BACKGROUND:C-reactive protein (CRP) has emerged as a promising peripheral biomarker of neuroinflammatory processes implicated in the pathophysiology of major depressive disorder (MDD). We combined structural magnetic resonance imaging (MRI) with resting-state functional MRI to examine plasma CRP-related cortical thickness and resting-state functional connectivity (RSFC) changes. METHODS:Seventy-six patients with MDD and 65 healthy controls (HCs) were included in this study. The association between whole-brain cortical thickness and plasma CRP levels in the total sample (MDD+ HC, n = 141) was investigated. Seed-to-voxel RSFC analysis was performed using plasma CRP level-related cortical regions as seeds. RESULTS:Compared to the HC group, the MDD group showed significantly higher plasma CRP levels (p = 0.039). Higher plasma CRP levels were associated with cortical thinning in the prefrontal cortices and precentral gyrus, and cortical thickening in the left postcentral and right medial orbitofrontal gyrus. RSFC analysis showed a lower RSFC between the left dorsolateral prefrontal cortex and bilateral insula and between the precentral gyrus and superior parietal cortex, and higher RSFC between the dorsomedial prefrontal cortex and postcentral gyrus in MDD compared to HCs. CONCLUSIONS:Our findings suggest that systemic inflammation may be associated with structural and functional disruption of emotion regulation and cognitive control networks in MDD.
BACKGROUND:The long-term effects of antipsychotic treatment on brain structure in schizophrenia remain uncertain. Previous meta-analyses have primarily compared patients with controls rather than examining dose-related associations between cumulative exposure and brain structure. OBJECTIVE:To quantify the association between cumulative antipsychotic exposure (chlorpromazine equivalents) and structural brain changes measured by MRI in patients with schizophrenia. METHODS:Following prospective registration (PROSPERO CRD420261352526), 19 studies involving 1607 patients were included. Pearson's r or standardized β coefficients were pooled using random-effects models with the DerSimonian-Laird estimator. Subgroup analyses were conducted for the primary outcome. RESULTS:Higher cumulative antipsychotic exposure was associated with lower global gray matter volume (10 studies, n = 993; r = -0.259, 95% CI -0.337 to -0.179; p < 0.001). Publication bias was detected for this outcome (Egger's test p = 0.00574). Negative associations were also observed for global cortical thickness (r = -0.293), frontal lobe volume (r = -0.339), temporal lobe volume (r = -0.351), and total brain volume (r = -0.310). Lateral ventricular volume showed a positive association (r = 0.228), whereas white matter volume was not significantly associated. The dose-related association with gray matter volume was consistent across study designs and illness stages, with no significant subgroup differences. CONCLUSIONS:Higher cumulative antipsychotic exposure was modestly associated with greater structural brain alterations. However, these observational findings do not establish causality because medication exposure is closely linked to illness progression. Interpretation is further limited by publication bias, the small number of studies, and heterogeneity across secondary outcomes. Well-designed prospective longitudinal studies are needed to better distinguish medication effects from disease-related changes.
Severe worry and anxiety in late life are linked to increased risk of Alzheimer’s disease (AD), but underlying neural mechanisms remain unclear. Severe worry is associated with reduced hippocampal volume. One proposed mechanism is stress-related glutamate excitotoxicity, which may contribute to hippocampal atrophy and cognitive decline. This pilot study examined whether anxiety, worry, and cognitive function are associated with markers of glutamate excitotoxicity, specifically hippocampal glutamate and N-acetyl aspartate (NAA). Eighteen older adults with varying worry underwent 7T magnetic resonance spectroscopy of left and right hippocampus and clinical and cognitive assessments. Linear regression analyses examined associations between anxiety, cognitive performance, and hippocampal metabolites, controlling for volume. Exploratory analyses assessed additional metabolites, including creatine, gamma-aminobutyric acid (GABA), glutathione, and myo-inositol. Greater worry severity and lower overall cognitive function were associated with reduced hippocampal NAA levels. Higher global anxiety was associated with lower hippocampal glutamate. Exploratory analyses revealed additional relationships between other metabolites and both mood symptoms and cognition. These preliminary findings suggest hippocampal metabolic markers may play a role in late-life anxiety and worry. However, results provide only partial support for glutamate excitotoxicity as a mechanism linking anxiety to cognitive impairment, highlighting the need for larger studies.
