Tractometry analysis represents a significant advancement in neuroimaging, offering a detailed examination of the brain's white matter at a micro level. Unlike traditional ROI or voxel-based methods, tractometry precisely reconstructs and characterizes white matter tracts. Using advanced diffusion MRI and tractography algorithms, it maps the trajectory, shape, and connectivity patterns of individual white matter bundles. Accurate alignment of these tracts across different groups is crucial for reliable and reproducible results. Nonlinear registration techniques are essential for achieving this alignment, harmonizing bundle shapes, and improving sensitivity to disease-related changes. However, nonlinear registration is complex, especially with tractography data, which digitally represents the brain's white matter anatomy. Potential structural changes in the bundle's shape during registration can lead to artifacts that obscure critical anatomical details needed for disease identification. We introduce BundleWarp, a streamline-based nonlinear deformable registration method designed specifically for white matter tracts. BundleWarp employs a sophisticated approach to align two white matter bundles while preserving their topological and anatomical features. It is formulated as a probability density estimation problem with motion coherence penalties, ensuring coherent movement of points along streamlines and maintaining the anatomical integrity of tracts through displacement field regularization. Additionally, we introduce a tract morphometry framework utilizing the displacement field generated by BundleWarp to analyze white matter tract shape differences. Our results show that BundleWarp effectively quantifies bundle shape differences and enhances structural harmonization in tractometry analysis for diverse subjects, including those with Alzheimer's and Parkinson's disease. Test-retest experiments further demonstrate that BundleWarp substantially improves subject fingerprinting by increasing within-subject reproducibility of both bundle shape and microstructural profiles (FA, MD, RD, AD). It precisely maps the brain's neuronal pathways, offering a robust tractometry framework with enhanced sensitivity for detecting disease-related structural and microstructural changes in white matter tracts associated with Mild Cognitive Impairment (MCI), dementia, and early-stage Alzheimer's biomarkers, including amyloid-beta plaques and tau neurofibrillary tangles.

Tractometry enables detailed mapping of white matter microstructure along individual tracts and is widely used to study disease effects such as those seen in Alzheimer's disease (AD). However, how different tractography algorithms influence tractometry outcomes remains unclear. Here, we compared whole-brain deterministic and probabilistic tractography using the BUndle ANalytics (BUAN) framework in the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, including 118 AD and 728 cognitively normal (CN) participants. Both approaches revealed the expected pattern of lower fractional anisotropy (FA) and higher mean, radial, and axial diffusivity (MD, RD, AxD) in AD, consistent with white matter degeneration. Despite broadly similar global trends, substantial bundle-level differences emerged between the two tractography methods. Probabilistic tracking produced stronger and more spatially extended effects in the fornix, a small and highly curved limbic pathway vulnerable to AD-related degeneration, whereas deterministic tracking showed greater sensitivity in the posterior segments of the right superior longitudinal fasciculus (SLF_R). These discrepancies highlight that the choice of tractography algorithm can alter detecting disease effects, emphasizing the need for cross-method validation to ensure the robustness and interpretability of along-tract measures.

Epilepsy is characterized by widespread structural brain alterations extending beyond the epileptic zone, involving both cortical and subcortical regions. Importantly, the clinical manifestation of epilepsy, including seizure types, psychiatric comorbidities, and treatment responses, has been shown to differ between sexes. However, sex differences in structural alterations in epilepsy have been seldomly reported in neuroimaging studies, partly due to limited sample sizes and single-center designs. Here, we systematically investigated sex differences in common epilepsies and their related clinical variables using structural neuroimaging biomarkers in an international multi-center cohort of 1,253 epilepsy patients and 1,077 healthy controls. We studied cortical thickness and subcortical volume in two types of epilepsy: temporal lobe epilepsy (TLE) and genetic generalized epilepsy (GGE). Both male and female patients with TLE showed widespread cortical and subcortical thinning compared with controls. In GGE, when compared separately to controls, male patients showed only subtle structural alterations, whereas female patients exhibited more widespread structural alterations. Sex-stratified analyses revealed some variation in the extent and distribution of cortical thickness and subcortical volume alterations between male and female patients in both epilepsy cohorts. Yet, we did not find significant sex-by-diagnosis interaction effects in TLE and GGE. Similarly, no significant interaction effects were observed between sex and age of onset or disease duration in either patient group. Overall, although we observed some differences in regional cortical thickness and subcortical volume between male and female patients with epilepsy, we did not find significant sex-by-diagnosis interactions. Our findings indicate that sex differences in behavioral and clinical outcomes of epilepsy may involve biological or functional processes that require further investigation.

