003e(30.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7,728 \u003c/p\u003e\u003cp\u003e(30.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e297 \u003c/p\u003e\u003cp\u003e(29.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e8,025 (30.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e50\u0026ndash;59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6,864 \u003c/p\u003e\u003cp\u003e(27.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e324 \u003c/p\u003e\u003cp\u003e(27.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6,910 \u003c/p\u003e\u003cp\u003e(27.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e278 \u003c/p\u003e\u003cp\u003e(27.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e7,188 (27.2%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e60\u0026ndash;74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4,663 \u003c/p\u003e\u003cp\u003e(18.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e157 \u003c/p\u003e\u003cp\u003e(13.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4,650 \u003c/p\u003e\u003cp\u003e(18.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e170 \u003c/p\u003e\u003cp\u003e(17.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4,820 (18.3%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGAD-7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.6 (2.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.7 (2.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.8 (2.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8.6 (4.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.0 (3.1)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eGAD-7\u0026thinsp;\u0026ge;\u0026thinsp;10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ehealthy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25,217 \u003c/p\u003e\u003cp\u003e(100.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0 \u003c/p\u003e\u003cp\u003e(0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e24,580 \u003c/p\u003e\u003cp\u003e(96.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e637 \u003c/p\u003e\u003cp\u003e(63.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e25,217 (95.6%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eanxious\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 \u003c/p\u003e\u003cp\u003e(0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1,161 \u003c/p\u003e\u003cp\u003e(100.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e796 \u003c/p\u003e\u003cp\u003e(3.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e365 \u003c/p\u003e\u003cp\u003e(36.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,161 (4.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eLifetime Panic \u003c/p\u003e\u003cp\u003eAttacks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eno\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e24,580 \u003c/p\u003e\u003cp\u003e(97.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e796 \u003c/p\u003e\u003cp\u003e(68.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25376 \u003c/p\u003e\u003cp\u003e(100.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0 \u003c/p\u003e\u003cp\u003e(0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e25,376 (96.2%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eyes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e637 \u003c/p\u003e\u003cp\u003e(2.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e365 \u003c/p\u003e\u003cp\u003e(31.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0 \u003c/p\u003e\u003cp\u003e(0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1,002 \u003c/p\u003e\u003cp\u003e(100.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,002 (3.8%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLifetime Cigarettes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.0 (192.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e10.6 (260.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6.0 (191.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11.6 (280.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6.2 (195.7)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDepressive Symptoms (PHQ-9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.3 (2.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.8 (5.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.5 (3.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9.4 (5.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.7 (3.6)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStress\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.1 (2.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e9.2 (3.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.2 (2.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e7.5 (4.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.4 (3.0)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChildhood Trauma Sum Score (CT-S)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003emean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.0 (2.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.8 (3.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7.1 (2.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8.5 (3.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e7.1 (2.5)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eBonferroni-corrected Pearson correlations between pre-selected neuroimaging variables and GAD symptoms and PA stratified by sex revealed some significant but very weak associations (\u003cem\u003ep\u003c/em\u003e-values reported as \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003eadj\u003c/em\u003e\u003c/sub\u003e). Overall, the sex-stratified correlation values of any of the neuroimaging variables with target anxiety phenotypes were extremely low, with significant correlations of .032 \u0026le; |r| \u0026le; .098 for GAD symptoms after correction for age, depressive symptoms, PA, and childhood trauma. These associations were predominantly present in the female group (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In females, categorical GAD symptoms had the highest significant correlation with total grey matter and cerebral white matter volumes, cortical gray matter volume and hemispheric surface area, as well as, more specifically, ventral diencephalon volumes (after correction for depression). Furthermore, GAD symptoms were most consistently correlated with hemispheric surface area bilaterally for both sexes.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTop correlations between clinically relevant GAD symptoms (GAD-7 \u0026ge;10) and various neuroimaging variables, corrected for age, depressive symptoms (PHQ-9), panic attacks (PHQ-Panic), and childhood trauma (dimensional CT-S), stratified by sex.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e\u003cp\u003eCorrelation with GAD-7 (cutoff)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeuroimaging variables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003efemale\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003emale\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003er\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ep_adj\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003er\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep_adj\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTotal volume\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal cerebral white matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e5.6E-06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.034\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0035\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal gray matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.061\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e5.9E-09\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.00018\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft hemispheric cerebral white matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e6.7E-06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0043\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSubcortical brain region volume\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft ventral diencephalon\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.052\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e2.6E-06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight ventral diencephalon\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.048\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e2.2E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBrainstem\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e9.5E-04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight amygdala\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.047\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e4.7E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft amygdala\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.021\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0063\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight thalamus\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.044\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e3.3E-04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.010\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft thalamus\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.040\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.025\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft hippocampus\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e5.7E-04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.045\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight hippocampus\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0040\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight accumbens area\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0082\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft accumbens area\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.018\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.041\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eCortical/subcortical gray matter volume\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight cortical gray matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.057\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e8.8E-08\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.00142\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft cortical gray matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.058\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e6.1E-08\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.036\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.00076\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSubcortical gray matter volume\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e4.5E-06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.00023\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft medial orbitofrontal cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.049\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e1.5E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0083\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft rostral anterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e6.1E-04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight rostral anterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.020\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft insula\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.049\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.034\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0033\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight insula\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.004\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.043\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e1.6E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft isthmus cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight isthmus cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.017\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSurface area\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft hemisphere\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.050\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e8.9E-06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e5.9E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight