Objective:Polygenic risk scores (PRS) for posttraumatic stress disorder (PTSD) often account for a low amount of variance. Ancestry-related differences in PRS scale and variance limit cross-group comparisons. This methodological challenge further complicates gene-by-environment (GxE) analyses, given that socioenvironmental exposures are inequitably distributed across ethnoracial groups. We constructed an ancestry-calibrated polygenic risk score (AC-PRS) for PTSD in the largest longitudinal study of trauma survivors to date and investigated GxE interactions. Method:Recent trauma survivors (N=1,801) provided a blood specimen for genotyping. Six PTSD trajectories were previously identified from PTSD Checklist for DSM-5 (PCL-5) scores at 2-weeks, 8-weeks, 3-months, and 6-months post-trauma. Greenspace (normalized difference vegetation index [NDVI) and socioeconomic disadvantage (area deprivation index [ADI]) were derived from residential addresses. Logistic regressions examined interactions between newly developed AC-PRS and neighborhood factors on trajectories after adjusting for sociodemographic and trauma-related covariates. Secondary linear models considered GxE interactions on 6-month PCL-5 scores. Results:AC-PRS performed well across ethnoracial groups, explaining significant variability in PTSD trajectories (R2=.053). ADI moderated the association between AC-PRS and the likelihood of assignment in a high nonremitting trajectory of PTSD symptoms and severity of symptoms at 6-months (ps < .05). There were no NDVI x AC-PRS interactions in any models. Conclusions:AC-PRS captures genetic risk for PTSD in admixed trauma survivors, demonstrating good discrimination between nonremitting and resilient courses of PTSD. However, neighborhood disadvantage may modify utility of PRS for PTSD, warranting careful consideration when applying these scores across contexts.

Schizophrenia (SZ) is characterized by widespread neural dysconnectivity, with particularly pronounced alterations in the default mode network (DMN)—a set of brain regions involved in self-referential thought and mind-wandering. We investigated dynamic functional connectivity within the DMN in SZ by analyzing resting-state fMRI data from two independent cohorts (FBIRN and COBRE). Using a novel iterative feature-removal clustering approach focusing on DMN independent components, we identified distinct recurring connectivity states while iteratively removing dominant connectivity features to reveal subtler network patterns. This approach iteratively refines the set of features considered, ensuring a more balanced representation and facilitating the identification of significant interactions that would otherwise be overlooked. Cluster centroids for the four connectivity states were highly similar across both datasets, reflecting stable shared DMN patterns. State occupancy was compared between participants with SZ and healthy controls, and associations with clinical symptom severity were examined. The SZ group exhibited significant alterations in DMN connectivity dynamics, spending more time than controls in certain connectivity states characterized by atypical DMN coupling and less time in a state reflecting a normative DMN configuration. A comparison between the FBIRN and COBRE datasets reveals both similarities and differences in occupancy rate (OCR) states. Overall, the pattern of group effects converged while state-specific divergences likely reflected protocol/sample differences rather than conflicting biology. Crucially, the explainability pipeline yielded a highly similar feature-removal order across datasets, indicating that the same DMN edges drive the diagnostic signal in both cohorts. Several OCR effects also tracked symptom ratings, linking abnormal DMN dynamics to clinical expression. Our findings suggest that SZ is characterized by reproducible disturbances in the temporal organization of DMN connectivity. These dynamic DMN features may serve as potential biomarkers for SZ, offering diagnostic and clinical utility by capturing network dysfunction that static connectivity measures could overlook. ### Competing Interest Statement The authors have declared no competing interest. National Institute of Health, R01MH123610

This article discusses the potential applications of dry electroencephalography (EEG) and its advantages over traditional wet EEG.

Objective. Functional network connectivity (FNC) estimated from resting-state functional magnetic resonance imaging showed great information about the neural mechanism in different brain disorders. But previous research has mainly focused on standard statistical learning approaches to find FNC features separating patients from control. While machine learning models can improve classification accuracy, they often lack interpretability, making it difficult to understand how they arrive at their decisions. Approach. Explainable machine learning helps address this issue by identifying which features contribute most to the model's predictions. In this study, we introduce a novel framework leveraging SHapley Additive exPlanations (SHAPs) to identify crucial FNC features distinguishing between two distinct population classes. Main results. Initially, we validate our approach using synthetic data. Subsequently, applying our framework, we ascertain FNC biomarkers distinguishing between, controls and schizophrenia (SZ) patients with accuracy of 81.04% as well as middle aged adults and old aged adults with accuracy 71.38%, respectively, employing random forest, XGBoost, and CATBoost models. Significance. Our analysis underscores the pivotal role of the cognitive control network (CCN), subcortical network (SCN), and somatomotor network in discerning individuals with SZ from controls. In addition, our platform found CCN and SCN as the most important networks separating young adults from older.