Diagnostic MRI evaluation of temporal lobe epilepsy (TLE) depends on the subjective visual interpretation of MRI images. These interpretations could be enhanced by quantitative artificial intelligence (AI) support tools. Humans often make sequential and conditional decisions during their radiological interpretations, such as whether an abnormality is present and, if present, characterizing the abnormality. It is not known whether it is superior to train AI to treat every decision separately in a similar step-wise manner or to train a model holistically on all decisions simultaneously. Here, we analysed three large epilepsy MRI datasets [n = 3676, 2320 people with epilepsy and 1356 healthy controls (HC)] to perform two tasks: (i) establish the presence of a TLE pattern on MRI and (ii) determine TLE pattern lateralization. We compared Step-wise models that independently classify TLE versus HC and lateralize patients as left TLE (L-TLE) or right TLE (R-TLE), against a simultaneous model trained to distinguish all three classes in a single step. To do this, 3D volumetric T1-weighted images were input into an EfficientNetV2 model multiple times to ensure reproducibility of results. Class prediction, model classification confidence and saliency maps were output for interpretability. Step-wise models outperformed the Simultaneous model on both tasks (both Ps < 0.001), with an average ∼2.8% accuracy increase for discriminating HC from TLE and an average 12.7% accuracy increase for distinguishing L-TLE from R-TLE. For both the Step-wise and Simultaneous models, important features discriminating TLE from HC included the known TLE limbic pattern involving the hippocampus, parahippocampal cortical regions, cingulate cortex and lateral temporal regions. However, there was less concordance between the Step-wise and Simultaneous models for the L-TLE versus R-TLE task (all Fisher's Zs > 10.5, Ps < 0.001); the Step-wise model focused less on subcortical regions such as the thalamus and hippocampus and focused more on distributed cortical pathology. Across the two Step-wise models, 95.1% of TLE patients had accurate classifications in either HC versus TLE and/or L-TLE versus R-TLE tasks. These results included 69.6% of patients being both correctly labelled as TLE and lateralized, 13.9% being correctly labelled TLE but lateralized incorrectly and 11.6% being lateralized correctly but not detected as TLE. These findings provide evidence that diagnostic tasks with simpler, Step-wise AI models may enhance diagnostic performance and interpretability in clinical workflows. Future AI clinical support tools can leverage this step-wise approach in the early identification of TLE-related structural patterns, supporting timely diagnosis and treatment decisions.

BACKGROUND:Childhood maltreatment (CM), encompassing abuse and neglect, is highly prevalent and associated with elevated risk for major depressive disorder (MDD), posttraumatic stress disorder (PTSD), and other related conditions. However, the extent to which neuroanatomical alterations in MDD and PTSD are attributable to CM is uncertain. METHODS:Here, we analyzed CM and whole-brain magnetic resonance imaging (MRI) data from 3711 participants in the ENIGMA (Enhancing Neuro Imaging Genetics through Meta Analysis) MDD and PTSD Working Groups (25 sites; mean age = 33.3 ± 13.0 years; 59.9% female). Normative modeling estimated deviation z scores for 14 subcortical volume, 68 cortical thickness (CT), and 68 surface area (SA) measures. To identify transdiagnostic effects, associations between CM and brain deviation scores were evaluated across all participants (patients and healthy control participants) stratified by sex and 3 age bins (pediatric, young adult, older adult). RESULTS:In young adults (ages 18-35), abuse was associated with larger volumes in the thalamus and pallidum, thinner isthmus cingulate and middle frontal regions, and thicker medial orbitofrontal cortex; there were no significant effects in pediatric (≤18 years) participants. The strongest effects were observed in young female adults (|β| = 0.07-0.22, q < .05): Greater abuse and neglect were correlated with smaller hippocampus and putamen volumes, thinner entorhinal cortex, and smaller SA in fusiform/inferior parietal regions and with larger SA in the orbitofrontal and occipital cortices. In males, abuse had widespread effects on CT and SA (|β| = 0.1-0.18, q < .05); effects for neglect were minimal. CONCLUSIONS:Our findings of age- and sex-specific instantiations of CM on brain morphometry highlight the importance of developmental context in understanding how adverse experiences shape neurobiological vulnerability to MDD and PTSD.