BACKGROUND:Mitochondrial dysfunction is a hallmark of Alzheimer's disease (AD), yet specific molecular targets remain to be fully characterized. METHODS:A summary-data-based Mendelian randomization (SMR) framework integrated AD genome-wide association study (GWAS) statistics (39,918 cases) with blood DNA methylation quantitative trait loci (mQTL), gene expression (eQTL), and protein (pQTL) data for 1136 mitochondria-related genes. Associations were assessed using Bayesian colocalization and HEIDI testing. Tissue relevance was evaluated in four brain regions (hippocampus, amygdala, cortex, frontal cortex) using GTEx and external transcriptomic datasets. RESULTS:Screening identified eight candidates supported across blood mQTL and eQTL layers. Stepwise central nervous system (CNS) evaluation singled out biphenyl hydrolase-like (BPHL) as the consistent candidate. Higher genetically predicted BPHL expression was associated with reduced AD risk across the hippocampus (OR=0.920, 95% CI 0.873-0.970), amygdala (OR=0.925, 95%CI 0.880-0.973), cortex (OR=0.943, 95% CI 0.908-0.978), and frontal cortex (OR=0.938, 95%CI 0.901-0.976). These findings aligned with protein-protein interactions connecting BPHL to respiratory complexes and lower BPHL expression in independent AD brains. Functional enrichment converged on oxidative phosphorylation pathways. CONCLUSIONS:By integrating multi-omics data with tissue-specific validation, this study nominates BPHL as a consistent protective candidate in the brain. These findings provide genetic support for mitochondrial molecular perturbations in AD, offering insights for future validation.
Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by atypical large-scale brain network organization. Using the comprehensive Autism Brain Imaging Data Exchange (ABIDE-I and ABIDE-II; n = 2013 participants), this study systematically evaluated three representative graph neural network (GNN) architectures for ASD classification based on resting-state functional MRI. The Graph Convolutional Network (GCN), Graph Attention Network (GAT), and Self-Attention Graph Pooling (SAGPool) models were applied to graph representations of functional connectivity among 200 brain regions, following standardized preprocessing and ComBat harmonization. Across 100 repetitions of 10-fold stratified cross-validation, all three models achieved consistent and reproducible accuracy (∼61% balanced accuracy), establishing a robust performance benchmark for connectome-based ASD classification. Permutation-based feature importance analysis revealed convergent brain regions across models, including the bilateral thalamus, right superior temporal gyrus, right middle occipital gyrus, and bilateral precuneus, which are involved in sensory integration, social cognition, and default mode network functioning. These findings indicate that distinct GNN architectures converge on common neurobiological signatures of ASD, highlighting their potential for reliable brain network modeling and biomarker discovery in psychiatry research.
Hemispheric asymmetry is a fundamental feature of the healthy human brain, with consistent differences observed across individuals. However, disruptions in asymmetry are associated with a range of disorders, including schizophrenia. Large-scale studies reported subtle alterations in subcortical asymmetry in schizophrenia. However, these findings are primarily based on macrostructural measures, and the extent to which tissue-level changes happen remains unclear. In this work, subcortical microstructural asymmetry in schizophrenia was investigated using quantitative T1 (qT1) mapping. A total of 23 patients with schizophrenia spectrum disorders (SSD) and 12 healthy controls underwent 3T-MRI scanning. Asymmetry indices were calculated for several subcortical regions, and group differences were assessed using multiple linear regression, controlling for age and sex. A trend toward diminished caudate asymmetry in the SSD group compared to healthy controls was observed. However, this did not survive multiple-comparisons correction (β = 0.032, punadjusted = 0.01, pFDR = 0.07). Additionally, no significant relationships were found between asymmetry indices and symptom severity or antipsychotic dose equivalents. Our study shows a trend towards microstructural asymmetry in SSD, in line with the limited existing literature. These findings highlight the need for larger, well-powered studies to further elucidate the role of tissue-level asymmetry in the pathophysiology of SSD.