Background:Posttraumatic stress disorder (PTSD) is a debilitating condition that disproportionately impacts females. Prior research indicates that males with PTSD exhibit hypoconnectivity of frontal brain regions measured with resting electroencephalography (EEG). In the current study, we examined functional connectivity among females with PTSD and trauma-exposed control females, as well as the impact of sex hormones. Methods:Participants included 61 females (mean age = 31.41 years, SD = 8.64) who endorsed criterion A trauma exposure. Resting-state EEG data were recorded for 5 minutes in the eyes-open position. Using a linear mixed-effects model, functional connectivity of the theta band (4-7 Hz) served as the response variable. Results:Compared with the control group, the PTSD group showed hyperconnectivity between visual brain regions and the rest of the cerebral cortex (false discovery rate-corrected p [p FDR] < .05). Additionally, participants with PTSD demonstrated enhanced connectivity between the default mode network and frontoparietal control network compared with control participants (p FDR < .05), as well as increased connectivity between the ventral attention network and the rest of the cerebral cortex (p FDR < .05). Estradiol was associated with higher connectivity, while progesterone was associated with lower connectivity, but these associations did not survive correction. Conclusions:The results are consistent with prior research indicating that PTSD is associated with altered connectivity in visual brain regions, which may reflect disrupted visual processing related to reexperiencing symptoms (e.g., intrusive memories). Our findings provide additional support for the relevance of the theta frequency range in PTSD given its role in fear learning and regulation processes.

Identifying robust neuroimaging biomarkers for Alzheimer's disease (AD) and mild cognitive impairment (MCI) is essential for early diagnosis and intervention. In this study, we introduce a novel, fully automated, guided dynamic functional connectivity (dFNC) framework for extracting multiple dynamic measures to distinguish MCI/AD from cognitively normal (CN) individuals. Resting-state fMRI data were used to extract subject-specific brain networks via spatially constrained independent component analysis (scICA), using a multi-objective optimization framework to ensure alignment with known functional networks while preserving individual variability. Using these components, dFNC was computed through a sliding-window approach. ICA was then applied to the concatenated dFNC matrices from the UK Biobank (UKBB) dataset to identify five canonical brain states, each representing a replicable, independent pattern of connectivity. These states served as biologically informed priors in a state-constrained ICA (St-cICA), which was applied to each subject in the combined OASIS-3 and ADNI datasets to guide individual-level decomposition and ensure interpretable connectivity states guided by state priors derived from the normative UKBB sample. St-cICA extracted subject-specific dFNC features and associated weighted timecourses. To characterize dFNC patterns, we computed metrics from the most strongly expressed (primary) state and introduced estimation of the second-most expressed (secondary) state at each time point, including dwell time, occupancy rate, and transition probabilities. Group comparisons using two-sample t-tests revealed widespread and significant alterations in AD/MCI compared to CN individuals. AD/MCI participants exhibited higher dwell times and increased self-transitions, indicating reduced neural flexibility and a tendency to remain in specific connectivity states. In contrast, CN individuals showed more diverse and recurrent transitions, reflecting greater adaptability. Secondary transitions revealed widespread selective switching in CN, whereas AD/MCI showed reduced cross-state engagement. A classification model trained on 6,960 dynamic features achieved strong performance in distinguishing AD/MCI from CN (mean AUC ≈ 0.85). These findings highlight the potential of guided dFNC as a biomarker framework for early-stage AD detection using resting-state fMRI.