Abstract Brain disorders are increasingly understood as disorders of distributed brain circuits, yet functional connectivity (FC), the dominant framework for mapping them, treats the brain as a collection of pairwise relationships between regions and cannot represent pathology distributed across coordinated sets of connections. We introduce a Hodge-Laplacian topological framework that localizes higher-order “loop” (1-cycle) organization within functional connectome, maps each loop to specific edges and networks, and yields a subject-level measure of loop expression. Applied to resting-state fMRI from the ENIGMA-OCD consortium (1,024 patients and 1,028 controls across 28 sites), the framework identified 93 loop-level abnormalities in obsessive–compulsive disorder (OCD), concentrated in frontoparietal and somatomotor systems. The edges forming these loops largely showed no significant differences between groups, indicating that the abnormalities were invisible to conventional FC analysis. The frontoparietal and somatomotor loop clusters recurred across the clinical subgroups, suggesting convergence on a shared higher-order phenotype. Robustness analyses showed the loop signal reflected higher-order organization rather than an artifact of individual edges, the network backbone, or any single site. These results indicate that coordinated, multi-edge pathology exists and can be localized even when pairwise analyses fail to detect it, positioning higher-order topology as a generalizable axis for mapping circuit pathology across psychiatric and neurological disorders.

Importance Childhood trauma is associated with increased risk for bipolar disorder, but the biological mechanisms of this association remain incompletely defined. Gray matter differences observed after trauma exposure overlap with those reported in bipolar disorder, suggesting that the association of childhood trauma with bipolar disorder might be mediated through brain morphology. Objective To determine whether cortical thickness, cortical surface, or subcortical volume mediate the association of childhood trauma with bipolar disorder. Design, Setting, and Participants This case-control study conducted a cross-sectional analysis of individuals with bipolar disorder and healthy controls from 19 international cohorts (Enhancing NeuroImaging Genetics Through Meta-Analyses [ENIGMA] Bipolar Disorder Working Group) from January 2010 to December 2022. Data were analyzed from January 2025 to January 2026. Exposures The primary exposure was the severity of total childhood trauma assessed with the Childhood Trauma Questionnaire, with secondary analyses of 5 subscales (emotional neglect and abuse, physical neglect and abuse, and sexual abuse). Main Outcomes and Measures The primary outcome was bipolar disorder diagnosis (case vs control). The primary measure was the mediation effects of childhood trauma on diagnosis via gray matter (75 bilateral-averaged cortical thickness, surface, and subcortical volume measures). The mediation pathway from severity of childhood trauma to bipolar disorder through brain morphology was specified a priori. High-dimensional mediation analysis, with leave-one-site-out cross-validation and permutation testing for significance (false discovery rate [FDR]), was conducted. Results The final sample included 2221 healthy controls (mean [SD] age, 35.6 [13.2] years; 1274 female [57%]) and 1031 participants with bipolar disorder (mean [SD] age, 38.6 [13.7] years; 579 female [56%]). Severity of childhood trauma was directly associated with higher likelihood of having a bipolar disorder diagnosis (median coefficient, 0.841; 95% CI, 0.834-0.851; range, 0.776-0.893; FDR P < .001). Less than 1% of the association between childhood trauma and bipolar disorder was mediated by brain morphology. Statistically significant mediators were hippocampal volume (median coefficient, 0.004; 95% CI, 0.002-0.005; range, 0-0.008; FDR P < .001), medial orbitofrontal gray matter thickness (median coefficient, 0.002; 95% CI, 0.002-0.003; range, 0-0.004; FDR P < .001), and superior frontal gyrus gray matter thickness (median coefficient, 0.002; 95% CI, 0.002-0.003; range, 0-0.005; FDR P < .001). Conclusions and Relevance This study found that severity of childhood trauma exposure was associated with bipolar disorder diagnosis in part through a smaller hippocampus, thinner cortex in the medial orbitofrontal gyrus, and thinner cortex in the superior frontal gyrus. The identification of this mechanistic pathway improves understanding of the disorder, could help to identify those at risk, and enable the development of new interventions.