hemisphere\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.050\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e1.1E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.040\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e7.6E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft insula\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0002\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight insula\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0002\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft isthmus cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0081\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight isthmus cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0084\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft caudal anterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.043\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight caudal anterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.006\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.043\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft posterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.031\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.014\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight posterior cingulate cortex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ens\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.031\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.015\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNote. All significant correlations between clinically relevant GAD symptoms and neuroimaging variables with increasing numbers of controlled confounds can be found in the Data Table S1. Significant correlations between panic attacks and several neuroimaging variables \u0026ndash; adjusted for different numbers of regressed-out confounders \u0026ndash; are reported in the Extended Data Table S1.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ePA were statistically significantly correlated with neuroimaging variables only in the female group after controlling for age and depression. Volumes of the total grey matter, cerebral white matter, ventral diencephalon, thalamus as well as total cortical, subcortical and the right isthmus cingulate cortex volume were significantly correlated with PA. Interestingly, after correction for childhood trauma score, only the correlations with the diencephalon volume and total gray matter volume remained significant. All significant correlations were very low with .032 \u0026le; |r| \u0026le; .040 and \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003eadj.\u003c/em\u003e\u003c/sub\u003e \u0026lt; .047. In males, none of the neuroimaging variables were significantly correlated with PA after adjustment.\u003c/p\u003e\u003cp\u003eIn summary, the correlation analyses revealed statistically significant correlations of GAD symptoms and PA with a number of different neuroimaging variables. However, the correlation coefficients are extremely small and hard to interpret regarding their importance for predicting GAD and PA. For all correlations, see Extended Data, tables S2 and S3.\u003c/p\u003e\u003cp\u003eIn the next step, we used machine learning to identify the brain structures and psychosocial variables that are most important for predicting these clinically relevant outcomes. As can be seen in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the random forest classifier outperformed other P-models (i.e., models using only psychosocial variables as input data) for both GAD symptoms (AUROC\u0026thinsp;=\u0026thinsp;0.976) and PA (AUROC\u0026thinsp;=\u0026thinsp;0.933) in the test dataset. The random forest classifier performed best on the datasets containing only neuroimaging data (N-models) but still showed relatively poor performances for classifying GAD symptoms (AUROC\u0026thinsp;=\u0026thinsp;0.557) and PA (AUROC\u0026thinsp;=\u0026thinsp;0.557). ELNET showed the best performance using the combination of neuroimaging data and psychosocial variables (N\u0026thinsp;+\u0026thinsp;P) to classify GAD symptoms (AUROC\u0026thinsp;=\u0026thinsp;0.961) and PA (AUROC\u0026thinsp;=\u0026thinsp;0.876), but did not outperform random forest P-models. AUROCs for the winning models for each dataset (P, N, P\u0026thinsp;+\u0026thinsp;N) are depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eArea under receiver operating characteristic curve (AUROC) values for each model and each algorithm.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"10\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e\u003cp\u003eGAD Symptoms\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c10\" namest=\"c7\"\u003e\u003cp\u003ePanic Attacks\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003emodel\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSVM_L\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSVM_R\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eELNET\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSVM_L\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eSVM_R\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003eELNET\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.517\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.553\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003e0.557\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.528\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.546\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.520\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e\u003cem\u003e0.557\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.522\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.962\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.950\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.973\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.962\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.875\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.866\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e\u003cb\u003e0.933\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.877\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP\u0026thinsp;+\u0026thinsp;N\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.949\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.954\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.957\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.961\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.852\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.831\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.874\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cem\u003e0.876\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"10\"\u003e\u003cem\u003eNote. N: neuroimaging imaging variables; P: psychosocial and psychometric variables; SVM_L\u0026thinsp;=\u0026thinsp;linear support vector machine; SVM_R\u0026thinsp;=\u0026thinsp;radial support vector machine, RF\u0026thinsp;=\u0026thinsp;random forest, ELNET\u0026thinsp;=\u0026thinsp;elastic net regression. The algorithms resulting in the highest AUROC for each model are printed in italic type and the highest value for each outcome additionally in bold type.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eConfusion matrices for the models with the highest AUROC.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eGAD Symptoms\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003ePanic Attacks\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMetric\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003cp\u003e(RF)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP\u003c/p\u003e\u003cp\u003e(RF)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP\u0026thinsp;+\u0026thinsp;N\u003c/p\u003e\u003cp\u003e(ELNET)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003cp\u003e(RF)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eP\u003c/p\u003e\u003cp\u003e(RF)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP\u0026thinsp;+\u0026thinsp;N\u003c/p\u003e\u003cp\u003e(ELNET)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAccuracy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.499\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.896\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.905\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.544\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.802\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.846\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKappa\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.383\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.398\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.214\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.221\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAccuracyLower\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.485\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.887\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.897\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.530\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.791\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.836\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAccuracyUpper\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.512\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.904\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.913\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.557\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.813\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.856\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAccuracyNull\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.958\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.958\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.958\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.961\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.961\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.961\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAccuracyPValue\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMcnemarPValue\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSensitivity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.575\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.910\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.873\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.522\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.918\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.729\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSpecificity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.496\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.895\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.907\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.544\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.797\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.851\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePos Pred Value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.047\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.275\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.290\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.156\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.166\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeg Pred Value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.964\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.996\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.994\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.965\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.996\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.987\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.047\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.275\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.290\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.156\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.166\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.575\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.910\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.873\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.522\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.918\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.729\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eF1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.088\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.422\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.436\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.082\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.267\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.271\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrevalence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.039\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDetection Rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.024\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.037\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.020\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.036\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.029\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDetection Prevalence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.507\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.139\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.126\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.458\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.231\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.172\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBalanced Accuracy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.535\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.902\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.890\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.533\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.858\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.790\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUROC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.557\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.973\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.959\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.557\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.933\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.876\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNote.\u003c/em\u003e RF\u0026thinsp;=\u0026thinsp;random forest, ELNET\u0026thinsp;=\u0026thinsp;elastic net regression.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTo investigate which variables had the highest predictive importance for the outcome variables, we computed the AUROC importance for the P\u0026thinsp;+\u0026thinsp;N models (Kuhn, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Variables with higher AUROC importance scores are considered more predictive. We depicted the top 10 variables (4% of all variables; for similar approach, see Wei\u0026szlig; et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) separately for GAD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA) and PA (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The importance scores for all variables of all winning models are provided in the Extended Data as follows: the winning P models are detailed in S4 (GAD) and S5 (PA); the winning N models in S6 (GAD) and S7 (PA); and the winning P\u0026thinsp;+\u0026thinsp;N models in S8 (GAD) and S9 (PA).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\u003ccolgroup cols=\"1\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e[FIGURE \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e here]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn addition to AUROC, we computed performance metrics derived from the confusion matrices for the top-performing models (P, N, P\u0026thinsp;+\u0026thinsp;N), as summarized in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Interestingly, while the elastic net P\u0026thinsp;+\u0026thinsp;N models did not achieve higher AUROCs compared to the random forest P models, both P\u0026thinsp;+\u0026thinsp;N models for GAD and PA exhibited higher unbalanced accuracy and positive predictive value (i.e., the likelihood that a positive classification is a true positive, also referred to as precision) as well as higher specificity values compared to random forest P models.\u003c/p\u003e\u003cp\u003e[FIGURE \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003eThe most important psychosocial variables (P) for classifying individuals with clinically relevant symptoms of GAD were depressive symptoms (PHQ-9 score), psychological distress (dimensional PHQ-Stress), childhood trauma (dimensional CT-S), PA, female sex, and age (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The most important neuroimaging (N) variables were gray matter volume in the regions of the left internal intermediate fiber masses (IF) of the amygdala, the left area OP8 in the frontal operculum, the right superficial fiber masses (SF) of the amygdala, and the left area 7M in the superior parietal lobule.\u003c/p\u003e\u003cp\u003eFor the classification of panic attacks, the most important psychosocial variables (P) were generalized anxiety disorder symptoms (GAD-7 score), depressive symptoms (PHQ-9 score), psychological distress (dimensional PHQ-Stress), childhood trauma (dimensional CT-S), female sex, and lifetime cigarettes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The most important neuroimaging (N) variables were gray matter volume in the regions of the left internal intermediate fiber masses of the amygdala, the left area OP8 in the frontal operculum, the right internal medial fiber masses of the amygdala, and the right area 7A in the superior parietal lobule.\u003c/p\u003e\u003cp\u003eTo evaluate the direction of these associations, the results of the simple logistic regression models for outcomes and the predictor variables are depicted in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCoefficients and Odds-Ratios of the logistic regression on GAD symptoms and panic attacks with top 10 variables of the winning P\u0026thinsp;+\u0026thinsp;N model.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e\u003cp\u003eGAD symptoms\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIndependent variable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ez\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOR\u003c/p\u003e\u003cp\u003e[95%CI]\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDepression (PHQ)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1.34\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e33.44\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e3.84\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[3.55\u003c/b\u003e,\u003c/p\u003e\u003cp\u003e\u003cb\u003e4.15]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStress\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.64\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e16.03\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.90\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.75, 2.05]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChildhood Trauma\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.813\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.01\u003c/p\u003e\u003cp\u003e[0.94, 1.08]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePanic Attacks\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1.08\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.11\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e9.71\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e2.93\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[2.36, 3.64]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.686\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.02\u003c/p\u003e\u003cp\u003e[0.93, 1.12]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft IF (Amygdala)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.080\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.08\u003c/p\u003e\u003cp\u003e[0.99, 1.17]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.212\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.94\u003c/p\u003e\u003cp\u003e[0.86, 1.04]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft Area OP8 (Frontal Operculum)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.221\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.94\u003c/p\u003e\u003cp\u003e[0.84, 1.04]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight SF (Amygdala)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e-0.11\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e-1.98\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.048\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.9\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[0.81, 1.00]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft Area 7M (SPL)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.16\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e3.05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.18\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.06, 1.31]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePanic Attacks\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eIndependent variable\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eEstimate\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eSE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003ez\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003ep\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eOR\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[95%CI]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGAD\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.76\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e18.87\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e2.15\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.98, 2.33]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDepression (PHQ)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.13\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e3.25\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.14\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.05, 1.23]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStress\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.24\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e6.21\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.27\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.18, 1.36]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChildhood Trauma\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.12\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.03\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e4.01\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.12 [1.06, 1.19]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.10\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e2.56\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.010\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.11\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.03, 1.20]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft IF (Amygdala)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.635\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.98 \u003c/p\u003e\u003cp\u003e[0.91, 1.06]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft Area OP8 (Frontal Operculum)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.92\u003c/p\u003e\u003cp\u003e[0.85, 1.00]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight MF (Amygdala)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.08\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.04\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e2.06\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.040\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e1.08\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e[1.00, 1.65]\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoking (Lifetime Cigarettes)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.542\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.98\u003c/p\u003e\u003cp\u003e[0.92, 1.03]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight Area 7A (SPL)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.076\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003cp\u003e[0.85, 1.01]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003eNote. Significant predictors in bold. Gray matter volume data are confound regressed for sex, age, total intracranial volume, lifetime cigarettes smoked, childhood trauma, and scanner site. SPL\u0026thinsp;=\u0026thinsp;superior parietal lobule, SE\u0026thinsp;=\u0026thinsp;standard error of the mean, OR\u0026thinsp;=\u0026thinsp;odds ratio, CI\u0026thinsp;=\u0026thinsp;confidence interval.