Brain Imaging and Behavior最新文献

{"title":"Altered brain function and its spatial correlations with neurotransmitter maps in anterior shoulder pain patients.","authors":"Zhijun Diao, Huan Li, Yujia Li, Chengxiang Liu, Yunsong Zheng, Jingdong Lu, Yintao Liu, Zi Wei, Peng Liu, Haifa Qiao","doi":"10.1007/s11682-026-01171-5","DOIUrl":"10.1007/s11682-026-01171-5","url":null,"abstract":"<p><p>Neuroimaging studies have reported that chronic pain may be associated with brain functional changes and related molecular mechanisms. However, there are few studies in anterior shoulder pain (ASP). Thus, we aimed to assess the possible brain functional changes and their associations with neurotransmitter receptor maps in ASP patients. Resting-state functional magnetic resonance imaging data were obtained from 27 ASP patients and 20 healthy controls (HCs). Regional homogeneity (ReHo) and fractional amplitude of low-frequency fluctuation (fALFF) methods were used to examine altered brain functions. JuSpace toolbox was employed for evaluating the spatial correlations between ReHo and fALFF changes and neurotransmitter receptor maps. Regression analysis was conducted to explore the associations between neuroimaging findings and clinical characteristics in ASP patients. Furthermore, we investigated reliable biomarkers to distinguish ASP patients from HCs using receiver operating characteristic (ROC) curve analysis. ASP patients had abnormal ReHo/fALFF in the middle temporal gyrus, putamen, caudate, pallidum, precentral gyrus, medial prefrontal cortex, precuneus, posterior cingulate cortex and inferior temporal gyrus. ReHo/fALFF alterations in ASP patients were associated with the serotonin, dopamine, GABA, choline, glutamate and norepinephrine neurotransmitter systems. Visual analog scale was positively correlated with ReHo in the right insula. Moreover, ROC analysis showed higher classification accuracy in the middle temporal gyrus, precentral gyrus and putamen. Our study investigated abnormal brain function and associated neurotransmitter receptor maps in ASP patients, which may contribute to improving our understanding of the pathophysiological mechanisms in ASP.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 6","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13680527/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149223159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The role of neighborhood socioeconomic environment in the association between glycemic control and the developing brain.","authors":"Anisha Chandra, Eustace Hsu, Shan Luo","doi":"10.1007/s11682-026-01211-0","DOIUrl":"10.1007/s11682-026-01211-0","url":null,"abstract":"<p><p>Metabolic health and socioeconomic environments may jointly shape neurodevelopment during adolescence, a sensitive period of brain maturation. This cross-sectional study investigated overall and neighborhood socioeconomic deprivation moderated associations between glycemic control and brain structure in youth. Data were analyzed from 705 healthy 10-13-year-olds participating in the Adolescent Brain and Cognitive Development (ABCD) Study® across 21 sites in the United States. Glycemic control was assessed using hemoglobin A1c (HbA1c), brain structure was evaluated via magnetic resonance imaging (MRI), and neighborhood deprivation was measured with the Area Deprivation Index (ADI). Mixed-effects models examined associations between HbA1c and brain structural measures while controlling for sociodemographic covariates, with additional analyses stratified by ADI tertiles. Higher HbA1c was associated with lower mean cortical thickness and smaller total cortical gray matter volume. Each one percent increase in HbA1c corresponded to a 0.024 mm reduction in mean cortical thickness and a 9,611 mm³ reduction in total cortical gray matter volume. Regionally, higher HbA1c was associated with thinner cortex and smaller gray matter volumes primarily within frontal, cingulate and occipital areas. Exploratory analysis found that higher HbA1c was associated with lower total cortical gray matter volume among youth living in medium and high ADI neighborhoods, but not low ADI neighborhoods, whereas higher HbA1c was associated with lower mean cortical thickness only among youth living in high ADI neighborhoods. These findings highlight the interplay between metabolic health and neighborhood deprivation, supporting a model in which socioeconomic disadvantage may magnify the neurodevelopmental impact of metabolic variation, even within the non-diabetic range.