Resting-state functional connectivity (rsFC) studies have revealed altered regional homogeneity (ReHo) and degree centrality (DC) in individuals with anorexia nervosa (AN) compared to healthy controls (HC), but the underlying mechanisms remain unclear. Here we explored the spatial alignment with neurotransmitter receptor and transporter densities (i.e., “chemoarchitecture”, based on “reference” PET studies) as a potential explanatory factor. We investigated rsFC alterations in acutely underweight patients with AN (n = 87) and age-matched HC (n = 87) cross-sectionally at admission and then again after successful weight-restoration treatment. Global ReHo and DC maps were associated with the spatial distribution of neurotransmitter receptors, transporters and/or metabolic glucose uptake. First, the correlation between rsFC alterations in AN and chemoarchitecture was evaluated at the group/timepoint-level. Second, individual-level correlations of neuroreceptor maps with rsFC alterations were calculated to test for possible associations with early weight restoration. The acute state of AN was characterized by higher DC (but not ReHo) in brain regions with a higher cortical density of vesicular acetylcholine transporter (VAChT), dopamine transporter (DAT) and serotonin transporter (SERT). Conversely, weight restoration was associated with normalization of DC, especially in areas with a higher DAT density. Importantly, individual-level spatial correlations between VAChT, DAT and SERT densities and DC alterations at admission significantly predicted early weight gain over first 90 days of treatment. These results suggest that neurochemical context may underlie functional brain alterations, providing a preliminary step toward identifying biological risk signatures. Replication with individualized PET data will be crucial to validate their potential utility for treatment stratification and personalization.

BACKGROUND:In a recent coordinated meta-analysis of neuroimaging data, we reported gray matter (GM) alterations in acutely underweight patients with anorexia nervosa (AN). Here, we extend these findings by examining individual variation in brain structure within AN, individual-level differentiation between AN and healthy controls (HC), and differences between AN subtypes, with potential relevance for understanding clinical heterogeneity. METHODS AND FINDINGS:We analyzed individual-level data from 11 international sites in the ENIGMA Eating Disorders Working Group, including 570 female participants with AN and 739 HC. We examined cortical thickness, cortical surface area and subcortical volumes in AN versus HC using three complementary approaches: (i) group-level differences in a mega-analysis correcting for age effects, (ii) frequencies of extreme deviations (infra-/supranormal; z < -1.96/z > 1.96) based on normative reference models by the CentileBrain Initiative, and (iii) individual-level classification performance using machine learning. The same analytic framework was applied to compare AN restricting versus binge-eating/purging subtype, additionally correcting for BMI effects. Mega-analyses reinforced previous meta-analytic findings of pronounced and widespread GM deficits in AN compared to HC. Normative modelling revealed that the frequency of infranormal z-scores (23/68 cortical thickness, 13/14 subcortical volume metrics) and supranormal z-scores (35/68 cortical thickness, 17/68 cortical surface area metrics) was significantly higher in AN than expected based on reference data. Individuals with AN could be reliably differentiated from HC using machine-learning classifiers (ROC-AUC = 0.75-0.81). In contrast, neither group-level differences nor frequency of extreme z-scores differed between AN subtypes, and individuals with different subtypes could not be reliably differentiated from each other. Importantly, the observational design cannot distinguish neurobiological differences related to AN from the effects of starvation or low BMI in the AN versus HC analyses. The lack of differences between subtypes does not exclude brain structural differences between AN subtypes that might be detectable with other modalities or analytic approaches. CONCLUSION:Using a mega-analytic approach, we confirm widespread GM deficits in AN, show that these alterations are (in some patients) extreme, and demonstrate that they enable robust classification with superior performance compared to most MRI-based psychiatric classification studies. The absence of differences between AN subtypes may reflect shared neurobiology, though other imaging modalities may reveal distinctions beyond brain structure.