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn the present study, we employed multiple machine learning models to model clinically relevant symptoms of GAD and PA in a sample of nearly 26,378 individuals. To classify individuals with or without these symptoms, we utilized regional gray matter volume of a whole-brain parcellation, psychosocial variables (age, sex, number of lifetime cigarettes, symptoms of GAD, PA, depression, stress, and childhood trauma), or a combination of both. We found that a random forest model using only psychosocial variables achieved the highest AUROC for classifying both individuals with GAD and PA. ELNET models combining brain volume data and psychosocial variables demonstrated the highest unbalanced accuracy. Depressive symptoms, childhood trauma, psychological distress, and female sex were the psychosocial variables with the highest predictive importance for classifying clinically relevant GAD symptoms and PA. Current and lifetime PA were among the most important variables for classifying GAD symptoms, while GAD symptoms were the most important statistical predictor for classifying PA. Among brain regions, amygdala volume, areas in the frontal operculum, and the superior parietal lobule showed the highest importance for both symptom categories. Generally, models using only psychosocial variables (such as other psychiatric symptoms, psychological stressors, and demographic variables) demonstrated the best-balanced performance in classifying clinically relevant GAD symptoms (Random Forest AUROC\u0026thinsp;=\u0026thinsp;0.973) and PA (Random Forest AUROC\u0026thinsp;=\u0026thinsp;0.933). Gray matter volume data did not substantially improve predictive quality and even led to slightly worse performances (ELNET AUROC\u0026thinsp;=\u0026thinsp;0.959 for GAD symptoms and 0.876 for PA, respectively). These results align with a recent study by Chavanne et al. (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), who predicted the onset of ANX in adolescents using gray matter volume and psychometric data. In their study, psychometric variables alone had sufficient predictive power, and performance sometimes declined when brain volume data was included. Furthermore, our gray matter volume data alone showed only poor classification performance of GAD (Random Forest AUROC\u0026thinsp;=\u0026thinsp;0.557) or PA (Random Forest AUROC\u0026thinsp;=\u0026thinsp;0.557), a finding in line with previous other machine learning studies (Bruin et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Chavanne et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Harrewijn et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e; Hill et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, adding brain matter volume to psychosocial variables increased the unbalanced accuracy and specificity. This indicates that the ELNET model with a combination of gray matter volume and psychosocial variables might be more reliable in detecting negative cases, i.e. individuals without GAD or PA, than the model without gray matter volume. Thus, our findings align with those recently reported in the ENIGMA mega-analyses. In one study, Hilbert et al. (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) highlighted distinct cortical and subcortical structural changes in specific phobia, with pronounced differences based on subtype. Similarly, our results underscore the importance of subcortical volumes in anxiety-related psychopathology. Furthermore, Groenewold et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) emphasize the role of subcortical structures, such as the putamen and pallidum, in ANX, consistent with the involvement of neural circuits important for fear and emotional regulation observed in our data. Of note, our simple correlation analyses indicated significant associations between pre-selected neuroimaging variables and GAD symptoms and PA; however, the correlation values were extremely low and thus not meaningful in terms of clinical application at this stage. Together, these findings reinforce the necessity of integrating neuroanatomical and psychosocial dimensions for a comprehensive understanding of ANX.\u003c/p\u003e\u003cp\u003eIn our case, variable importance analyses revealed that psychosocial characteristics generally held greater predictive power than variations in gray matter volume. Depressive symptoms were identified as the most important predictor for clinically relevant GAD symptoms and the second most important predictor for PA. This finding aligns with existing research showing that anxiety disorders are frequently followed by depression (Solmi et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ter Meulen et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, there is substantial comorbidity between GAD and depression (with 45\u0026ndash;98% of GAD patients suffering from both conditions, Noyes, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), and between PD \u0026ndash; defined as the disorder with PA as a core symptom \u0026ndash; and major depression (with about 50% of PD patients suffering from a depressive episode, Gorman \u0026amp; Coplan, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). Additionally, symptom overlap between GAD and depression as assessed with GAD-7 and PHQ-9 might contribute to relevant predictive power of depression for GAD symptoms. The most important predictors for classifying PA were symptoms of GAD. Conversely, lifetime PA ranked as the fourth most important variable for classifying symptoms of GAD. This relationship can, again, be explained by the high degree of comorbidity between these syndromes and shared neurobiological factors (Noyes, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2001\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eInterestingly, childhood trauma (i.e., experiences of abuse and neglect) emerged as a relatively important factor for classifying symptoms of both GAD and PA. This aligns with previous research demonstrating that childhood trauma is a substantial risk factor for both syndromes (Hovens et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Klinger-K\u0026ouml;nig et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kuzminskaite et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). It should be noted, however, that while the relatively complex ELNET classifier trained on the combination of psychosocial and neuroimaging data found associations with both GAD and PA, the simple logistic regression models only demonstrated a significant association with PA.\u003c/p\u003e\u003cp\u003eAdditionally, female sex proved to be a relatively important variable for machine-learning model performance for both conditions. This finding is consistent with a substantial body of research showing that females are affected by anxiety and depressive disorders more frequently than males, and that machine-learning analyses account for sex-related differences better than simple regression (Jalnapurkar et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Kessler, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Leach et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Lower age showed relatively modest but notable importance for classifying GAD symptoms in the elastic net regression \u0026ndash; although not statistically significant, the effect was descriptively in the same direction as in the logistic regressions. This finding could be explained by the typical age at onset of GAD, which is usually during middle/late adulthood (McGrath et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Solmi et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), and there is evidence that the intensity and number of symptoms might decrease with increasing age (Le Roux et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Miloyan et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Additionally, the lifetime number of cigarettes smoked showed a subtle relationship with PA; however, this relationship was not significant in the simpler logistic regression. While studies suggest that smoking might be a risk factor for developing panic disorder and ANX in general (Breslau \u0026amp; Klein, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Ter Meulen et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zvolensky et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), it is also possible that smoking develops as a coping mechanism for PA (Lavallee et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In line with this, one recent systematic review on the prospective bidirectional association of smoking and anxiety pathology showed inconsistent results (Fluharty et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe amygdala displays a complex anatomical and cellular organization with several subregions having distinct functions and brain connections (Janak \u0026amp; Tye, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The individual internal fiber masses consist of medial (MF) and intermediate (IF) fiber bundles and separate the amygdaloid substructures (Kedo et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The IF and MF are attached to the ventromedial part of the basomedial nucleus (BMA) of the laterobasal complex of the amygdala. The BMA has been shown to be the major target of ventral medial prefrontal cortex projections (mPFC), which constitute one top-down regulation pathway to control anxiety states (Adhikari et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Additionally, previous work implies the existence of a BMA microcircuit involved in fear acquisition, generalization, recall, and extinction in a distinct and sex-dependent manner (Rajbhandari et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In our study, higher gray matter volume in the left amygdala IF was associated with clinically relevant GAD symptoms (although this only was detected by the machine-learning models, not the logistic regressions), while lower gray matter volume was linked to PA. These differences may reflect distinct clinical profiles between GAD and PA, with potentially higher mPFC involvement in GAD (Via et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, as the specific functional implications of the IF have not been evaluated so far, further studies are needed to clarify its utility as a phenotypic marker for different ANX.