The common form of dementia is known as Alzheimer’s disease (AD) that mostly affects the elderly individuals also, it is incurable and degenerative brain disease. According to progression level of AD, it led to memory loss initially and it affects functionality in different degrees. Even before the appearances of symptoms, AD cause severe damage in brain tissue and cell. Identifying AD at its earliest stages offers affected patients chance to get together with their healthcare providers, other support network members and loved ones to create personalized care plans, ensuring a more effective method to handle their condition. Here, ResSpikingNet has been developed to detect AD earlier using MRI image. Initially, the input image is given into the pre-processing process that is performed using median filter. After that, brain area segmentation is done by employing O-Segnet. Furthermore, the feature extraction techniques are employed to extract specific features, including ALDP with info gain, entropy, and statistical features. Finally, early detection of AD is done utilizing ResSpikingNet that is developed by the combination of Spiking ResNet and DRN. In addition, developed ResSpikingNet obtained maximal value of accuracy as 91.648%, sensitivity as 91.780%, and specificity as 91.938%.
Obsessive-compulsive disorder (OCD) exhibits substantial clinical heterogeneity that may reflect underlying neurobiological diversity. Neuroimaging-based subtyping may advance precision psychiatry by identifying biologically distinct subgroups with differential treatment responses. This study systematically synthesized evidence from functional neuroimaging subtyping studies in OCD to identify reproducible neurobiological subtypes, characterize their clinical profiles, and establish a consensus-based classification framework. We reviewed 40 original studies employing machine learning, clustering, normative modeling, or classification approaches, encompassing approximately 8,150 patients. Consensus clustering identified three reproducible neurobiological subtypes. The Limbic-Hyperactive subtype, comprising approximately 40% of patients, exhibited amygdala and insula hyperconnectivity, elevated anxiety levels, predominant contamination and washing symptoms, and favorable response to cognitive-behavioral therapy. The Fronto-Striatal-Hypoconnected subtype, comprising approximately 35% of patients, demonstrated reduced orbitofrontal-striatal connectivity, cognitive inflexibility, predominant checking and ordering symptoms, and a favorable response to selective serotonin reuptake inhibitors. The Global-Disrupted subtype, comprising approximately 25% of patients, exhibited widespread connectivity disruption, greater symptom severity, and poor treatment response. Support vector machine classification achieved 81.5% accuracy for subtype assignment, though classification of OCD versus healthy controls showed limited generalizability in multisite settings (AUC 0.567-0.673). These findings support a neuroimaging-based framework for personalized treatment selection but require prospective validation.
The pathological mechanisms of depression are not yet fully clear, and the causative factors remain somewhat ambiguous. Clinical diagnosis of depression is often challenging and complex, frequently leading to misdiagnosis and missed diagnoses. Combining deep learning with resting-state fMRI can quantify the degree of abnormal brain function caused by depression and automatically screen for discriminative features that aid in the classification and identification of depression, which may serve as hypothesis-generating discriminative features within the current dataset, providing candidate neuroimaging signatures that warrant further investigation. This paper proposes a cross-site fMRI data analysis framwork for depression. First , it contains a graph deep learning-based auxiliary diagnostic method, which fully leverages the topological structure of brain networks to achieve higher classification accuracy compared to existing models, with interpretable results. Building on this network, the framework also contains a domain-adversarial-based cross-site semi-supervised transfer method is proposed, making full use of multi-site data to analyze depression-related brain networks and ROIs.Finally, based on cross site data, the distribution of brain networks and brain regions was discussed. The research findings are consistent with existing studies, confirming the reliability of this method. Furthermore, we validated the cross-dataset generalizability of our framework on an independent OpenNeuro dataset, where adversarial transfer consistently outperformed direct transfer, demonstrating the potential of our approach to generalize beyond the original consortium.
