Frontiers in Neuroimaging

Independent component analysis (ICA) is widely used in resting-state fMRI to identify large-scale functional networks; however, existing approaches provide limited means of quantifying how network representations are distributed across independent components. We introduce an entropy-based network integration framework that characterizes the organizational architecture of canonical resting-state networks by quantifying the distribution of ICA-derived functional contributions within Yeo atlas networks. Spatial overlap between independent components and network templates is normalized to generate a probability distribution, from which Shannon entropy and a normalized integration index are derived. The resulting metric provides a continuous measure of network representational integration, ranging from specialized configurations dominated by a small number of components to distributed configurations involving multiple functional modes. The framework was evaluated and validated using resting-state fMRI data from healthy controls, Parkinsons disease patients with normal cognition, and Parkinsons disease patients with mild cognitive impairment. Global entropy and integration measures were complemented by network-specific analyses, dominance profiling, principal component analysis (PCA), and multivariate centroid-distance assessments. The proposed framework revealed selective alterations in Ventral Attention and Limbic network organization associated with cognitive-status differences, while preserving overall within-group heterogeneity. Group-wise PCA independently further identified these networks as major contributors to altered network organization, and centroid-distance analyses demonstrated that observed differences reflected coherent shifts in network architecture rather than increased variability. By quantifying the distribution of network representations across ICA-derived functional modes, this framework provides a simple, interpretable, and generalizable measure of large-scale brain organization, offering a complementary approach for studying network reorganization in health and disease.

Identifying informative voxels is a critical, yet challenging step in functional magnetic resonance imaging (fMRI), particularly for multivariate analyses involving multiple related conditions. Existing approaches often rely on predefined regions of interest (ROIs) or activation-based criteria, which may be insufficient for capturing fine-grained representational differences. This challenge becomes particularly relevant in experimental settings and interventions such as neurofeedback training, where voxels are not only measured as neural responses but also used as targets for intervention based on their previously observed activity patterns. In this study, we propose a subject-level searchlight optimization framework that integrates voxel-wise general linear model (GLM)-based univariate analysis with representational similarity analysis (RSA)-based multivariate refinement to identify voxels that are both task-relevant and condition-sensitive. To enhance practical applicability, the framework further incorporates a data-driven hyperparameter tuning step based on Bayesian optimization, enabling efficient identification of high-performing configurations from small pilot datasets, with consistent performance when applied to larger samples. The proposed framework was evaluated using an emotion imagery fMRI dataset with four affective conditions. Results demonstrate that the multivariate refinement improves alignment between empirical and target representational structures compared with univariate selection alone. Compared with a classifier-based voxel selection approach, the RSA-based approach better preserves the representational geometry of emotional states while maintaining discriminative capacity. These findings highlight the effectiveness, efficiency, and robustness of the proposed RSA framework, providing a practical solution for identifying condition-sensitive voxels and supporting more precise multivariate investigation of affective brain states in multi-condition fMRI studies.

While Blood Oxygenation Level-Dependent (BOLD) fMRI remains the gold standard for mapping functional brain networks with MRI, its vascular origins inherently conflate haemodynamic effects with neural activity, limiting its sensitivity in white matter (WM) or its interpretation in neurovascular diseases. Apparent Diffusion Coefficient (ADC) fMRI offers an alternative, diffusion-based contrast that is theoretically more sensitive to neuromorphological coupling and therefore more specific to neuronal activation, though investigated primarily during task-based conditions. This study aimed to comprehensively evaluate the efficacy of isotropic ADC-fMRI in detecting established resting-state networks (RSNs) and to extend this methodology to the investigation of grey-to-white matter (GM-WM) functional connectivity. Our analyses revealed a gradient of ADC detectability shaped by the degree of static functional cohesion and structural tethering of each network. The visual and somatomotor networks, being both highly segregated and strongly anchored to underlying structural pathways, yielded the most robust detection. The default mode network (DMN) and dorsal attention network (DAN) reached group-level significance but with lower effect sizes, and their detection proved fragile across analytical approaches. The frontoparietal network (FPN) and salience network (SAN), whose functional identity is defined by dynamic cross-network reconfiguration, did not reach significance. This gradient partially mirrors the established hierarchy of network segregation observed in BOLD, while further suggesting that ADC sensitivity depends on the structural grounding of each network. Furthermore, ADC demonstrated superior sensitivity to GM-WM functional coupling compared to BOLD. GM-WM functional connectivity profiles derived from ADC were significantly more aligned with underlying structural WM architecture across subjects. Taken together, these findings position isotropic ADC-fMRI as a viable complementary modality to BOLD, offering a more direct window into the neural and structural foundations of brain connectivity.