Scaling laws describe how model performance improves as the amount of training data increases, and recent theories such as the zeta law suggest that scaling behavior is influenced by the eigenspectrum of the model's latent representation. Here, we evaluated whether the distribution of discriminative signals across spectral modes predicts the future scaling behavior, for MRI transformers trained for disease classification. We trained three supervised 3D vision transformers (ViT3D, MINiT, and NIT) for Alzheimer's disease classification using 2,822 training scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI); we compared their encoder spectra with that of a frozen self-supervised DINO ViT-B/16 encoder adapted to 3D MRI. The supervised models learned highly concentrated representations, with 90-96% of CLS-token variance captured by a single principal component, whereas DINO distributed signal across many latent directions. Via spectral expansion of the Mahalanobis signal, we found that supervised training concentrated disease information into a single dominant mode, while self-supervised training produced a richer spectral geometry with higher effective rank and discoverability. This led to different scaling behavior: supervised models exhibited flatter A U C ( N ) curves, yet DINO continued to improve as sample size increased, gaining 11.0 percentage points from N = 50 to N = 2,822 . Overall, the spectral distribution of the discriminative signal, for these different encoder types, influenced how much performance remained discoverable as sample size increased. Distributed representations may retain signal across many latent modes and continue to improve with additional data, whereas concentrated representations tend to exhaust most of the discoverable signal at much lower sample sizes.

BACKGROUND:Previous large-scale structural MRI analyses of the brain in autism have identified gray matter (GM) differences when using region-of-interest analyses based on gross anatomical regions. However, such analyses have limited spatial specificity and may obscure subtle focal differences. Whole brain voxel-based morphometry (VBM) analyses enable greater spatial precision to identify and localize previously undetected neuroanatomical alterations. PURPOSE:To rigorously identify voxel-wise GM and white matter (WM) volume differences in autism in the largest VBM mega-analysis to date. MATERIALS AND METHODS:This retrospective mega-analysis included structural 3D volumetric T1-weighted MRI brain scans from 3,051 participants (15.0 ± 8.2 yrs; 76.8% male; 1,519 autism; 1,532 neurotypicals) collected across 51 sites/scanners. Voxel-wise GM and WM volumes were quantified using the ENIGMA CAT12 VBM pipeline. Linear mixed-effects regression was performed at each voxel to evaluate the association between diagnostic group and voxel-wise volume while adjusting for standard nuisance covariates. RESULTS:Autism was associated with widespread lower GM volume involving cortical, subcortical, and cerebellar regions (peak t=7.39, peak β=0.13); such GM differences were most notably detected in the bilateral orbitofrontal cortex, amygdala, thalamus, and posterior lobes of the cerebellum. WM volume was lower in autism across major projection, commissural, association, and cerebellar/brainstem tracts (peak t=6.74, peak β=0.08), including the corona radiata, internal capsule, corpus callosum, and cerebellar peduncles. These findings remained consistent in sensitivity analyses, including covarying for full-scale IQ and the application of increasingly strict motion exclusion criteria. CONCLUSION:Autism is associated with smaller voxel-wise GM and WM volume involving widespread cortical, subcortical, and cerebellar regions. This high-resolution identification and localization of structural brain differences support the involvement of distributed neural systems in autism that underlie reward processing, sensory integration, and motor functioning in autism.

Major depressive disorder (MDD) is common and disabling, yet reported brain structural differences vary across studies. Here we performed a large vertex-wise (point-by-point) meta-analysis of cortical thickness and surface area using harmonized magnetic resonance imaging processing across 64 cohorts from the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) MDD and Depression Imaging Research Consortium (DIRECT) consortia (5,736 patients; 6,538 controls). We show significantly lower cortical thickness in patients with MDD in multiple brain regions, including the inferior parietal, lateral occipital, superior parietal, medial and lateral orbitofrontal, anterior and posterior cingulate, and precentral gyri, with cortical surface area showing no significant differences. Effects were most pronounced in adults with acute depression, whereas adolescents showed no significant case-control differences. Antidepressant medication use at scanning was associated with more extensive thinning, although effect sizes remained modest (mostly |Cohen's d| < 0.20). This high-resolution, globally generalizable map can support studies of mechanisms and help evaluate structural markers of the clinical course and treatment response.

Genome-wide association studies (GWAS) have advanced the quest to understand how specific genetic variants influence human brain structure and function. Recent work has identified hundreds of common variants associated with subcortical brain volumes, sparking interest in how these genetic markers overlap across brain networks. While this can be estimated by hierarchical clustering of the genetic correlation matrix to identify modular patterns of shared architecture, no brain-wide maps of these effects are available. To address this, we computed polygenic scores (PGS) from loci associated with ten brain volume regions of interest (ROIs): nine major subcortical structures and intracranial volume, with each locus weighted by its association with regional volume. In an independent sample from the discovery GWAS, we performed large-scale segmentation of 3D volumetric T1-weighted MRI scans using voxel-based morphometry (VBM) to map 3D profile of regions where gray matter volume (GMV) was associated with each PGS. We found statistically significant, localized effects for PGS defined for the amygdala, thalamus, and basal ganglia, but PGS for brainstem volume was associated with widespread differences throughout the brain. These brain-wide maps reveal patterns consistent with both localized and distributed genetic influences, offering a novel approach to interpret the genomic architecture of brain structure.