Autism Spectrum Disorder (ASD) in toddlers is characterized by neurodevelopmental deficits while its early detection remains challenging due to the lack of specific biomarkers and developmental variability. Therefore, a TinyDINO VHaar based Bi-directional Factorization with Chameleon optimized LightASDNet (TiDI-ASDNet) is proposed. In this framework, input images are initially pre-processed using Optimized Hierarchical Guided Image Filter (OHGF) to denoise and outlier removal, while Self-Distillation with No Labels version 2-Network Vector of Locally Aggregated Descriptors (DINOv2-NetVLAD) extracts the global visual patterns. Simultaneously, the input texts are pre-processed using One-Hot SMOTE (OHS) for easier interpretation. Besides, the standard structures interrupt syntactic/structural parsing due to autistic language corpora which clusters n-gram pattern, weakening the pragmatic deficits. To address these issues a Hybrid Tiny Encoder based HaarNet (TE-HNet) that extracts discriminative questionnaire features and maintains the lexical stringency. Since, the manifold torsion occurs due to discrepancy in encoding levels from non-isomorphic latent structures with non-diffeomorphic mapping, Bi-directional Encoder based Cross Factorization (BiE-xF) is employed and it learns about the shared semantics of visual features and behavioral linguistic features, which reduces Heteroscedastic Ambiguity. Besides, the motor stimming behaviour produces recurring self-stimulatory motor patterns that confound temporal alignment in vision-based ASD models Thus, Modified Chameleon optimized LightASDNet (MC-LAN) is presented for classifying ASD and non-ASD thereby mitigates Dyspraxic Gait Aberrations. Simulations revealed the robustness of the framework with 99.2% AUC and 98.2% accuracy.
Objective This systematic review investigates brain changes in youths with anxiety disorders following cognitive behavioral therapy (CBT) and neural markers that predict CBT responses. Methods We conducted a systematic search using the electronic databases PubMed, Web of Science, and ProQuest. The inclusion deadline was set to October 27, 2025. We included fifteen peer-reviewed neuroimaging studies that examined the effects of CBT in youths under 19 years old with a primary clinical diagnosis of an anxiety disorder based on DSM-5 criteria. Results Although the existing literature is marked by substantial diversity in methods and outcomes, task-related neural response in the anterior cingulate cortex (ACC, 2/8, 25.0%), insula (1/8, 12.5%) increased from pre to post CBT and these changes were further correlated with clinical symptom improvements. Moreover, CBT outcomes were predicted by pre-treatment activity or connectivity in the ACC and amygdala (3/13, 23.0%). A smaller proportion of studies (2/13, 15.3%) found that activity or connectivity in the insula, precuneus/cuneus, postcentral gyrus, and activity or structure in the nucleus accumbens (NAcc) predicted response to CBT. The low consistency of these findings was driven by methodological variability, low reliability of the neural markers, and relatively small sample sizes. Conclusions This review highlights promises of neural predictors and outcomes to enhance anxiety disorder treatments in children and adolescents, facilitating future personalized and effective CBT. Beyond this initial promise, the field is hindered by methodological inconsistencies and limited replications. While longitudinal and personalized approaches are important next steps, the central challenge remains: identifying neural markers that are both reliable and robust.
State-dependent neural dynamics are central to models of large-scale brain regulation in neurodevelopmental and psychotic disorders, but it remains unclear whether autism spectrum disorder (ASD) and schizophrenia (SCH) show altered oscillatory activity across internally and externally oriented cognitive states. This study examined EEG spectral power across resting, interoceptive, and cognitive conditions in adults with ASD, SCH, and matched neurotypical controls. Absolute and relative power were calculated for canonical frequency bands to distinguish overall power differences from spectral redistribution; group and condition effects were analysed while controlling for covariates. Robust condition effects emerged across frequency bands. Absolute delta power decreased during interoception compared to rest and task, while relative delta power was lowest during rest, suggesting spectral reweighting. The most prominent group effect was a condition-invariant elevation of theta power in SCH, suggesting a baseline shift with preserved state-dependent modulation. The ASD group showed modest detectable spectral differences. Condition-dependent modulation patterns were broadly similar across groups. Alpha, beta, and gamma activity were primarily condition-driven with less group differences. Group-by-condition interactions, altered state-dependent oscillatory modulation were not reliably detectable. Absolute and relative EEG power captured partly distinct spectral characteristics across conditions and groups.