\u003c/p\u003e\u003cp\u003eThe superficial amygdala (SF) is located close to the hippocampal formation. Lower grey matter volume of the SF was associated with GAD symptoms. Studies investigating reactivity to emotional faces and activation of SF suggest its implication in social relevance processing and also fear-related memory (Sedwick VM et al., 2022; Zhu Xiao et al., 2023). In general, previous research has yielded mixed findings regarding the relationship between amygdala gray matter volume and GAD. Some studies report increased amygdala volumes, while others show decreased volumes compared to healthy controls (Schienle et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; for a review, see Madonna et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). A recent study suggests that anomalies in amygdala volume among GAD patients may be driven by comorbid ANX, as individuals with GAD without comorbidities may not exhibit amygdala volume differences compared to healthy controls (Suor et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Thus, the relationship between amygdala gray matter volume anomalies and GAD may be complex and warrants further investigation in future research. Previous research has shown that smaller amygdala volume was associated with panic disorder (Hayano et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; for a review, see Wang et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), which is in line with our findings of smaller gray matter volume in the left amygdala being an important classification variable for PA. However, we also found \u003cem\u003ehigher\u003c/em\u003e gray matter volume of the right MF amygdala to be associated with PA. These findings are in contrast to previous research and need further investigation. Given they have different locations within the amygdala, the IF and MF might have individual implications in anxiety-related circuits, which need to be evaluated in further studies. So far, the evidence on volumetric associations with panic (and GAD) is mostly based on total amygdala grey matter volume measurements; however, gray matter volume of subnuclei and fiber masses might more precisely reflect underlying anxiety-related pathological processes, e.g. functional adaptations in response to specific patterns of neural activity inside the amygdala subdivisions or in connection with other brain regions such as the mPFC. Using a detailed microstructural anatomical atlas defining relevant brain areas, like the Julich-Brain Atlas used in this study, could assist in further specifying the differential roles of specific substructures in GAD and panic attacks. In terms of laterality, sex might play a role, as a recent study reported clear sex-differences in the left and right amygdala with differential sex-specific genetic correlation only in the left amygdala (Gui et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWe also found that reduced gray matter volume in the frontal operculum was associated with both GAD and PA, although these relationships were only found with the complex machine-learning analyses and were not significant in the simple logistic regressions. Area OP8 is located in the frontal part of the frontal operculum, the latter being part of a network controlling activity in other brain areas during performance of cognitive tasks in the sense of cognitive control (Higo et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Previous research has shown that the operculum is associated with negative cognition and worrying (Makovac et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Excessive worry is the core symptom of GAD and worrying about future PA is an important symptom of panic disorder (\u003cem\u003eDiagnostic and Statistical Manual of Mental Disorders\u003c/em\u003e, 2013). Thus, structural anomalies in the frontal operculum might be an underlying factor or a consequence of anxious ruminations within ANX. Last, we found that larger gray matter volume of the left area of the superior parietal lobule (SPL) was associated with GAD symptoms and smaller gray matter volume of the right area of the SPL was associated with PA. Again, these relationships only emerged in the machine-learning model and were not significant in the logistic regression models. There has been some research showing that ANX might be associated with anomalies and activity changes in superior parietal regions (Wang et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Jin et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In summary, though there was limited contribution of the neuroimaging variables to the total prediction of GAD symptoms and PA, the analyses detected brain areas clearly involved in anxiety-related circuits, suggesting that neuroanatomical structures could potentially have an additive value in biomarker panels as an imaging-based objective measurement and indication of a present pathological anxiety phenotype.\u003c/p\u003e\u003cp\u003eFinally, we would like to point out some limitations and potential directions for future research. First, although the sample size was significantly larger than many previous studies, the population was still limited to an epidemiological cohort consisting of residents of Germany, which may affect the generalizability of the results to other geographical populations and to clinical populations with diagnosed ANX displaying higher symptom severity and potentially higher degrees of neuroanatomical changes. Consequently, we might underestimate the contribution of structural imaging features to GAD and PA classification, and a higher grade of anxiety symptomatology might improve the overall performance of neuroanatomical variables in complex models. Moreover, future studies should aim to replicate these findings in more ethnically, clinically and geographically diverse samples to better understand cross-ethnical differences in the neurobiological underpinnings of GAD and PA. Furthermore, the low prevalence of GAD and PA may have limited the predictive validity in subgroup analyses. Second, the study was based on cross-sectional data, which limits the ability to draw causal conclusions and to make temporal predictions. Longitudinal studies that track brain structure and symptom development over time are needed to assess how structural changes in the brain might influence or be influenced by the progression of ANX. Exploring these additional factors in future research could lead to a more comprehensive understanding of the neurological correlates of GAD and panic-related pathology. Third, despite the use of machine learning algorithms, the predictive power of neuroimaging data alone was relatively low compared to psychosocial variables. This suggests that structural MRI using the imaging protocol, gray/white matter contrast and spatial resolution used in the NAKO study may not fully capture the complexity of ANX. Functional neuroimaging or multimodal approaches, including genetics, may provide additional layers of information that could improve predictive accuracy. Along these lines, we must address the fact that some of the most important features identified by the machine learning analyses did not show significant relationships with the outcome variables when examined using a conventional logistic regression analysis. While this makes interpretation of the results somewhat more challenging, it also highlights the advantages of machine learning approaches in detecting associations that would be difficult to uncover through conventional methods. The applied machine learning approaches can accommodate numerous multicollinear predictors, a known limitation of traditional analyses. We therefore interpret this as a clear strength of our methodology (Altelbany, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eWe applied machine learning techniques to a large cohort dataset to classify the presence vs. absence of GAD and PA. The results indicate that psychosocial variables, including symptoms of depression, childhood trauma, and psychological distress are more predictive of GAD and PA than neuroimaging measures of gray matter volume. While the integration of neuroimaging data improved specificity for GAD and PA, it did not significantly enhance overall predictive performance. These findings underscore the importance of considering psychological and demographic factors when developing predictive models for ANX, while also highlighting the potential limitations of relying solely on state-of-the-art structural MRI data. We advocate multimodal approaches and longitudinal designs to better understand the dynamic interplay between neurobiological factors and clinical symptoms, ultimately advancing the identification of objective and measurable biomarkers for ANX.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are part of the German National Cohort (NAKO Gesundheitsstudie). NAKO data are subject to the EU / EEA General Data Protection Regulation and to the NAKO Terms of Use. They can therefore not be deposited in a public repository. Qualified researchers affiliated to EU or EEA institutions may apply for access through the NAKO TransferHub (https://transfer.nako.de). Applications are evaluated by the NAKO Use \u0026amp; Access Committee; successful applicants must sign a Data-Use Agreement and cover associated handling fees.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e\u003cstrong\u003eExtended Data\u003c/strong\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eExtended Data is available at https://osf.io/wt9yf/?view_only=95484d1d3363458eb7d3cd127cb35d78\u003c/p\u003e\n\u003cp\u003eAll significant correlations between the preselected neural variables and GAD or panic attacks (PA) are presented in Extended Data Table S1. Table S2 provides all correlations between neural variables and GAD, while Table S3 lists those for PA. Variable importance scores from models using only psychosocial predictors are reported in Table S4 (for GAD) and Table S5 (for PA). For models using only neuroimaging variables, importance scores are shown in Table S6 (GAD) and Table S7 (PA). Finally, importance scores for models combining psychosocial and neural variables are presented in Table S8 for GAD and Table S9 for PA.\u003c/p\u003e\n\u003cp\u003eAnalysis code is available at https://osf.io/wt9yf/?view_only=95484d1d3363458eb7d3cd127cb35d78\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis project was conducted with data (Application No. NAKO-689) from the German National Cohort (NAKO) (www.nako.de). The NAKO is funded by the Federal Ministry of Education and Research (BMBF) [project funding reference numbers: 01ER1301A/B/C, 01ER1511D, 01ER1801A/B/C/D and 01ER2301A/B/C], DZPG (German Centre for Mental Health Research) and by the BMBF (German Ministry of Education and Research) grant 01EE2303E, federal states of Germany and the Helmholtz Association, the participating universities and the institutes of the Leibniz Association.\u003c/p\u003e\n\u003cp\u003eWe thank all participants who took part in the NAKO study. We also thank the staff at the NAKO study centres, the data management and integration centre, and the NAKO head office who enabled the study completion and made the collection of all data possible.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatement of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJ. Gutzeit, M. Wei\u0026szlig;, T. Kuhn, J. Deckert, G. Hein, A. Erhardt-Lehmann, L. Krist, C. Jockwitz, B. Brandes, M.N. Wright, C.M. Friedrich, M. Woeckel, R. Mikolajczyk, T. Keil, S. Castell, P. Betker, M. Rietschel, K. Berger, S. Caspers, M.F. Leitzmann, P. Bohmann, K. Wirkner, Y. Wang, J. Klinger-K\u0026ouml;nig, F. Streit, M. Rietschel, and M. Wei\u0026szlig; declare that they have no conflicts of interest.\u003c/p\u003e\n\u003cp\u003eH. J. Grabe has received travel grants and speaker honoraria from Neuraxpharm, Servier, Indorsia, and Janssen Cilag. F. Bamberg: \u0026nbsp; Speaker Bureau and unrestricted research grants Siemens Healthineers. C. L. Schlett: Speaker Bureau Siemens Healthineers and Bayer Healthcare; Research Grants Siemens Healthineers.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbi-Dargham, A., Moeller, S. J., Ali, F., DeLorenzo, C., Domschke, K., Horga, G., Jutla, A., Kotov, R., Paulus, M. P., Rubio, J. M., Sanacora, G., Veenstra-VanderWeele, J., \u0026amp; Krystal, J. H. (2023). Candidate biomarkers in psychiatric disorders: State of the field. \u003cem\u003eWorld Psychiatry : Official Journal of the World Psychiatric Association (WPA\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(2), 236\u0026ndash;262. https://doi.org/10.1002/wps.21078\u003c/li\u003e\n\u003cli\u003eAdhikari, A., Lerner, T. N., Finkelstein, J., Pak, S., Jennings, J. H., Davidson, T. J., Ferenczi, E., Gunaydin, L. A., Mirzabekov, J. J., Ye, L., Kim, S.-Y., Lei, A., \u0026amp; Deisseroth, K. (2015). Basomedial amygdala mediates top-down control of anxiety and fear. \u003cem\u003eNature\u003c/em\u003e, \u003cem\u003e527\u003c/em\u003e(7577), 179\u0026ndash;185. https://doi.org/10.1038/nature15698\u003c/li\u003e\n\u003cli\u003eAltelbany, S. (2021). Evaluation of Ridge, Elastic Net and Lasso Regression Methods in Precedence of Multicollinearity Problem: A Simulation Study. \u003cem\u003eJournal of Applied Economics and Business Studies\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), Article 1. https://doi.org/10.34260/jaebs.517\u003c/li\u003e\n\u003cli\u003eAlves, P. N., Foulon, C., Karolis, V., Bzdok, D., Margulies, D. S., Volle, E., \u0026amp; Thiebaut de Schotten, M. (2019). An improved neuroanatomical model of the default-mode network reconciles previous neuroimaging and neuropathological findings. \u003cem\u003eCommunications Biology\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 1\u0026ndash;14. https://doi.org/10.1038/s42003-019-0611-3\u003c/li\u003e\n\u003cli\u003eAmunts, K., Mohlberg, H., Bludau, S., \u0026amp; Zilles, K. (2020). Julich-Brain: A 3D probabilistic atlas of the human brain\u0026rsquo;s cytoarchitecture. \u003cem\u003eScience\u003c/em\u003e, \u003cem\u003e369\u003c/em\u003e(6506), 988\u0026ndash;992. https://doi.org/10.1126/science.abb4588\u003c/li\u003e\n\u003cli\u003eAsselmann, E., Wittchen, H.