Posttraumatic stress disorder (PTSD) is a psychiatric condition that may develop after trauma exposure. PTSD is characterized by considerable clinical heterogeneity. The amygdala's key role in fear conditioning makes it an important focus for investigating the neurobiology of PTSD. However, associations between amygdala volume and PTSD have been inconsistent. The amygdala consists of functionally distinct nuclei. Specific associations between amygdala nuclei volumes and PTSD may account for previous discrepancies between PTSD and whole amygdala volume. This study investigates the associations between amygdala nuclei volumes, PTSD diagnosis, severity, symptom cluster scores, age of onset and childhood trauma. Individuals with a PTSD diagnosis (n = 771) and controls (n = 1 081, 72% trauma-exposed) were sourced from the Enhancing Neuro-Imaging Genetics through Meta-Analysis and Psychiatric Genomics Consortium (mean age = 32.4 years, (SD = 13 years), 60% male). Nine amygdala nuclei volumes were compared to PTSD diagnosis, age of onset, overall severity, symptom cluster scores (re-experiencing, arousal, and avoidance/emotional numbing), and childhood trauma subscales. Analyses were performed using ordinary least-squares regression, corrected for age, sex, intracranial volume, and whole amygdala volume. PTSD diagnosis was not significantly associated with amygdala nuclei volumes. PTSD severity scores were associated with smaller right lateral nucleus volume (β = -0.26, pBON = 0.01). Smaller right lateral nucleus volume was also associated with re-experiencing (β = -1.01, pBON = 0.04) and arousal (β = -0.9, pBON = 0.04), smaller left paralaminar nucleus volume was associated with re-experiencing (β = -0.1, pBON = 0.04), smaller left corticoamygdaloid transition area volume was associated with avoidance (β = -0.31, pBON = 0.02). Larger left and right central nucleus volumes were significantly associated with childhood physical abuse (β = 0.24, pBON = 9 × 10-3) and neglect (β = 0.29, pBON = 0.04), respectively. Differences in select amygdala nuclei volumes among adults are associated with PTSD severity, symptom cluster scores, and childhood physical abuse and neglect. These findings demonstrate nuclei-specific patterns consistent with their functional roles in fear learning and expression.

INTRODUCTION:Ambient air pollution increases Alzheimer's disease (AD) risk, yet exposure associations with cortical thickness (CTh) in AD-vulnerable brain regions is unclear. Here we examined the associations between PM2.5 and NO2 with CTh in an AD meta-region of interest (ROI), and across the cortex, in participants from the Vietnam Era Twin Study of Aging (VETSA) and Women's Health Initiative Memory Study (WHIMS). METHODS:We conducted a cross-sectional study using data from 387 VETSA men (Mage=61.9 ± 2.6) and 1097 WHIMS women (Mage=77.9 ± 3.7) without dementia or stroke prior to MRI. Long-term residential exposures to PM2.5 and NO2 were quantified as the 3-year average of monthly estimates prior to MRI that were derived from spatiotemporal models with regionalized universal kriging. Brain MRI scans were processed using FreeSurfer-v.5.3.0 to estimate CTh in 34 bilateral regions parcellated with the Desikan-Killiany atlas. An AD meta-ROI was calculated as the surface-area weighted average of CTh in four bilateral regions (entorhinal, fusiform, inferior temporal, and middle temporal cortices) that are vulnerable to AD. Linear mixed models were conducted separately in each cohort with appropriate covariates. RESULTS:In WHIMS, exposures were negatively associated with CTh in the AD meta-ROI (pPM2.5<0.001; pNO2=0.018) and diffusely across the cortex. In VETSA, exposures were positively associated with CTh in the AD meta-ROI (pPM2.5=0.017; pNO2=0.021) and temporal pole, but effects were age-dependent, becoming negative (though nonsignificant) after age 64. DISCUSSION:Positive associations in younger VETSA men, coupled with negative associations in older WHIMS women, may suggest nonmonotonic AD-related neurodegeneration.