This study examines the association between brain dynamic functional network connectivity (dFNC) and current/future posttraumatic stress (PTS) symptom severity, and the impact of sex on this relationship. By analyzing 275 participants' dFNC data obtained ~2 weeks after trauma exposure, we noted that brain dynamics of an inter-network brain state link negatively with current (r=-0.179, p corrected = 0.021) and future (r=-0.166, p corrected = 0.029) PTS symptom severity. Also, dynamics of an intra-network brain state correlated with future symptom intensity (r = 0.192, p corrected = 0.021). We additionally observed that the association between the network dynamics of the inter-network brain state with symptom severity is more pronounced in females (r=-0.244, p corrected = 0.014). Our findings highlight a potential link between brain network dynamics in the aftermath of trauma with current and future PTSD outcomes, with a stronger protective effect of inter-network brain states against symptom severity in females, underscoring the importance of sex differences.

Background:Schizophrenia is characterized by deficits in attention and working memory. In recent years, the brain age gap (BAG), defined as the difference between neuroimaging-predicted and chronological age, has emerged as a biomarker of brain dysfunction. Prior studies primarily use structural MRI or static functional network connectivity (sFNC), while the potential of dynamic functional network connectivity (dFNC) to quantify BAG in relationship with cognition remains underexplored. Methods:Leveraging one of the largest resting-state fMRI datasets to date (N=22,569; UK Biobank, HCP, HCP-Aging), we developed robust brain age prediction models incorporating both wide-brain and sub-network variables derived from sFNC and dFNC. These models were validated in an independent clinical sample (FBIRN; N=153) including individuals with schizophrenia and healthy controls. Associations between BAGs and cognitive measures (attention vigilance, working memory) were evaluated using general linear models, controlling for key demographic and clinical covariates. Results:Both sFNC and dFNC models demonstrated robust prediction accuracy in healthy individuals (sFNC: r=0.8755; dFNC: r=0.8675). Wide-brain BAGs (wBAGs) showed strong negative associations with attention vigilance (sFNC: r=-0.2923, FDR p=0.0013; dFNC: r=-0.2715, FDR p=0.0016) and working memory (sFNC: r=-0.2237, FDR p=0.0088; dFNC: r=-0.2508, FDR p=0.0064). Sub-network BAGs (subBAGs) within subcortical, sensorimotor, cognitive control, and default mode networks robustly predicted cognitive deficits, with dFNC-derived subBAGs showing the strongest effects (FDR p<0.01). Conclusions:Our findings establish dFNC-based BAGs as a sensitive and clinically relevant biomarker of cognitive impairment in schizophrenia, outperforming sFNC and highlighting the translational potential of dynamic connectivity for precision diagnosis and treatment.

Posttraumatic Stress Disorder (PTSD) is a heterogeneous mental health disorder that can develop following a traumatic experience. Understanding its neurobiological basis is crucial to advance early diagnosis and treatment. Electroencephalography (EEG) can be used to explore the neurobiological basis of PTSD. However, only limited research has explored mobile EEG, which is important for scalability. This proof-of-concept study delves into mobile EEG-derived biomarkers for posttraumatic stress (PTS) symptom severity and their potential implications. Participants with partial PTSD, defined as meeting for at least three out of four symptom clusters, including hyperarousal symptoms, were enrolled in the study. Over four weeks, we measured PTS symptom severity using the PTSD checklist for DSM-5 (PCL-5) at multiple timepoints, and we recorded multiple EEG sessions from 21 individuals using a mobile EEG device. In total, we captured 38 EEG sessions, each comprising two recordings (“Recording A” and “Recording B”) that lasted approximately 180 s, to evaluate reproducibility. Next, we extracted Shannon entropy, as a measure of the brain flexibility and complexity of the signal and spectral power for the fronto-temporal regions of interest, including electrodes at AF3, AF4, T7, and T8 for each EEG recording session. We calculated the partial correlation between the EEG variables and PCL- 5 measured closest to the EEG session, using age, sex, and the grouping variable ‘batch’ as covariates. We observed a significant negative correlation between Shannon entropy in fronto- temporal regions and PCL-5 scores. Specifically, this association was evident in the AF3 (r = -0.456, FDR-corrected p = 0.01), AF4 (r = -0.362, FDR-corrected p = 0.04), and T7 (r = -0.472, FDR-corrected p = 0.01) regions. Additionally, we found a significant negative association between the alpha power estimated from AF4 and PCL-5 (r = -0.429, FDR-corrected p = 0.04). Our findings suggest that EEG markers acquired using a mobile EEG device are associated with PTS symptom severity, offering valuable insights into the neurobiological mechanisms underlying PTSD and highlighting the potential benefits of this innovative technology in assessing and monitoring PTSD.