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13668964/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149179100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Duration-related regional patterns of gray matter alterations in hypertension and their cognitive implications.","authors":"Xile Zhu, Yanbo Liang, Xiaoqing Yuan, Yiwen Gao, Luyao Wang, Jiehui Jiang, Han Lv","doi":"10.1007/s11682-026-01210-1","DOIUrl":"https://doi.org/10.1007/s11682-026-01210-1","url":null,"abstract":"<p><p>Hypertension is a major modifiable vascular risk factor for cognitive decline. However, how regional gray matter (GM) alterations differ according to hypertension duration remains incompletely understood. We hypothesized that different brain regions would exhibit heterogeneous GMV profiles across hypertension-duration groups and that regional GMV alterations would be associated with cognitive performance. Structural MRI and cognitive data were obtained from 487 participants with hypertension and 165 normotensive controls. Participants with hypertension were stratified by disease duration (≤ 4, 5-11, and ≥ 12 years). Voxel-based morphometric analysis was used to identify whole-brain GMV alterations across groups. Regional associations with hypertension duration and cognitive performance were subsequently examined, adjusting for age, sex, total intracranial volume, and, where appropriate, education. Regression models further assessed whether GMV-cognition associations differed across hypertension-duration groups. Three cross-sectional duration-related GMV patterns were identified: early and relatively stable alterations, duration-sensitive alterations, and intermediate-to-long-duration alterations. Longer hypertension duration was associated with lower GMV in the right rectus, left middle temporal, and left angular gyri, and higher GMV in the bilateral putamen and right insula after FDR correction. Only greater left thalamic GMV remained associated with higher MoCA scores after correction. Interaction analyses suggested duration-group differences in GMV-cognition associations in the right insula and right precuneus. Hypertension was associated with heterogeneous, region-specific GMV alteration across duration groups rather than a uniform cumulative atrophy pattern. The direction and cognitive relevance of these alterations varied across regions, highlighting the importance of considering hypertension duration when evaluating brain structural alterations. Longitudinal studies are needed to clarify their temporal evolution.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149165743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterizing hippocampal activation during pattern separation with multi-echo fMRI in cognitively normal older adults.","authors":"Kevin Grant Solar, Sriranga Kashyap, Nicolas Deom, Kâmil Uludağ, Ljubica Zotovic, Krista Lanctôt, Sandra E Black, Mary Pat McAndrews","doi":"10.1007/s11682-026-01208-9","DOIUrl":"https://doi.org/10.1007/s11682-026-01208-9","url":null,"abstract":"<p><p>Hippocampal hyperactivity, detected with mnemonic similarity task (MST) fMRI, is both a promising marker of risk and a probable driver of degeneration in Alzheimer's disease (AD). However, implementation of such fMRI in interventional research is limited by long scan times and the absence of a framework for identifying prospective participants with hyperactivity who could be targeted with therapeutic approaches aimed at reducing this activity. We used multi-echo fMRI during the MST, an episodic memory task previously shown to engage hippocampal subfields CA3 and dentate gyrus (DG), to characterize activity profiles in 33 cognitively normal older adults. We compared MST-related CA3 and DG activation and behavioural performance across subsets of eight task runs. Behavioural and activation effects were equivalent for the first set of four runs (Runs-A) and the second set (Runs-B), suggesting that robust data can be obtained with four runs. Left and right CA3 + DG activation did not differ. We provide an exemplar framework for the derivation of a threshold for single-participant classification of elevated hippocampal activation as an indicator of AD risk in cognitively normal older individuals. Scalable (~22 min) multi-echo MST fMRI robustly captures hippocampal activation and provides a framework for individual-level characterization of hippocampal function in cognitively normal older adults. This enables individualized detection of elevated hippocampal activation as a modifiable risk factor and may guide enrolment into prevention trials targeting preclinical AD.