BACKGROUND:Loss of dopaminergic neurones and iron deposition in the substantia nigra pars compacta (SNpc) are two major pathological hallmarks of Parkinson's disease (PD). Such changes can be visualised by advanced techniques including neuromelanin-sensitive MRI (NM-MRI) and quantitative susceptibility mapping (QSM). This systematic review investigates the diagnostic performance and methodological development of the integrated use of NM-MRI and QSM in PD. METHODS:The systematic search was performed in four databases (Scopus, PubMed, ScienceDirect, and Web of Science) according to the PRISMA 2020 guidelines until July 2026. Bias was assessed using QUADAS-2 and certainty of evidence was assessed using GRADE. RESULTS:Seventeen studies with 2228 participants were included. Combined NM-MRI and QSM consistently showed reduced neuromelanin volume/contrast and increased iron deposition in the SNpc of PD patients compared to healthy controls. Multimodal integration yielded a significant improvement in diagnostic accuracy (AUC values 0.86-0.99), and was able to successfully differentiate PD. Recent methodological advances included simultaneous acquisition sequences (e.g. MTC-GRE, STAGE, setMag) and AI-driven automated segmentation, which led to significantly reduced scan times and improved reproducibility. CONCLUSION:The combination of NM-MRI and QSM has a synergistic effect and provides powerful complementary biomarkers for the diagnosis and differential diagnosis of PD.
BACKGROUND:Attention-Deficit/Hyperactivity Disorder (ADHD) often persists into adulthood and causes significant functional impairments. Resting-state functional MRI (rs-fMRI) offers valuable insights into intrinsic brain connectivity; however, connectivity (FC) alterations specific to adults with ADHD remain unclear. This systematic review aimed to synthesize rs-fMRI studies comparing adults with ADHD and healthy controls, identify consistent FC patterns, and assess methodological quality. METHODS:Following PRISMA guidelines, PubMed, Embase, and Scopus were searched for rs-fMRI studies involving adults diagnosed with ADHD per DSM-IV/DSM-5 criteria. Eligible studies examined FC differences between adults with ADHD and healthy controls. Data extraction included study characteristics, imaging parameters, and FC methodologies. Study quality was assessed using the Newcastle-Ottawa Scale. RESULTS:Eight studies (n = 529; 264 ADHD, 265 controls) met inclusion criteria. FC analyses employed Seed-Based Analysis (SBA), Independent Component Analysis (ICA), and graph theory methods. Despite methodological heterogeneity, consistent FC alterations were observed: increased FC in the default mode (DMN), visual (VN), and central executive (CEN) networks, and decreased FC in the ventral attention (VAN) and somatomotor (SMN) networks. Alterations were predominantly left-hemispheric. CONCLUSION:Adults with ADHD exhibit distinct rs-FC disruptions, mainly involving DMN, VAN, CEN, and SMN. Standardized analytic approaches and subtype-specific analyses are needed for improved clinical relevance.
BACKGROUND AND OBJECTIVES:Objective, scalable markers for monitoring treatment response in treatment-resistant depression remain limited. This pilot study evaluated the feasibility of repeated smartphone-based facial landmark analysis during repetitive transcranial magnetic stimulation (rTMS) and explored associations with clinical change. METHODS:Six patients with treatment-resistant depression underwent a standardized 20-session rTMS course. At each session, smartphone videos were recorded during five facial expressions: neutral, smile, surprised, angry, and sad. Automated facial landmark tracking was used to extract facial motion metrics. Hamilton Depression Rating Scale and Hamilton Anxiety Rating Scale scores were assessed at baseline and after treatment. RESULTS:Depression and anxiety scores decreased from baseline to end-of-treatment in this uncontrolled pilot cohort. Facial analyses showed nominal expression-dependent longitudinal changes, most consistently during the sad expression. Exploratory correlations identified preliminary nominal associations between selected facial motion changes and symptom-score reductions; however, these findings did not survive correction for multiple comparisons. CONCLUSION:Repeated smartphone-based facial landmark analysis during rTMS was technically and procedurally feasible and generated longitudinal facial motion metrics. Larger controlled studies are needed to determine whether these changes reflect treatment-related effects, nonspecific longitudinal factors, or clinically meaningful behavioral signals.