The molecular pathology of stress-related disorders remains elusive. Our brain multiregion, multiomic study of posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) included the central nucleus of the amygdala, hippocampal dentate gyrus, and medial prefrontal cortex (mPFC). Genes and exons within the mPFC carried most disease signals replicated across two independent cohorts. Pathways pointed to immune function, neuronal and synaptic regulation, and stress hormones. Multiomic factor and gene network analyses provided the underlying genomic structure. Single nucleus RNA sequencing in dorsolateral PFC revealed dysregulated (stress-related) signals in neuronal and non-neuronal cell types. Analyses of brain-blood intersections in >50,000 UK Biobank participants were conducted along with fine-mapping of the results of PTSD and MDD genome-wide association studies to distinguish risk from disease processes. Our data suggest shared and distinct molecular pathology in both disorders and propose potential therapeutic targets and biomarkers.

Functional network connectivity has previously been shown to distinguish patient groups from healthy controls (HC). However, the overlap across schizophrenia (SZ), bipolar (BP), and schizoaffective disorder (SAD) is not clear yet. This study focuses on finding overlap across these three psychotic disorders using dynamic FNC (dFNC) and compares it with static FNC (sFNC). We used resting-state fMRI, demographics, and clinical information from the Bipolar– Schizophrenia Network on Intermediate Phenotypes cohort. The data includes three groups of patients with schizophrenia (SZP, N=102), bipolar (BPP, N=102), and schizoaffective (SADP, N=102), their relatives SZR (N=102), BPR (N=102), SADR (N=102), and HC (N=118) groups. After estimating each individual’s dFNC, we put them into three identical states. We estimated five different features, including occupancy rate (OCR), number of transitions, the total number of transitions, and the total distance traveled. Finally, the extracted features are tested statistically across patients and HC groups. In addition, we explored the link between the clinical scores and the extracted features. We found that the OCR difference between SZP and SZR in state2, between BPP and HC in state1, and between SADP and HC in state2. Also, state2 OCR separates SZP from BPP, state 3 OCR separates BPP from SZP and SADP. Moreover, the OCR and traveled distance feature extracted from SZ and BP could significantly predict PANSS Total and PANSS General scores. Finally, combined distance features of all disorders showed a significant relationship to PANSS Total and PANSS General scores.

A common analysis approach for resting state functional magnetic resonance imaging (rs-fMRI) dynamic functional network connectivity (dFNC) data involves clustering windowed correlation time-series and assigning time windows to clusters (i.e., states) that can be quantified to summarize aspects of the dFNC dynamics. However, those methods can be dominated by a select few features and obscure key dynamics related to less dominant features. This study presents an iterative feature learning approach to identify a maximally significant and minimally complex subset of dFNC features within the default mode network (DMN) in schizophrenia (SZ). Utilizing dFNC data from individuals with SZ and healthy controls (HC), our approach uncovers a subset of features that has a greater number of dFNC states with disorder-related dynamics than is found when all features are present in the clustering. We find that anterior cingulate cortex/posterior cingulate cortex (ACC/PCC) interactions are consistently related to SZ across the most significant iterations of the feature learning analysis and that individuals with SZ tend to spend more time in states with greater intra-ACC anticorrelation and almost no time in a state of high intra-ACC correlation that HCs periodically enter. Our findings highlight the need for nuanced analyses to reveal disorder-related dynamics and advance our understanding of neuropsychiatric disorders.

Posttraumatic Stress Disorder (PTSD) is a heterogeneous mental health disorder that occurs following traumatic experience. Understanding its neurobiological basis is crucial to advance early diagnosis and treatment. Electroencephalography (EEG) can be used to explore the neurobiological basis of PTSD. However, only limited research has explored mobile EEG, which is important for scalability. This proof-of-concept study delves into mobile EEG-derived biomarkers for PTSD and their potential implications. Over four weeks, we measured PTSD symptoms using the PTSD checklist for DSM-5 (PCL-5) at multiple timepoints, and we recorded multiple EEG sessions from 21 individuals using a mobile EEG device. In total, we captured 38 EEG sessions, each comprising two recordings that lasted approximately 180 seconds, to evaluate reproducibility. Next, we extracted Shannon entropy, as a measure of the randomness or unpredictability of the signal and spectral power for the fronto-temporal regions of interest, including electrodes at AF3, AF4, T7, and T8 for each EEG recording session. We calculated the partial correlation between the EEG variables and PCL-5 measured closest to the EEG session, using age, sex, and the grouping variable 'batch' as covariates. We observed a significant negative correlation between Shannon entropy in fronto-temporal regions and PCL-5 scores. Specifically, this association was evident in the AF3 (r = -0.456, FDR-corrected p = 0.01), AF4 (r = -0.362, FDR-corrected p = 0.04), and T7 (r = -0.472, FDR-corrected p = 0.01) regions. Additionally, we found a significant negative association between the alpha power estimated from AF4 and PCL-5 (r=-0.429, FDR-corrected p=0.04). Our findings suggest that EEG data acquired using a mobile EEG device is associated with PTSD symptom severity, offering valuable insights into the neurobiological mechanisms underlying PTSD.