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149165190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Altered hippocampal subfield functional connectivity and its association with back pain in axial spondyloarthritis: a multimodal neuroimaging study.","authors":"Fei Xiong, Yamin Luo, Yang Liu, Fan Lu, Yang Xue, Guiping Shen, Ye Wang, Wei Wang","doi":"10.1007/s11682-026-01207-w","DOIUrl":"https://doi.org/10.1007/s11682-026-01207-w","url":null,"abstract":"<p><p>To investigate hippocampal subfield volumes and resting-state functional connectivity (rsFC) in patients with axial spondyloarthritis (axSpA), and to assess their clinical relevance. In this cross-sectional study, 39 patients with axSpA and 41 healthy controls (HCs) underwent 3-T structural T1-weighted and resting-state functional MRI. Hippocampal subfields were segmented using FreeSurfer, and seed-based rsFC maps were computed for 12 subfields per hemisphere. Group differences were assessed with general linear models (FDR-corrected) for volumes and permutation-based family-wise error correction with threshold-free cluster enhancement for rsFC. Partial correlations adjusted for age, sex, and education tested associations between rsFC and clinical parameters (FDR-corrected). No hippocampal subfield volume differed significantly between groups after FDR correction, although patients with axSpA showed nominally larger right CA4 and bilateral GC-ML-DG volumes. Increased rsFC between the left presubiculum and left precentral gyrus/supplementary motor area (PreCG/SMA) was observed in patients relative to HCs (permutation-based FWER-corrected p < 0.05). Presubiculum-PreCG/SMA rsFC was negatively associated with overall back pain (partial r = -0.52, p = 0.001, FDR-corrected q = 0.028), the only association surviving FDR correction across all clinical parameters. No significant associations were observed with inflammatory markers or disease activity indices. axSpA is associated with hippocampal-motor cortical connectivity, with increased presubiculum-PreCG/SMA rsFC linked to lower back pain severity, whereas hippocampal subfield volumes are largely preserved. These findings support reconceptualizing axSpA as a systemic disease with measurable central nervous system involvement and pain-related network reorganization. Clinical trials register number: ChiCTR2500111024 (October 23th 2025).</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149147508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gray matter subtyping with SuStaIn reveals neuroanatomical patterns in autism spectrum disorder.","authors":"Hao Ren, Yumei Duan, Hechun Li, Sisi Jiang, Jianwei Zhang, Meropi Pavlidou, Roberto Rodriguez-Labrada, Tao Peng, Cheng Luo, Chang Liu","doi":"10.1007/s11682-026-01203-0","DOIUrl":"https://doi.org/10.1007/s11682-026-01203-0","url":null,"abstract":"<p><p>Autism spectrum disorder (ASD) is highly heterogeneous, and structural magnetic resonance imaging (MRI) studies have reported both increased and decreased gray matter volume relative to typically developing controls. This variability may obscure distinct neuroanatomical patterns in conventional group-level analyses. It therefore motivates subtyping approaches that can characterize neuroanatomical heterogeneity beyond conventional group-level analyses. We applied Subtype and Stage Inference (SuStaIn) to structural MRI data from the Autism Brain Imaging Data Exchange (ABIDE) to identify gray matter subtypes and stage-like variation in ASD. Original and sign-reversed z-score representations were modeled to capture gray matter deviations above and below the control reference. We then evaluated whether the organization of subtype-related brain regions was consistent across deviation directions. Cross-direction consistency, within-ABIDE cross-cohort separability, and structural covariance network (SCN) differences between stage groups were assessed using subtype alignment, Dice coefficients, cross-cohort machine-learning classification, and structural covariance analyses. We identified two gray matter subtype patterns in ASD that were consistent across deviation directions. Subtype 1 was centered on cerebellar-limbic-striatal regions, whereas Subtype 2 involved anterior fronto-cingulo-insular regions. In exploratory covariate-adjusted analyses, higher stage in sign-reversed-model S1 was associated with lower verbal intelligence quotient (IQ) scores, whereas S2 showed modestly higher Autism Diagnostic Observation Schedule (ADOS) scores. The two subtypes showed regionally distinct volume profiles. The two-subtype solution showed strong cross-direction agreement, and within-ABIDE cross-cohort classification supported structural separability of the aligned subtypes. In the sign-reversed model, SCN analyses further identified stage-group structural covariance differences within each subtype, with little evidence for a stable covariance difference between subtypes. This study identifies two gray matter-based ASD subtype patterns that were consistent across the original and sign-reversed z-score representations, with distinct regional volume profiles, selective clinical associations, and stage-group structural covariance differences. The findings provide a neuroanatomical framework for organizing ASD heterogeneity and for examining brain-behavior variation in future studies.