Accurate in vivo prediction of neuropathology is critical for advancing diagnosis and treatment of Alzheimer’s disease and related dementias (ADRDs). As many individuals with ADRDs have mixed pathologies (β-amyloid, pathologic tau, cerebrovascular disease, vascular brain injury, pathologic TDP-43, hippocampal sclerosis, Lewy bodies), there is interest in determining how accurately we can infer these pathologic changes from clinical data, biofluid assays (e.g., CSF), and neuroimaging. Here we evaluated automated machine learning models trained on data curated by the AD Sequencing Project Phenotype Harmonization Consortium (N=7,894 individuals), to predict 26 autopsy-confirmed neuropathological outcomes. Predictors included in vivo clinical and cognitive composite scores, brain measures from 3D structural MRI and diffusion tensor imaging, image-derived measures of white matter hyperintensities (WMH), and CSF biomarkers. Predictive models were trained using ensemble learning with stratified cross-validation. We assessed performance using Spearman’s rank correlation and Matthews correlation coefficient, to accommodate co-occurring pathologic changes. The added value of neuroimaging and CSF versus clinical features alone was quantified. Braak stage was among the most consistently predicted outcomes. CSF biomarkers best predicted β-amyloid and tau pathology, but diffusion MRI metrics best captured vascular brain injury and white matter injury, and outperformed clinical and cognitive measures and anatomical MRI in predicting Lewy body disease. Anatomical measures from structural MRI outperformed standard clinical assessments in assessing neurodegeneration and hippocampal sclerosis, and WMH complemented cognitive measures in predicting TDP-43 pathology. These results establish a baseline for comparing modalities for inferring neuropathology.

Predicting the trajectory of clinical decline in aging individuals is a pressing challenge, especially for people with mild cognitive impairment, Alzheimer’s disease, Parkinson’s disease, or vascular dementia. Accurate predictions can guide treatment decisions, identify risk factors, and optimize clinical trials. In this study, we compared two deep learning approaches for forecasting changes, over a 2-year interval, in the Clinical Dementia Rating scale ‘sum of boxes’ score (sobCDR), as a continuous outcome (regression). This is a key metric in dementia research and clinical trials, and scores range from 0 (no impairment) to 18 (severe impairment). To predict decline, we trained a hybrid convolutional neural network (CNN) that integrates 3D T1-weighted brain MRI scans with tabular clinical and demographic features (including age, sex, body mass index (BMI), and baseline sobCDR). We benchmarked its performance against AutoGluon, an automated multimodal machine learning framework that selects an appropriate neural network architecture (an ‘autoML’ approach). We evaluated the models using data from 2,319 unique participants drawn from three independent cohorts—ADNI, OASIS-3, and NACC. For each participant, we used one T1-weighted brain MRI scan along with corresponding clinical and demographic information. Our results demonstrate the importance of combining image and tabular data in predictive modeling for this clinical application. Deep learning algorithms can fuse information from image-based brain signatures and tabular clinical data, with potential for personalized prognostics in aging and dementia. Rather than concluding that multimodal fusion uniformly improves performance, our results show that deep learning applied to volumetric MRI data may struggle to add predictive value, particularly when clinical covariates explain substantial variance and provide a strong baseline. In other conditions and tasks, it may help to have a hybrid system that can learn from both data types, and their relative value may be different. Conversely, AutoML-based multimodal fusion provides a robust baseline when tabular data already provide strong predictive value for the task. These insights clarify how different multimodal strategies could be selected in clinical prognostic applications.