Re-experiencing symptoms in posttraumatic stress disorder (PTSD) often involve vivid visual imagery, yet occipital resting-state functional connectivity (rsFC) remains understudied, particularly in women exposed to interpersonal violence. Clarifying these neural mechanisms may advance understanding of sensory and contextual processing disturbances that contribute to intrusive memories. Notably, occipital rsFC has not been evaluated in relation to re-experiencing using the Posttraumatic Stress Diagnostic Scale (PDS), a validated self-report measure that supports fine-grained symptom-level analyses. Sixty-three participants meeting DSM-IV criteria for PTSD completed clinical assessments and resting-state fMRI. PTSD severity was measured with the Clinician-Administered PTSD Scale (CAPS-IV) and the PDS. We conducted seed-based voxelwise rsFC analyses using bilateral visual cortex seeds (V1, V2, V3) and tested associations with re-experiencing symptoms. Greater V3-left cerebellar connectivity was positively associated with PDS re-experiencing subscores (t(61)=5.32, pFWE < .05). Item-level analyses suggested this effect was driven by endorsements of nightmares (t(61)=4.18, pFWE < .05) and physiological reactivity (t(61)=4.77, pFWE < .05). Increased V3-cerebellar connectivity may reflect heightened coupling between visual and sensorimotor/regulatory systems supporting PTSD-related re-experiencing. These interactions may relate more specifically to re experiencing symptoms than to global PTSD severity, suggesting a potential symptom targeted neural marker in trauma-exposed females.
BACKGROUND:Non-suicidal self-injury (NSSI) is highly prevalent among adolescents and a strong risk factor for suicidal behavior. However, its neurobiological mechanisms remain incompletely understood, and traditional treatments face challenges of delayed efficacy and high relapse rates. OBJECTIVE:This review synthesizes advances in the neurobiology of non-suicidal self-injury (NSSI) from multimodal studies and discusses prospects for precision, mechanism-based interventions. METHODS:We focused on integrating findings from multimodal neuroimaging (fMRI, DTI, MRS, EEG/MEG), neuroimmunology, and the gut-brain axis literature. KEY FINDINGS:NSSI arises from dynamic imbalances across interconnected circuits: (1) a central pain disinhibition/attenuation phenotype; (2) prefrontal-limbic connectivity abnormalities underlying emotion and impulse dysregulation; (3) neuroimmune-metabolic dysregulation linking peripheral inflammation to brain connectivity and metabolism; and (4) emerging gut-brain axis influences on neural function. CONCLUSION AND PROSPECTS:NSSI is a disorder of "brain-body" interaction. Advancing its treatment requires longitudinal multi-omics data and advanced techniques (e.g., closed-loop neuromodulation) to validate mechanisms and propel the field toward precision medicine models.
Dementia is a neurodegenerative disorder marked by cognitive decline affecting daily activities, including various conditions, however, early detection and classification of dementia using EEG signals is crucial. Hence, a novel "Dynamic Bayesian Morlet-Hilbert Consensus Spectrum Optimization" is proposed to enhance the accuracy of dementia diagnosis, specifically for Alzheimer's Disease and Mild Cognitive Impairment. Additionally, Cholinergic System Dysfunction, characterized by non-uniform brain degeneration, complicates EEG feature extraction due to heterogeneous neural activity patterns, and existing models struggle to accurately interpret these patterns. Thus, a novel Adaptive Morlet Consensus Beamforming Transform is introduced to capture heterogeneous neural activity patterns and adapt to complex brain regions, thereby improving the identification of biomarkers associated with cholinergic system dysfunction. Furthermore, existing models often overlook the striatum due to its intricate structure and the cholinergic system's impact on cognitive functions, making it challenging to accurately assess cholinergic dysfunction. So, a novel Multilayer Adaptive Dynamic Causal Frequency Decomposition is introduced to detect subtle variations in brain function, improving the identification of early-stage dementia-related changes linked to cholinergic system dysfunction. Moreover, the striatum, a complex structure with subregions, exhibits varying impacts of cholinergic dysfunction, causing significant variability in cognitive and motor functions. Therefore, a novel Bayesian Hilbert-ElasticNet Spectrum Optimization is introduced to effectively capture the spectral and temporal dynamics of EEG signals to classify subtle variations in dementia. The results demonstrate that the methodology significantly improves classification performance, achieving an accuracy of 98.23%, along with precision and F1-Score rates of 98.84% and 97.52%, respectively, while minimizing RMSE and enhancing processing efficiency.