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149147232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimodal MRI markers of white matter hyperintensities and their impact on cognition in cerebral small vessel disease: a systematic review.","authors":"Babina Aryal, Saikiran Pendem, Priya P S, Rajagopal Kadavigere, Priyanka, Rahul Ranjan","doi":"10.1007/s11682-026-01209-8","DOIUrl":"10.1007/s11682-026-01209-8","url":null,"abstract":"<p><p>Cerebral small vessel disease (CSVD) is a leading cause of stroke, dementia, and age-related cognitive decline, with white matter hyperintensities (WMH) being a key imaging feature in MRI. This review aims to evaluate the association between multimodal magnetic resonance imaging markers of white matter hyperintensities and cognition in cerebral small vessel disease. A systematic search of PubMed, Scopus, Embase, and Web of Science (2016-2026) was done. Studies involving adults with MRI visible CSVD markers who underwent MRI standardized cognitive assessment were eligible. Fifteen studies were included with a total of 5752 participants. Included studies used conventional and advanced MRI techniques to evaluate the association of WMH with cognitive function. The findings were synthesized narratively according to MRI techniques. Conventional MRI studies showed that greater WMH burden and CSVD severity were associated with poorer cognition, particularly processing speed and global cognition, with individual studies reporting β values of - 0.371 (p < 0.001) and - 0.363 (p < 0.001), with good predictive accuracy reported in one study (AUC = 0.800, 95% CI: 0.710-0.891). While Advanced MRI also showed associations with microstructural and microvascular damage in white matter and with reduced blood flow was associated with poorer cognitive performance, individual studies reported β values of 0.268-0.273, moderate-to-strong associations (r = 0.310-0.441), good diagnostic accuracy (AUC = 0.86-0.89), and significantly lower scores on cognitive assessments (Mini-Mental State Examination: p < 0.001; Montreal Cognitive Assessment: p < 0.05). These findings suggest that WMH burden and disrupted white matter microstructure are associated with cognitive impairment. Larger, standardized studies are needed to determine the clinical value of MRI markers as reliable biomarkers for early detection and monitoring of cognitive impairment in CSVD.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13649864/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149135033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Brainstem volume and arterial stiffness interact to support memory in midlife.","authors":"Yanrong Li, Tianyu Wang, Marie Caillaud, Cherry Youn, Isabelle Gallagher, Jessica Park, Hirofumi Tanaka, Andreana P Haley","doi":"10.1007/s11682-026-01206-x","DOIUrl":"https://doi.org/10.1007/s11682-026-01206-x","url":null,"abstract":"<p><p>Historically regarded as a hub for autonomic control, the brainstem is now increasingly recognized for its role in cognition. However, little information is available regarding its specific contribution to cognitive function, particularly in midlife, when modifiable cardiovascular risks begin to shape brain health trajectories. Because the brainstem houses key nuclei that regulate cardiovascular dynamics and cerebral perfusion, we focused on carotid-femoral pulse wave velocity (cfPWV), a well-established index of central arterial stiffness that directly reflects cerebral hemodynamic load. This study investigated whether brainstem volume is associated with cognitive performance in healthy midlife adults and whether this relation is moderated by cfPWV. A total of 198 adults aged 40-65 years underwent neuropsychological testing, fasting cardiovascular and metabolic assessments, and structural MRI. Linear regression models examined brainstem volume as a predictor of attention, executive function, and memory, adjusting for age, sex, education, and total intracranial volume. Additional models tested the interaction between cfPWV and brainstem volume, controlling for cardiometabolic covariates. Greater brainstem volume was significantly associated with better memory performance (p = .010, FDR q = 0.030), but not attention or executive function. Greater brainstem volume was also associated with lower systolic and pulse pressure. Notably, cfPWV moderated the relationship between brainstem volume and memory, such that the positive association was strongest among individuals with higher arterial stiffness (p = .001, FDR q = 0.002). These findings suggest that the brainstem may serve as a structural hub linking peripheral vascular health and cognitive function in midlife, particularly under conditions of elevated arterial stiffness.