Cortical brain morphology in early-onset psychosis (EOP; age of onset < 19 years) is poorly understood, partly due to recruitment constraints linked to its low incidence. We pooled T1-weighted magnetic resonance imaging (MRI) data from 387 adolescents with EOP (mean age=16.1±1.5; 49.6% female) and 338 healthy controls (CTR; mean age=15.8±1.9, 54.4% female) from nine research sites worldwide. Using harmonized processing protocols with FreeSurfer, we extracted cortical brain metrics from 34 bilateral regions. Univariate regression analysis revealed widespread lower bilateral cortical thickness (left/right hemisphere: d=-0.36/-0.31), surface area (left/right: d=-0.42/-0.41), cortical volume (left/right: d=-0.58/-0.56), and Local Gyrification Index (LGI; left/right: d=-0.39/-0.52) in EOP relative to CTR. Subgroup analyses showed broader and more pronounced case-control differences in early-onset schizophrenia for area, volume, and LGI. We found no associations with antipsychotic medication use, illness duration, age of onset, or positive symptoms. Negative symptoms were related to smaller left lingual volumes (partial r=-0.21; p FDR =0.014) and antidepressant users had smaller area (d=-0.43; p FDR =0.034) and volume (d=-0.50; p FDR =0.003) of the right rostral anterior cingulate compared to non-users. Cortical alterations in EOP showed a similar pattern to those observed in prior studies on adults with schizophrenia (SCZ; r=0.62) and bipolar disorders (BD; r=0.61). However, surface area alterations were overall 1.5 times greater for EOP than adult SCZ and 4.6 times greater than adult BD. In the largest study of its kind, we observed an extensive pattern of cortical alterations in adolescents with psychotic disorders, highlighting the potential impact of aberrant neurodevelopment on cortical morphology in this clinical group.

Extensive neuroimaging research in temporal lobe epilepsy with hippocampal sclerosis (TLE-HS) has identified brain atrophy as a disease phenotype. While it is also related to a complex genetic architecture, the transition from genetic risk factors to brain vulnerabilities remains unclear. Using a population-based approach, we examined the associations between epilepsy-related polygenic risk for HS (PRS-HS) and brain structure in healthy developing children, assessed their relation to brain network architecture, and evaluated its correspondence with case-control findings in TLE-HS diagnosed patients relative to healthy individuals. We used genome-wide genotyping and structural T1-weighted MRI of 3826 neurotypical children from the Adolescent Brain Cognitive Development (ABCD) study. Surface-based linear models related PRS-HS to cortical thickness measures, and subsequently contextualized findings with structural and functional network architecture based on epicentre mapping approaches. Imaging-genetic associations were then correlated to atrophy and disease epicentres in 785 patients with TLE-HS relative to 1512 healthy controls aggregated across multiple sites. Higher PRS-HS was associated with decreases in cortical thickness across temporo-parietal as well as fronto-central regions of neurotypical children. These imaging-genetic effects were anchored to the connectivity profiles of distinct functional and structural epicentres. Compared with disease-related alterations from a separate epilepsy cohort, regional and network correlates of PRS-HS strongly mirrored cortical atrophy and disease epicentres observed in patients with TLE-HS and were highly replicable across different studies. Findings were consistent when using statistical models controlling for spatial autocorrelations and robust to variations in analytic methods. Capitalizing on recent imaging-genetic initiatives, our study provides novel insights into the genetic underpinnings of structural alterations in TLE-HS, revealing common morphological and network pathways between genetic vulnerability and disease mechanisms. These signatures offer a foundation for early risk stratification and personalized interventions targeting genetic profiles in epilepsy.

Structural variants, including copy number variants (CNVs), confer substantial risk for neurodevelopmental and psychiatric disorders (NPDs), yet whether their cortical effects relate to those observed in the psychiatric conditions they predispose to remains unclear. Here, we present the first systematic comparison of cortical phenotypes across 18 NPD-associated CNVs and aneuploidies, disorder-associated common variants, and 8 psychiatric disorders. Rare CNVs preferentially affected total surface area (SA), with 11-fold larger effects than psychiatric diagnoses, while NPDs preferentially affected mean cortical thickness (CT), with most CT effects observed in medicated subgroups, suggesting non-genetic contributions. NPD-associated common variants showed enrichment in SA but not CT associations. Regionally, both rare and common genetic variants showed larger effects in sensorimotor regions, aligning with the sensorimotor-to-association cortical gradient as well as regional heritability estimates. In contrast, psychiatric diagnoses showed larger effects in association regions. Individual NPD-associated variants were evenly split between those increasing and decreasing surface area. This heterogeneity likely explains why aggregating variants using polygenic scores shows only weak associations with SA. Overall, cortical signatures of psychiatric diagnoses diverge from those associated with genetic risk. Genetic variants preferentially impact SA and sensorimotor regions through early developmental mechanisms, while psychiatric diagnoses are associated with CT and association regions likely reflecting medication, illness chronicity, and environmental factors.