-U., Lieb, R., H\u0026ouml;fler, M., \u0026amp; Beesdo-Baum, K. (2014). Associations of fearful spells and panic attacks with incident anxiety, depressive, and substance use disorders: A 10-year prospective-longitudinal community study of adolescents and young adults. \u003cem\u003eJournal of Psychiatric Research\u003c/em\u003e, \u003cem\u003e55\u003c/em\u003e, 8\u0026ndash;14. https://doi.org/10.1016/j.jpsychires.2014.04.001\u003c/li\u003e\n\u003cli\u003eBamberg, F., Kauczor, H. U., Weckbach, S., Schlett, C. L., Forsting, M., Ladd, S. C., Greiser, K. H., Weber, M. A., Schulz-Menger, J., Niendorf, T., Pischon, T., Caspers, S., Amunts, K., Berger, K., B\u0026uuml;low, R., Hosten, N., Hegenscheid, K., Kr\u0026ouml;ncke, T., Linseisen, J., \u0026hellip; ... (2015). German National Cohort MRI Study Investigators. \u003cem\u003eRadiology\u003c/em\u003e, \u003cem\u003e277\u003c/em\u003e(1), 206\u0026ndash;220. https://doi.org/10.1148/radiol.2015142272\u003c/li\u003e\n\u003cli\u003eBatelaan, N. M., Rhebergen, D., de Graaf, R., Spijker, J., Beekman, A. T. F., \u0026amp; Penninx, B. W. J. H. (2012). Panic attacks as a dimension of psychopathology: Evidence for associations with onset and course of mental disorders and level of functioning. \u003cem\u003eThe Journal of Clinical Psychiatry\u003c/em\u003e, \u003cem\u003e73\u003c/em\u003e(9), 1195\u0026ndash;1202. https://doi.org/10.4088/JCP.12m07743\u003c/li\u003e\n\u003cli\u003eBernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., Stokes, J., Handelsman, L., Medrano, M., Desmond, D., \u0026amp; Zule, W. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. \u003cem\u003eChild Abuse \u0026amp; Neglect\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(2), 169\u0026ndash;190. https://doi.org/10.1016/s0145-2134(02)00541-0\u003c/li\u003e\n\u003cli\u003eBoeke, E. A., Holmes, A. J., \u0026amp; Phelps, E. A. (2020). Toward Robust Anxiety Biomarkers: A Machine Learning Approach in a Large-Scale Sample. \u003cem\u003eBiological Psychiatry. Cognitive Neuroscience and Neuroimaging\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(8), 799\u0026ndash;807. https://doi.org/10.1016/j.bpsc.2019.05.018\u003c/li\u003e\n\u003cli\u003eBreslau, N., \u0026amp; Klein, D. F. (1999). Smoking and Panic Attacks: An Epidemiologic Investigation. \u003cem\u003eArchives of General Psychiatry\u003c/em\u003e, \u003cem\u003e56\u003c/em\u003e(12), 1141\u0026ndash;1147. https://doi.org/10.1001/archpsyc.56.12.1141\u003c/li\u003e\n\u003cli\u003eBruin, W. B., Zhutovsky, P., van Wingen, G. A., Bas-Hoogendam, J. M., Groenewold, N. A., Hilbert, K., Winkler, A. M., Zugman, A., Agosta, F., \u0026Aring;hs, F., Andreescu, C., Antonacci, C., Asami, T., Assaf, M., Barber, J. P., Bauer, J., Bavdekar, S. Y., Beesdo-Baum, K., Benedetti, F., \u0026hellip; Aghajani, M. (2024). Brain-based classification of youth with anxiety disorders: Transdiagnostic examinations within the ENIGMA-Anxiety database using machine learning. \u003cem\u003eNature Mental Health\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 104\u0026ndash;118. https://doi.org/10.1038/s44220-023-00173-2\u003c/li\u003e\n\u003cli\u003eBzdok, D., \u0026amp; Meyer-Lindenberg, A. (2018). Machine Learning for Precision Psychiatry: Opportunities and Challenges. \u003cem\u003eBiological Psychiatry. Cognitive Neuroscience and Neuroimaging\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(3), 223\u0026ndash;230. https://doi.org/10.1016/j.bpsc.2017.11.007\u003c/li\u003e\n\u003cli\u003eCatani, M., Dell\u0026rsquo;Acqua, F., \u0026amp; Thiebaut de Schotten, M. (2013). A revised limbic system model for memory, emotion and behaviour. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(8), 1724\u0026ndash;1737. https://doi.org/10.1016/j.neubiorev.2013.07.001\u003c/li\u003e\n\u003cli\u003eChavanne, A. V., Paill\u0026egrave;re Martinot, M. L., Penttil\u0026auml;, J., Grimmer, Y., Conrod, P., Stringaris, A., van Noort, B., Isensee, C., Becker, A., Banaschewski, T., Bokde, A. L. W., Desrivi\u0026egrave;res, S., Flor, H., Grigis, A., Garavan, H., Gowland, P., Heinz, A., Br\u0026uuml;hl, R., Nees, F., \u0026hellip; Artiges, E. (2023). Anxiety onset in adolescents: A machine-learning prediction. \u003cem\u003eMolecular Psychiatry\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(2), 639\u0026ndash;646. https://doi.org/10.1038/s41380-022-01840-z\u003c/li\u003e\n\u003cli\u003eCichocki, A. C., Zinbarg, R. E., Craske, M. G., Chat, I. K., Young, K. S., Bookheimer, S. Y., \u0026amp; Nusslock, R. (2024). Transdiagnostic symptom of depression and anxiety associated with reduced gray matter volume in prefrontal cortex. \u003cem\u003ePsychiatry Research. Neuroimaging\u003c/em\u003e, \u003cem\u003e339\u003c/em\u003e, 111791. https://doi.org/10.1016/j.pscychresns.2024.111791\u003c/li\u003e\n\u003cli\u003eCore R Team. (2021). A Language and Environment for Statistical Computing. In \u003cem\u003eR Foundation for Statistical Computing\u003c/em\u003e (Bd. 2, S. https://www.R--project.org) [Software]. R Foundation for Statistical Computing. http://www.r-project.org\u003c/li\u003e\n\u003cli\u003eCraske, M. G., Stein, M. B., Eley, T. C., Milad, M. R., Holmes, A., Rapee, R. M., \u0026amp; Wittchen, H.-U. (2017). Anxiety disorders. \u003cem\u003eNature Reviews Disease Primers\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 1\u0026ndash;19. https://doi.org/10.1038/nrdp.2017.24\u003c/li\u003e\n\u003cli\u003ede Jonge, P., Roest, A. M., Lim, C. C. W., Florescu, S. E., Bromet, E. J., Stein, D. J., Harris, M., Nakov, V., Caldas-de-Almeida, J. M., Levinson, D., Al-Hamzawi, A. O., Haro, J. M., Viana, M. C., Borges, G., O\u0026rsquo;Neill, S., de Girolamo, G., Demyttenaere, K., Gureje, O., Iwata, N., \u0026hellip; Scott, K. M. (2016). Cross-national epidemiology of panic disorder and panic attacks in the world mental health surveys. \u003cem\u003eDepression and Anxiety\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(12), 1155\u0026ndash;1177. https://doi.org/10.1002/da.22572\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eDiagnostic and statistical manual of mental disorders: DSM-5\u003c/em\u003e (5th ed). (2013). American psychiatric association.\u003c/li\u003e\n\u003cli\u003eErhardt, A., Gelbrich, G., Klinger-K\u0026ouml;nig, J., Streit, F., Kleineidam, L., Riedel-Heller, S. G., NAKO Investigators, Schmidt, B., Schmiedek, F., Wagner, M., Grabe, H. J., Rietschel, M., Berger, K., \u0026amp; Deckert, J. (2023). Generalised anxiety and panic symptoms in the German National Cohort (NAKO). \u003cem\u003eThe World Journal of Biological Psychiatry: The Official Journal of the World Federation of Societies of Biological Psychiatry\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(10), 881\u0026ndash;896. https://doi.org/10.1080/15622975.2021.2011409\u003c/li\u003e\n\u003cli\u003eEtkin, A., Prater, K. E., Schatzberg, A. F., Menon, V., \u0026amp; Greicius, M. D. (2009). Disrupted amygdalar subregion functional connectivity and evidence of a compensatory network in generalized anxiety disorder. \u003cem\u003eArchives of General Psychiatry\u003c/em\u003e, \u003cem\u003e66\u003c/em\u003e(12), 1361\u0026ndash;1372. https://doi.org/10.1001/archgenpsychiatry.2009.104\u003c/li\u003e\n\u003cli\u003eFluharty, M., Taylor, A. E., Grabski, M., \u0026amp; Munaf\u0026ograve;, M. R. (2017). The Association of Cigarette Smoking With Depression and Anxiety: A Systematic Review. \u003cem\u003eNicotine \u0026amp; Tobacco Research\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(1), 3\u0026ndash;13. https://doi.org/10.1093/ntr/ntw140\u003c/li\u003e\n\u003cli\u003eGanganwar, V. (2012). \u003cem\u003eAn overview of classification algorithms for imbalanced datasets\u003c/em\u003e. \u003cem\u003e2\u003c/em\u003e(4).\u003c/li\u003e\n\u003cli\u003eGoodwin, R. D., Lieb, R., Hoefler, M., Pfister, H., Bittner, A., Beesdo, K., \u0026amp; Wittchen, H.-U. (2004). Panic attack as a risk factor for severe psychopathology. \u003cem\u003eThe American Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e161\u003c/em\u003e(12), 2207\u0026ndash;2214. https://doi.org/10.1176/appi.ajp.161.12.2207\u003c/li\u003e\n\u003cli\u003eGorman, J. M., \u0026amp; Coplan, J. D. (1996). Comorbidity of depression and panic disorder. \u003cem\u003eThe Journal of Clinical Psychiatry\u003c/em\u003e, \u003cem\u003e57 Suppl 10\u003c/em\u003e, 34\u0026ndash;41; discussion 42-43.\u003c/li\u003e\n\u003cli\u003eGroenewold, N. A., Bas-Hoogendam, J. M., Amod, A. R., Laansma, M. A., Van Velzen, L. S., Aghajani, M., Hilbert, K., Oh, H., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, J. R., Blair, K. S., Hirsch, J., Pantazatos, S. P., Schneier, F. R., Talati, A., Roelofs, K., \u0026hellip; Van Der Wee, N. J. A. (2023). Volume of subcortical brain regions in social anxiety disorder: Mega-analytic results from 37 samples in the ENIGMA-Anxiety Working Group. \u003cem\u003eMolecular Psychiatry\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(3), 1079\u0026ndash;1089. https://doi.org/10.1038/s41380-022-01933-9\u003c/li\u003e\n\u003cli\u003eGrummitt, L., Baldwin, J. R., Lafoa\u0026rsquo;i, J., Keyes, K. M., \u0026amp; Barrett, E. L. (2024). Burden of Mental Disorders and Suicide Attributable to Childhood Maltreatment. \u003cem\u003eJAMA Psychiatry\u003c/em\u003e, \u003cem\u003e81\u003c/em\u003e(8), 782\u0026ndash;788. https://doi.org/10.1001/jamapsychiatry.2024.0804\u003c/li\u003e\n\u003cli\u003eGuan, X., \u0026amp; Cao, P. (2024). Brain Mechanisms Underlying Panic Attack and Panic Disorder. \u003cem\u003eNeuroscience Bulletin\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(6), 795\u0026ndash;814. https://doi.org/10.1007/s12264-023-01088-9\u003c/li\u003e\n\u003cli\u003eGui, Y., Zhou, G., Cui, S., Li, H., Lu, H., \u0026amp; Zhao, H. (2025). The left amygdala is genetically sexually-dimorphic: Multi-omics analysis of structural MRI volumes. \u003cem\u003eTranslational Psychiatry\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(1), 1\u0026ndash;7. https://doi.org/10.1038/s41398-025-03223-8\u003c/li\u003e\n\u003cli\u003eHarrewijn, A., Cardinale, E. M., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Hilbert, K., Cardoner, N., Porta-Caster\u0026agrave;s, D., Gosnell, S., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, K. S., Blair, J. R., Hammoud, M. Z., Milad, M. R., Burkhouse, K. L., Phan, K. L., \u0026hellip; Pine, D. S. (2021a). Cortical and subcortical brain structure in generalized anxiety disorder: Findings from 28 research sites in the ENIGMA-Anxiety Working Group. \u003cem\u003eTranslational Psychiatry\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 1\u0026ndash;15. https://doi.org/10.1038/s41398-021-01622-1\u003c/li\u003e\n\u003cli\u003eHarrewijn, A., Cardinale, E. M., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Hilbert, K., Cardoner, N., Porta-Caster\u0026agrave;s, D., Gosnell, S., Salas, R., Jackowski, A. P., Pan, P. M., Salum, G. A., Blair, K. S., Blair, J. R., Hammoud, M. Z., Milad, M. R., Burkhouse, K. L., Phan, K. L., \u0026hellip; Pine, D. S. (2021b). Cortical and subcortical brain structure in generalized anxiety disorder: Findings from 28 research sites in the ENIGMA-Anxiety Working Group. \u003cem\u003eTranslational Psychiatry\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), Article 1. https://doi.org/10.1038/s41398-021-01622-1\u003c/li\u003e\n\u003cli\u003eHayano, F., Nakamura, M., Asami, T., Uehara, K., Yoshida, T., Roppongi, T., Otsuka, T., Inoue, T., \u0026amp; Hirayasu, Y. (2009). Smaller amygdala is associated with anxiety in patients with panic disorder. \u003cem\u003ePsychiatry and Clinical Neurosciences\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e(3), 266\u0026ndash;276. https://doi.org/10.1111/j.1440-1819.2009.01960.x\u003c/li\u003e\n\u003cli\u003eHigo, T., Mars, R. B., Boorman, E. D., Buch, E. R., \u0026amp; Rushworth, M. F. S. (2011). Distributed and causal influence of frontal operculum in task control. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e108\u003c/em\u003e(10), 4230\u0026ndash;4235. https://doi.org/10.1073/pnas.1013361108\u003c/li\u003e\n\u003cli\u003eHilbert, K., Boeken, O. J., Langhammer, T., Groenewold, N. A., Bas-Hoogendam, J. M., Aghajani, M., Zugman, A., \u0026Aring;hs, F., Arolt, V., Beesdo-Baum, K., Bj\u0026ouml;rkstrand, J., Blackford, J. U., Blanco-Hinojo, L., B\u0026ouml;hnlein, J., B\u0026uuml;low, R., Cano, M., Cardoner, N., Caseras, X., Dannlowski, U., \u0026hellip; Lueken, U. (2024). Cortical and Subcortical Brain Alterations in Specific Phobia and Its Animal and Blood-Injection-Injury Subtypes: A Mega-Analysis From the ENIGMA Anxiety Working Group. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e181\u003c/em\u003e(8), 728\u0026ndash;740. https://doi.org/10.1176/appi.ajp.20230032\u003c/li\u003e\n\u003cli\u003eHill, E. D., Kashyap, P., Raffanello, E., Wang, Y., Moffitt, T. E., Caspi, A., Engelhard, M., \u0026amp; Posner, J. (2025). Prediction of mental health risk in adolescents. \u003cem\u003eNature Medicine\u003c/em\u003e. https://doi.org/10.1038/s41591-025-03560-7\u003c/li\u003e\n\u003cli\u003eHovens, J. G. F. M., Giltay, E. J., Spinhoven, P., Hemert, A. M. van, \u0026amp; Penninx, B. W. J. H. (2015). Impact of Childhood Life Events and Childhood Trauma on the Onset and Recurrence of Depressive and Anxiety Disorders. \u003cem\u003eThe Journal of Clinical Psychiatry\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e(7), 8559. https://doi.org/10.4088/JCP.14m09135\u003c/li\u003e\n\u003cli\u003eHovens, J. G. F. M., Giltay, E. J., Wiersma, J. E., Spinhoven, P., Penninx, B. W. J. H., \u0026amp; Zitman, F. G. (2012). Impact of childhood life events and trauma on the course of depressive and anxiety disorders. \u003cem\u003eActa Psychiatrica Scandinavica\u003c/em\u003e, \u003cem\u003e126\u003c/em\u003e(3), 198\u0026ndash;207. https://doi.org/10.1111/j.1600-0447.2011.01828.x\u003c/li\u003e\n\u003cli\u003eJalnapurkar, I., Allen, M., \u0026amp; Pigott, T. (2018). Sex differences in anxiety disorders: A review. \u003cem\u003eJ Psychiatry Depress Anxiety\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(12), 3\u0026ndash;16.\u003c/li\u003e\n\u003cli\u003eJanak, P. H., \u0026amp; Tye, K. M. (2015). From circuits to behaviour in the amygdala. \u003cem\u003eNature\u003c/em\u003e, \u003cem\u003e517\u003c/em\u003e(7534), 284\u0026ndash;292. https://doi.org/10.1038/nature14188\u003c/li\u003e\n\u003cli\u003eJanssen, R. J., Mour\u0026atilde;o-Miranda, J., \u0026amp; Schnack, H. G. (2018). Making Individual Prognoses in Psychiatry Using Neuroimaging and Machine Learning. \u003cem\u003eBiological Psychiatry. Cognitive Neuroscience and Neuroimaging\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(9), 798\u0026ndash;808. https://doi.org/10.1016/j.bpsc.2018.04.004\u003c/li\u003e\n\u003cli\u003eJin, H., Zhang, B., Cui, H., Li, W., Li, H., Hu, Q., Wang, J., \u0026amp; Li, C. (2020). Altered Function of Superior Parietal Lobule Associated with Perceptive Awareness in First-Episode Drug-Na\u0026iuml;ve Panic Disorders: A Preliminary fMRI Study. \u003cem\u003eNeuropsychiatric Disease and Treatment\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e, 1653\u0026ndash;1659. https://doi.org/10.2147/NDT.S248453\u003c/li\u003e\n\u003cli\u003eKalisch, R., Russo, S. J., \u0026amp; M\u0026uuml;ller, M. B. (2024). Neurobiology and systems biology of stress resilience. \u003cem\u003ePhysiological Reviews\u003c/em\u003e, \u003cem\u003e104\u003c/em\u003e(3), 1205\u0026ndash;1263. https://doi.org/10.1152/physrev.00042.2023\u003c/li\u003e\n\u003cli\u003eKedo, O., Zilles, K., Palomero-Gallagher, N., Schleicher, A., Mohlberg, H., Bludau, S., \u0026amp; Amunts, K. (2018). Receptor-driven, multimodal mapping of the human amygdala. \u003cem\u003eBrain Structure and Function\u003c/em\u003e, \u003cem\u003e223\u003c/em\u003e(4), 1637\u0026ndash;1666. https://doi.org/10.1007/s00429-017-1577-x\u003c/li\u003e\n\u003cli\u003eKessler, R. C. (2003). Epidemiology of women and depression. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e74\u003c/em\u003e(1), 5\u0026ndash;13. https://doi.org/10.1016/S0165-0327(02)00426-3\u003c/li\u003e\n\u003cli\u003eKlinger-K\u0026ouml;nig, J., Erhardt, A., Streit, F., V\u0026ouml;lker, M. P., Schulze, M. B., Keil, T., Fricke, J., Castell, S., Klett-Tammen, C. J., Pischon, T., Karch, A., Teismann, H., Michels, K. B., Greiser, K. H., Becher, H., Karrasch, S., Ahrens, W., Meinke-Franze, C., Schipf, S., \u0026hellip; Grabe, H. J. (2024). Childhood Trauma and Somatic and Mental Illness in Adulthood. \u003cem\u003eDeutsches \u0026Auml;rzteblatt International\u003c/em\u003e, \u003cem\u003e121\u003c/em\u003e(1), 1\u0026ndash;8. https://doi.org/10.3238/arztebl.m2023.0225\u003c/li\u003e\n\u003cli\u003eKlinger-K\u0026ouml;nig, J., Streit, F., Erhardt, A., Kleineidam, L., Schmiedek, F., Schmidt, B., Wagner, M., Deckert, J., Rietschel, M., Berger, K., \u0026amp; Grabe, H. J. (2023). The assessment of childhood maltreatment and its associations with affective symptoms in adulthood: Results of the German National Cohort (NAKO). \u003cem\u003eThe World Journal of Biological Psychiatry\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(10), 897\u0026ndash;908. https://doi.org/10.1080/15622975.2021.2011406\u003c/li\u003e\n\u003cli\u003eKroenke, K., Spitzer, R. L., \u0026amp; Williams, J. B. (2001). The PHQ-9: Validity of a brief depression severity measure. \u003cem\u003eJournal of General Internal Medicine\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(9), 606\u0026ndash;613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x\u003c/li\u003e\n\u003cli\u003eKuhn, M. (2008). Building Predictive Models in R Using the caret Package. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e, 1\u0026ndash;26. https://doi.org/10.18637/jss.v028.i05\u003c/li\u003e\n\u003cli\u003eKuhn, M. (2012). Variable importance using the caret package. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e, 1\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eKuzminskaite, E., Penninx, B. W. J. H., van Harmelen, A.-L., Elzinga, B. M., Hovens, J. G. F. M., \u0026amp; Vinkers, C. H. (2021). Childhood Trauma in Adult Depressive and Anxiety Disorders: An Integrated Review on Psychological and Biological Mechanisms in the NESDA Cohort. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e283\u003c/em\u003e, 179\u0026ndash;191. https://doi.org/10.1016/j.jad.2021.01.054\u003c/li\u003e\n\u003cli\u003eLavallee, K. L., Zhang, X. C., Schneider, S., \u0026amp; Margraf, J. (2021). A longitudinal examination of the relationship between smoking and panic, anxiety, and depression in Chinese and German students. \u003cem\u003eAddictive Behaviors Reports\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e, 100347. https://doi.org/10.1016/j.abrep.2021.100347\u003c/li\u003e\n\u003cli\u003eLe Roux, H., Gatz, M., \u0026amp; Wetherell, J. L. (2005). Age at Onset of Generalized Anxiety Disorder in Older Adults. \u003cem\u003eThe American Journal of Geriatric Psychiatry\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 23\u0026ndash;30. https://doi.org/10.1097/00019442-200501000-00005\u003c/li\u003e\n\u003cli\u003eLeach, L. S., Christensen, H., Mackinnon, A. J., Windsor, T. D., \u0026amp; Butterworth, P. (2008). Gender differences in depression and anxiety across the adult lifespan: The role of psychosocial mediators. \u003cem\u003eSocial Psychiatry and Psychiatric Epidemiology\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(12), 983\u0026ndash;998. https://doi.org/10.1007/s00127-008-0388-z\u003c/li\u003e\n\u003cli\u003eLinden, D. E. J. (2012). The challenges and promise of neuroimaging in psychiatry. \u003cem\u003eNeuron\u003c/em\u003e, \u003cem\u003e73\u003c/em\u003e(1), 8\u0026ndash;22. https://doi.org/10.1016/j.neuron.2011.12.014\u003c/li\u003e\n\u003cli\u003eMadonna, D., Delvecchio, G., Soares, J. C., \u0026amp; Brambilla, P. (2019). Structural and functional neuroimaging studies in generalized anxiety disorder: A systematic review. \u003cem\u003eBrazilian Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e, 336\u0026ndash;362. https://doi.org/10.1590/1516-4446-2018-0108\u003c/li\u003e\n\u003cli\u003eMakovac, E., Meeten, F., Watson, D. R., Garfinkel, S. N., Critchley, H. D., \u0026amp; Ottaviani, C. (2015). Neurostructural abnormalities associated with axes of emotion dysregulation in generalized anxiety. \u003cem\u003eNeuroimage Clin\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e, 172\u0026ndash;181. https://doi.org/10.1016/j.nicl.2015.11.022.\u003c/li\u003e\n\u003cli\u003eMakovac, E., Smallwood, J., Watson, D. R., Meeten, F., Critchley, H. D., \u0026amp; Ottaviani, C. (2017). The verbal nature of worry in generalized anxiety: Insights from the brain. \u003cem\u003eNeuroImage : Clinical\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e, 882\u0026ndash;892. https://doi.org/10.1016/j.nicl.2017.12.014\u003c/li\u003e\n\u003cli\u003eMcGrath, J. J., Al-Hamzawi, A., Alonso, J., Altwaijri, Y., Andrade, L. H., Bromet, E. J., Bruffaerts, R., Almeida, J. M. C. de, Chardoul, S., Chiu, W. T., Degenhardt, L., Demler, O. V., Ferry, F., Gureje, O., Haro, J. M., Karam, E. G., Karam, G., Khaled, S. M., Kovess-Masfety, V., \u0026hellip; Zaslavsky, A. M. (2023). Age of onset and cumulative risk of mental disorders: A cross-national analysis of population surveys from 29 countries. \u003cem\u003eThe Lancet Psychiatry\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(9), 668\u0026ndash;681. https://doi.org/10.1016/S2215-0366(23)00193-1\u003c/li\u003e\n\u003cli\u003eMiloyan, B., Byrne, G. J., \u0026amp; Pachana, N. A. (2014). Age-related changes in generalized anxiety disorder symptoms. \u003cem\u003eInternational Psychogeriatrics\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(4), 565\u0026ndash;572. https://doi.org/10.1017/S1041610213002470\u003c/li\u003e\n\u003cli\u003eMore, S., Eickhoff, S. B., Caspers, J., \u0026amp; Patil, K. R. (2021). Confound Removal and Normalization in Practice: A Neuroimaging Based Sex Prediction Case Study. \u003cem\u003eMachine Learning and Knowledge Discovery in Databases. Applied Data Science and Demo Track\u003c/em\u003e, \u003cem\u003e12461\u003c/em\u003e, 3\u0026ndash;18. https://doi.org/10.1007/978-3-030-67670-4_1\u003c/li\u003e\n\u003cli\u003eNoyes, R. (2001). Comorbidity in Generalized Anxiety Disorder. \u003cem\u003ePsychiatric Clinics of North America\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1), 41\u0026ndash;55. https://doi.org/10.1016/S0193-953X(05)70205-7\u003c/li\u003e\n\u003cli\u003ePenninx, B. W., Pine, D. S., Holmes, E. A., \u0026amp; Reif, A. (2021). Anxiety disorders. \u003cem\u003eLancet (London, England)\u003c/em\u003e, \u003cem\u003e397\u003c/em\u003e(10277), 914\u0026ndash;927. https://doi.org/10.1016/S0140-6736(21)00359-7\u003c/li\u003e\n\u003cli\u003ePessoa, L. (2023). How many brain regions are needed to elucidate the neural bases of fear and anxiety? \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e, \u003cem\u003e146\u003c/em\u003e, 105039. https://doi.org/10.1016/j.neubiorev.2023.105039\u003c/li\u003e\n\u003cli\u003ePeters, A., Consortium, G. N. C., Peters, A., Greiser, K. H., G\u0026ouml;ttlicher, S., Ahrens, W., Albrecht, M., Bamberg, F., B\u0026auml;rnighausen, T., Becher, H., Berger, K., Beule, A., Boeing, H., Bohn, B., Bohnert, K., Braun, B., Brenner, H., B\u0026uuml;low, R., Castell, S., \u0026hellip; Zschocke, J. (2022). Framework and baseline examination of the German National Cohort (NAKO. \u003cem\u003eEuropean journal of epidemiology\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(10), 1107\u0026ndash;1124. https://doi.org/10.1007/s10654-022-00890-5\u003c/li\u003e\n\u003cli\u003eRajbhandari, A. K., Octeau, C. J., Gonzalez, S., Pennington, Z. T., Mohamed, F., Trott, J., Chavez, J., Ngyuen, E., Keces, N., Hong, W. Z., Neve, R. L., Waschek, J., Khakh, B. S., \u0026amp; Fanselow, M. S. (2021). A Basomedial Amygdala to Intercalated Cells Microcircuit Expressing PACAP and Its Receptor PAC1 Regulates Contextual Fear. \u003cem\u003eJournal of Neuroscience\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(15), 3446\u0026ndash;3461. https://doi.org/10.1523/JNEUROSCI.2564-20.2021\u003c/li\u003e\n\u003cli\u003eRuscio, A. M., Hallion, L. S., Lim, C. C. W., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., Andrade, L. H., Borges, G., Bromet, E. J., Bunting, B., Caldas de Almeida, J. M., Demyttenaere, K., Florescu, S., de Girolamo, G., Gureje, O., Haro, J. M., He, Y., Hinkov, H., Hu, C., \u0026hellip; Scott, K. M. (2017). Cross-sectional Comparison of the Epidemiology of DSM-5 Generalized Anxiety Disorder Across the Globe. \u003cem\u003eJAMA Psychiatry\u003c/em\u003e, \u003cem\u003e74\u003c/em\u003e(5), 465\u0026ndash;475. https://doi.org/10.1001/jamapsychiatry.2017.0056\u003c/li\u003e\n\u003cli\u003eSchienle, A., Ebner, F., \u0026amp; Sch\u0026auml;fer, A. (2011). Localized gray matter volume abnormalities in generalized anxiety disorder. \u003cem\u003eEuropean Archives of Psychiatry and Clinical Neuroscience\u003c/em\u003e, \u003cem\u003e261\u003c/em\u003e(4), 303\u0026ndash;307. https://doi.org/10.1007/s00406-010-0147-5\u003c/li\u003e\n\u003cli\u003eSchuppert, C., Rospleszcz, S., Hirsch, J. G., Hoinkiss, D. C., K\u0026ouml;hn, A., Kr\u0026uuml;chten, R., Russe, M. F., Keil, T., Krist, L., Schmidt, B., Michels, K. B., Schipf, S., Brenner, H., Kr\u0026ouml;ncke, T. J., Pischon, T., Niendorf, T., Schulz-Menger, J., Forsting, M., V\u0026ouml;lzke, H., \u0026hellip; Schlett, C. L. (2023). Automated image quality assessment for selecting among multiple magnetic resonance image acquisitions in the German National Cohort study. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 22745. https://doi.org/10.1038/s41598-023-49569-1\u003c/li\u003e\n\u003cli\u003eSnoek, L., Miletić, S., \u0026amp; Scholte, H. S. (2019). How to control for confounds in decoding analyses of neuroimaging data. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e184\u003c/em\u003e, 741\u0026ndash;760. https://doi.org/10.1016/j.neuroimage.2018.09.074\u003c/li\u003e\n\u003cli\u003eSolmi, M., Radua, J., Olivola, M., Croce, E., Soardo, L., Salazar de Pablo, G., Il Shin, J., Kirkbride, J. B., Jones, P., Kim, J. H., Kim, J. Y., Carvalho, A. F., Seeman, M. V., Correll, C. U., \u0026amp; Fusar-Poli, P. (2022). Age at onset of mental disorders worldwide: Large-scale meta-analysis of 192 epidemiological studies. \u003cem\u003eMolecular Psychiatry\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(1), 281\u0026ndash;295. https://doi.org/10.1038/s41380-021-01161-7\u003c/li\u003e\n\u003cli\u003eSpitzer, R. L., Kroenke, K., \u0026amp; Williams, J. B. (1999). Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. \u003cem\u003eJAMA\u003c/em\u003e, \u003cem\u003e282\u003c/em\u003e(18), 1737\u0026ndash;1744. https://doi.org/10.1001/jama.282.18.1737\u003c/li\u003e\n\u003cli\u003eStreit, F., Zillich, L., Frank, J., Kleineidam, L., Wagner, M., Baune, B. T., Klinger-K\u0026ouml;nig, J., Grabe, H. J., Pabst, A., Riedel-Heller, S. G., Schmiedek, F., Schmidt, B., Erhardt, A., Deckert, J., NAKO Investigators, Rietschel, M., \u0026amp; Berger, K. (2023). Lifetime and current depression in the German National Cohort (NAKO). \u003cem\u003eThe World Journal of Biological Psychiatry: The Official Journal of the World Federation of Societies of Biological Psychiatry\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(10), 865\u0026ndash;880. https://doi.org/10.1080/15622975.2021.2014152\u003c/li\u003e\n\u003cli\u003eStrom, N. I., Verhulst, B., Bacanu, S.-A., Cheesman, R., Purves, K. L., Gedik, H., Mitchell, B. L., Kwong, A. S., Faucon, A. B., Singh, K., Medland, S., Colodro-Conde, L., Krebs, K., Hoffmann, P., Herms, S., Gehlen, J., Ripke, S., Awasthi, S., Palviainen, T., \u0026hellip; Hettema, J. M. (2024). Genome-wide association study of major anxiety disorders in 122,341 European-ancestry cases identifies 58 loci and highlights GABAergic signaling. \u003cem\u003emedRxiv: The Preprint Server for Health Sciences\u003c/em\u003e, 2024.07.03.24309466. https://doi.org/10.1101/2024.07.03.24309466\u003c/li\u003e\n\u003cli\u003eSuor, J. H., Jimmy, J., Monk, C. S., Phan, K. L., \u0026amp; Burkhouse, K. L. (2020). Parsing differences in amygdala volume among individuals with and without social and generalized anxiety disorders across the lifespan. \u003cem\u003eJournal of Psychiatric Research\u003c/em\u003e, \u003cem\u003e128\u003c/em\u003e, 83\u0026ndash;89. https://doi.org/10.1016/j.jpsychires.2020.05.027\u003c/li\u003e\n\u003cli\u003eTeicher, M. H., Samson, J. A., Anderson, C. M., \u0026amp; Ohashi, K. (2016). The effects of childhood maltreatment on brain structure, function and connectivity. \u003cem\u003eNature Reviews Neuroscience\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(10), 652\u0026ndash;666. https://doi.org/10.1038/nrn.2016.111\u003c/li\u003e\n\u003cli\u003eTer Meulen, W. G., Draisma, S., van Hemert, A. M., Schoevers, R. A., Kupka, R. W., Beekman, A. T. F., \u0026amp; Penninx, B. W. J. H. (2021). Depressive and anxiety disorders in concert-A synthesis of findings on comorbidity in the NESDA study. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e284\u003c/em\u003e, 85\u0026ndash;97. https://doi.org/10.1016/j.jad.2021.02.004\u003c/li\u003e\n\u003cli\u003eter Meulen, W. G., Draisma, S., van Hemert, A. M., Schoevers, R. A., Kupka, R. W., Beekman, A. T. F., \u0026amp; Penninx, B. W. J. H. (2021). Depressive and anxiety disorders in concert\u0026ndash;A synthesis of findings on comorbidity in the NESDA study. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e284\u003c/em\u003e, 85\u0026ndash;97. https://doi.org/10.1016/j.jad.2021.02.004\u003c/li\u003e\n\u003cli\u003eVia, E., Fullana, M. A., Goldberg, X., Tinoco-Gonz\u0026aacute;lez, D., Mart\u0026iacute;nez-Zalaca\u0026iacute;n, I., Soriano-Mas, C., Davey, C. G., Mench\u0026oacute;n, J. M., Straube, B., Kircher, T., Pujol, J., Cardoner, N., \u0026amp; Harrison, B. J. (2018). Ventromedial prefrontal cortex activity and pathological worry in generalised anxiety disorder. \u003cem\u003eThe British Journal of Psychiatry: The Journal of Mental Science\u003c/em\u003e, \u003cem\u003e213\u003c/em\u003e(1), 437\u0026ndash;443. https://doi.org/10.1192/bjp.2018.65\u003c/li\u003e\n\u003cli\u003eWang, X., Cheng, B., Wang, S., Lu, F., Luo, Y., Long, X., \u0026amp; Kong, D. (2021). Distinct grey matter volume alterations in adult patients with panic disorder and social anxiety disorder: A systematic review and voxel-based morphometry meta-analysis. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e281\u003c/em\u003e, 805\u0026ndash;823. https://doi.org/10.1016/j.jad.2020.11.057\u003c/li\u003e\n\u003cli\u003eWei\u0026szlig;, M., Gutzeit, J., Appel, K. S., Bahmer, T., Beutel, M., Deckert, J., Fricke, J., Han\u0026szlig;, S., Hettich-Damm, N., Heuschmann, P. U., Horn, A., Jauch-Chara, K., Kohls, M., Krist, L., Lorenz-Depiereux, B., Otte, C., Pape, D., Reese, J.-P., Schreiber, S., \u0026hellip; Hein, G. (2024). Depression and fatigue six months post-COVID-19 disease are associated with overlapping symptom constellations: A prospective, multi-center, population-based cohort study. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e352\u003c/em\u003e, 296\u0026ndash;305. https://doi.org/10.1016/j.jad.2024.02.041\u003c/li\u003e\n\u003cli\u003eWittchen, H. U., Jacobi, F., Rehm, J., Gustavsson, A., Svensson, M., J\u0026ouml;nsson, B., Olesen, J., Allgulander, C., Alonso, J., Faravelli, C., Fratiglioni, L., Jennum, P., Lieb, R., Maercker, A., van Os, J., Preisig, M., Salvador-Carulla, L., Simon, R., \u0026amp; Steinhausen, H.-C. (2011). The size and burden of mental disorders and other disorders of the brain in Europe 2010. \u003cem\u003eEuropean Neuropsychopharmacology: The Journal of the European College of Neuropsychopharmacology\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(9), 655\u0026ndash;679. https://doi.org/10.1016/j.euroneuro.2011.07.018\u003c/li\u003e\n\u003cli\u003eZvolensky, M. J., Feldner, M. T., Leen-Feldner, E. W., \u0026amp; McLeish, A. C. (2005). Smoking and panic attacks, panic disorder, and agoraphobia: A review of the empirical literature. \u003cem\u003eClinical Psychology Review\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(6), 761\u0026ndash;789. https://doi.org/10.1016/j.cpr.2005.05.001\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e In this context, \u0026ldquo;predict\u0026rdquo; does not imply a temporal or causal forecast. Instead, it follows the common usage in machine learning and statistics, referring to the explanatory value of a model. The term \u0026ldquo;predict\u0026rdquo; is used throughout this paper in that sense, unless explicitly indicated otherwise.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6627834/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6627834/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAnxiety disorders (ANX) are common and impairing mental health conditions. This study aimed to classify self-reported symptoms of generalized anxiety disorder (GAD) and panic attacks as two psychopathological manifestations of ANX by applying machine learning to a cross-sectional dataset of 26,378 adults from the German National Cohort Study (NAKO). We first explored linear relationships between preselected neuroimaging correlates in MRI scans and anxiety phenotypes. Overall, sex-stratified correlation coefficients - while partly highly significant - were extremely low with \u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;.04 for panic attacks and \u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;.06 for GAD symptoms after correction for confounding variables like childhood trauma and depression. We then examined the combined classifying value of whole-brain imaging data of 246 ROIs in addition to psychosocial variables such as self-reported depression symptoms, stress, and childhood trauma, using four machine learning algorithms (support-vector machines with linear and radial kernels, elastic-net regression, and random forest). Neuroimaging data, particularly gray-matter volumes in regions such as the amygdala and superior parietal lobule, contributed to classification, but performance was substantially better when psychosocial variables were added. For both GAD symptoms and panic attacks, depression, stress and childhood trauma were the clearest indicators the classification would show the condition was present. Random forest models based on psychosocial variables alone achieved the highest discrimination performance for GAD symptoms (area under the receiver operating characteristic curve, AUROC\u0026thinsp;=\u0026thinsp;0.973) and panic attacks (AUROC\u0026thinsp;=\u0026thinsp;0.933). Combining neuroimaging and psychosocial variables in elastic-net regressions further improved specificity. These results support multimodal approaches to diagnose and investigate ANX that integrate structural brain abnormalities and psychosocial measures to capture the complexity of GAD and panic attacks, enabling the creation of individual risk profiles based on multiple biomarkers. These profiles may guide tailored therapeutic and preventive interventions.\u003c/p\u003e","manuscriptTitle":"Multimodal phenotypic classification of generalized anxiety and panic using structural MRI data and psychosocial factors: Machine learning results from the German National Cohort (NAKO) study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-15 06:32:57","doi":"10.21203/rs.3.rs-6627834/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"transferred","content":"Translational Psychiatry","date":"2025-10-02T13:40:45+00:00","index":"","fulltext":""},{"type":"decision","content":"Reject after peer review","date":"2025-09-26T15:11:43+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-09-21T04:14:52+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-09-13T23:51:20+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-09-09T04:09:43+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-09-08T23:53:33+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2025-09-08T06:48:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-12T09:27:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-12T08:54:47+00:00","index":"","fulltext":""},{"type":"submitted","content":"Molecular Psychiatry","date":"2025-05-09T10:50:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"14c9dce2-0326-47fc-bede-ff4c7968a64f","owner":[],"postedDate":"September 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":54337607,"name":"Health sciences/Diseases/Psychiatric disorders"},{"id":54337608,"name":"Biological sciences/Psychology"},{"id":54337609,"name":"Health sciences/Biomarkers/Diagnostic markers"}],"tags":[],"updatedAt":"2025-11-28T14:40:25+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-15 06:32:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6627834","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6627834","identity":"rs-6627834","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}