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13633365/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149103613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neural correlates of acute task-related stress in humans: a systematic review and coordinate-based ALE meta-analysis.","authors":"Xinyi Zhang, Sijun Liu, Yuyi Zhang, Yunxiao Guo","doi":"10.1007/s11682-026-01204-z","DOIUrl":"10.1007/s11682-026-01204-z","url":null,"abstract":"<p><p>Stress is a major risk factor for psychiatric disorders, yet the whole-brain neural signature of acute task-related stress remains incompletely defined, particularly for stress-related decreases in activation. We conducted a PROSPERO-registered systematic review and activation likelihood estimation (ALE) meta-analysis of English-language, peer-reviewed human task-based functional MRI (fMRI) studies published between January 2000 and November 2025. Eligible studies reported whole-brain stereotactic coordinates for stress versus matched non-stress control conditions. Stress > control and control > stress contrasts were analysed separately in GingerALE 3.0.2 using random-effects ALE with cluster-level family-wise error correction. Eighty-nine studies met inclusion criteria. The stress > control analysis included 85 experiments from 84 studies (2,810 participants; 1,350 foci), whereas the control > stress analysis included 25 experiments from 25 studies (922 participants; 263 foci). Stress > control showed convergent activation in a bilateral network including the anterior insula and inferior frontal operculum, dorsal anterior cingulate and supplementary motor area, thalamus, striatum, lateral frontal and parietal cortex, and left amygdala/parahippocampal gyrus. Control > stress yielded a single right parahippocampal-amygdala cluster. Thus, across heterogeneous acute stress-related challenges, the task-based fMRI literature was characterized more by reproducible activation increases across salience, arousal, and action-readiness systems than by consistent deactivation. These findings provide an updated trans-paradigm coordinate-based reference map for stress neurobiology, while acknowledging that the pooled convergence may partly reflect neural responses to salience or task demand that are shared across paradigms.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149103968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From voxels to models: a multi-center voxel-based morphometry study identifying robust biomarkers for deep learning classification of schizophrenia.","authors":"Mohammed Jajere Adamu, Li Qiang, Charles Okanda Nyatega","doi":"10.1007/s11682-026-01202-1","DOIUrl":"https://doi.org/10.1007/s11682-026-01202-1","url":null,"abstract":"<p><strong>Background: </strong>The application of deep learning in psychiatric disorders such as schizophrenia has lagged behind neuro-oncology, primarily due to a lack of identifiable neuroanatomical targets and concerns regarding generalizability across sites.</p><p><strong>Methods: </strong>To address this, we analyzed structural MRI data from a multi-site study of 401 subjects (194 schizophrenia, 207 healthy) across three independent datasets (COBRE, FBIRN, SLIM). We employed a two-stage conservative voxel-based morphometry (VBM) approach where neuroanatomical regions identified in a discovery dataset (COBRE) were rigorously validated in independent replication datasets (FBIRN, SLIM). The neuroanatomical features that survived this validation were then used to train a support vector machine (SVM) classifier with strictly nested cross-validation to prevent data leakage.</p><p><strong>Results: </strong>Fifteen robust structural biomarkers were identified, consistently showing gray matter reductions in frontal and temporal regions, white matter alterations, and ventricular enlargement across all datasets. The SVM classifier achieved an area under the curve (AUC) of 0.94 with 89.5% accuracy in initial analysis, with strictly nested cross-validation confirming robust performance (AUC = 0.93, accuracy = 88.7%). The model maintained high performance in leave-one-dataset-out validation (AUC = 0.90-0.93), demonstrating generalizability across unseen data from differing sites. A label-shuffling permutation test confirmed that the feature set reflects genuine neuroanatomical signal rather than statistical artifacts (p < 0.001).</p><p><strong>Conclusion: </strong>This study demonstrates a robust, multi-site validated set of structural biomarkers associated with schizophrenia, enabling highly accurate case-control classification. The identified feature set provides a candidate neurobiological benchmark to inform and validate future deep learning models for automated diagnostic support in psychiatry, while acknowledging the need for future multi-diagnostic validation to establish differential diagnostic utility.</p>","PeriodicalId":9192,"journal":{"name":"Brain Imaging and Behavior","volume":"20 5","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149103713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}