Objective.To treat neurological and psychiatric diseases with deep brain stimulation (DBS), a trained clinician must select parameters for each patient by monitoring their symptoms and side-effects in a months-long trial-and-error process, delaying optimal clinical outcomes. Bayesian optimization has been proposed as an efficient method to quickly and automatically search for optimal parameters. However, conventional Bayesian optimization does not account for patient safety and could trigger unwanted or dangerous side-effects.Approach.In this study we develop SAFE-OPT, a Bayesian optimization algorithm designed to learn subject-specific safety constraints to avoid potentially harmful stimulation settings during optimization. We prototype and validate SAFE-OPT using a rodent multielectrode stimulation paradigm which causes subject-specific performance deficits in a spatial memory task. We first use data from an initial cohort of subjects to build a simulation where we design the best SAFE-OPT configuration for safe and accurate searchingin silico. Main results.We then deploy both SAFE-OPT and conventional Bayesian optimization without safety constraints in new subjectsin vivo, showing that SAFE-OPT can find an optimally high stimulation amplitude that does not harm task performance with comparable sample efficiency to Bayesian optimization and without selecting amplitude values that exceed the subject's safety threshold.Significance.The incorporation of safety constraints will provide a key step for adopting Bayesian optimization in real-world applications of DBS.

Background: Posttraumatic stress disorder (PTSD) is a debilitating condition that disproportionately impacts individuals who are female. Prior research indicates that males with PTSD exhibit hypoconnectivity of frontal brain regions measured with resting electroencephalography (EEG). The present study examined functional connectivity among females with PTSD and trauma-exposed controls, as well as the impact of sex hormones. Methods: Participants included 61 females (Mage = 31.41, SD = 8.64) who endorsed Criterion A trauma exposure. Resting state EEG data were recorded for five minutes in the eyes open position. Using a Linear Mixed Effects model, paired region-of-interest power envelope connectivity of the theta band (4-7 Hz) served as the response variables. Results: Compared to controls, the PTSD group displayed hyperconnectivity between visual brain regions and the rest of the cerebral cortex (pFDR < 0.05). Additionally, participants with PTSD demonstrated enhanced connectivity between the default mode network and frontoparietal control network compared to controls (pFDR < 0.05), as well as increased connectivity between the ventral attention network and the rest of the cerebral cortex (pFDR < 0.05). Estradiol was associated with higher connectivity, while progesterone was associated with lower connectivity, but these did not survive correction. Conclusions: Results are consistent with prior research indicating that PTSD is associated with altered connectivity in visual brain regions, which may reflect disrupted visual processing related to reexperiencing symptoms (e.g., intrusive memories). Our findings provide additional support for the relevance of the theta frequency range in PTSD given its role in fear learning and regulation processes.

Introduction: Alzheimer’s disease (AD), the most prevalent age-related dementia, leads to significant cognitive decline. While genetic risk factors and neuroimaging biomarkers have been extensively studied, establishing a neuroimaging-based metric to assess AD risk has received less attention. This study introduces the Brain-wide Risk Score (BRS), a novel approach using multimodal neuroimaging data to assess the risk of mild cognitive impairment (MCI), a precursor to AD. Methods: Participants from the OASIS-3 cohort (N = 1,389) were categorized into control (CN) and MCI groups. Structural MRI (sMRI) data provided gray matter (GM) segmentation maps, while resting-state functional MRI (fMRI) data yielded functional network connectivity (FNC) matrices via spatially constrained independent component analysis. Similar imaging features were computed from the UK Biobank (N = 37,780). The BRS was calculated by comparing each participant’s neuroimaging features to the difference between average features of CN and MCI groups. Both GM and FNC features were used. The BRS effectively differentiated CN from MCI individuals within OASIS-3 and in an independent dataset from the ADNI cohort (N = 729), demonstrating its ability to identify MCI risk. Results: Unimodal analysis revealed that sMRI provided greater differentiation than fMRI, consistent with prior research. Using the multimodal BRS, we identified two distinct groups: one with high MCI risk (negative GM and FNC BRS) and another with low MCI risk (positive GM and FNC BRS). Additionally, 46 UK Biobank participants diagnosed with AD showed FNC and GM patterns similar to the high-risk groups. Conclusion: Validation using the ADNI dataset confirmed our results, highlighting the potential of FNC and sMRI-based BRS in early Alzheimer’s detection.

Brain stimulation holds promise for treating brain disorders, but personalizing therapy remains challenging. Effective treatment requires establishing a functional link between stimulation parameters and brain response, yet traditional methods like random sampling (RS) are inefficient and costly. To overcome this, we developed an active learning (AL) framework that identifies optimal relationships between stimulation parameters and brain response with fewer experiments. We validated this framework through three experiments: (1) in silico modeling with synthetic data from a Parkinson’s disease model, (2) in silico modeling with real data from a non-human primate, and (3) in vivo modeling with a real-time rat optogenetic stimulation experiment. In each experiment, we compared AL models to RS models, using various query strategies and stimulation parameters (amplitude, frequency, pulse width). AL models consistently outperformed RS models, achieving lower error on unseen test data in silico (p<0.0056, N=1000) and in vivo (p=0.0036, N=20). This approach represents a significant advancement in brain stimulation, potentially improving both research and clinical applications by making them more efficient and effective. Our findings suggest that AL can substantially reduce the cost and time required for developing personalized brain stimulation therapies, paving the way for more effective and accessible treatments for brain disorders.

Alzheimer's disease (AD) is the most common form of age-related dementia, leading to a decline in memory, reasoning, and social skills. While numerous studies have investigated the genetic risk factors associated with AD, less attention has been given to identifying a brain imaging-based measure of AD risk. This study introduces a novel approach to assess mild cognitive impairment MCI, as a stage before AD, risk using neuroimaging data, referred to as a brain-wide risk score (BRS), which incorporates multimodal brain imaging. To begin, we first categorized participants from the Open Access Series of Imaging Studies (OASIS)-3 cohort into two groups: controls (CN) and individuals with MCI. Next, we computed structure and functional imaging features from all the OASIS data as well as all the UK Biobank data. For resting functional magnetic resonance imaging (fMRI) data, we computed functional network connectivity (FNC) matrices using fully automated spatially constrained independent component analysis. For structural MRI data we computed gray matter (GM) segmentation maps. We then evaluated the similarity between each participant's neuroimaging features from the UK Biobank and the difference in the average of those features between CN individuals and those with MCI, which we refer to as the brain-wide risk score (BRS). Both GM and FNC features were utilized in determining the BRS. We first evaluated the differences in the distribution of the BRS for CN vs MCI within the OASIS-3 (using OASIS-3 as the reference group). Next, we evaluated the BRS in the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort (using OASIS-3 as the reference group), showing that the BRS can differentiate MCI from CN in an independent data set. Subsequently, using the sMRI BRS, we identified 10 distinct subgroups and similarly, we identified another set of 10 subgroups using the FNC BRS. For sMRI and FNC we observed results that mutually validate each other, with certain aspects being complementary. For the unimodal analysis, sMRI provides greater differentiation between MCI and CN individuals than the fMRI data, consistent with prior work. Additionally, by utilizing a multimodal BRS approach, which combines both GM and FNC assessments, we identified two groups of subjects using the multimodal BRS scores. One group exhibits high MCI risk with both negative GM and FNC BRS, while the other shows low MCI risk with both positive GM and FNC BRS. Moreover, in the UKBB we have 46 participants diagnosed with AD showed FNC and GM patterns similar to those in high-risk groups, defined in both unimodal and multimodal BRS. Finally, to ensure the reproducibility of our findings, we conducted a validation analysis using the ADNI as an additional reference dataset and repeated the above analysis. The results were consistently replicated across different reference groups, highlighting the potential of FNC and sMRI-based BRS in early Alzheimer's detection.