Individual brains are unique in structure and function. Functional differences are captured by neural fingerprints, which reflect individual differences in behavior and cognition as well as group-level changes related to neurodegenerative diseases. Most research efforts so far have focused on fingerprints com-prising full functional connectomes. However, the high dimensionality of the connectomes can increase computational load and impede performance of machine learning methods in potential applications. A low-dimensional alternative that retains individual features of the full connectomes would thus be beneficial. The present study employed latent-noise Bayesian Reduced Rank Regression (lnBRRR) to learn low-dimensional latent spaces that capture individual features in functional connectivity and power spectral density data derived from MEG recordings. LnBRRR performance was assessed with low training set sizes (N=20-44), and against principal component analysis and linear discriminant analysis. Model performance was also assessed with task data, and the solutions were compared across task conditions with cosine similarity to establish whether individual features are altered by different cognitive processes. LnBRRR captured generalizable individual patterns already at N=20 but N=30-35 was needed to reach optimal test accuracies and to prevent potential overfitting. The model also achieved comparable performance to the alternative models. Latent fingerprints derived from task data attained comparable performance to resting-state latent fingerprints, and lnBRRR solutions were shown to generalize across conditions. Additionally, the model solutions for power spectral density data were discovered to be notably similar, yet differently rotated, over task conditions, suggesting that similar patterns of individual features were captured by the model regardless of the task condition. Altogether, the present results highlight lnBRRR as a potential tool for neuroimaging data analysis and demonstrate that individual differences in power spectral density are largely intrinsic and unaffected by varying cognitive processes.

Emerging evidence suggests that saliva cell-free mitochondrial DNA (cf-mtDNA) increases in response to psychosocial and physical stress. Here, we quantified saliva cf-mtDNA changes in response to acute physical and cognitive stressors as well as identifying potential predictors of these responses, while also exploring the potential modulatory effects of transcranial infrared laser stimulation (TILS). In a crossover design, a total of 47 participants (53% female, ages 18-30) underwent up to three experimental sessions, including an exercise stress task and two cognitive stress tasks. Repeated saliva samples were collected for cf-mtDNA and cell-free nuclear DNA (cf-nDNA) quantification, alongside continuous measurement of heart rate, oxygen consumption, and blood pressure. Our results show that average cf-mtDNA levels increased by 90% after baseline during exercise experiments, and in cognitive stress experiments peaked 160% above average baseline levels during the stress task. Inter-individual differences in response trajectories were associated with differences in factors such as fitness, sleep quality, and stress perception. Notably, participants with higher cf-mtDNA elevations during the exercise experiment reported fewer recent stressful incidents, drank alcohol less frequently, had higher maximum VO2 during exercise, and had lower BMI. More dynamic responses to cognitive stress were observed in participants with poorer sleep quality and greater blood pressure reactivity. These findings provide a foundation for larger studies by highlighting the dynamic behavior of saliva cf-mtDNA following physical and cognitive stressors, and by suggesting potential drivers of individual differences in saliva cf-mtDNA stress reactivity.

PURPOSE: The UK7T travelling head dataset was used to characterise the reproducibility of 7T measurements of the susceptibility of the hippocampal subfields, focusing on the Cornu Ammonis (CA1, CA2 and CA3), dentate gyrus (DG), subiculum (SUB), tail of the hippocampus (TAIL) and entorhinal cortex (ERC). METHODS: Susceptibility maps were created from whole-brain 3D single-echo GRE data (TE=20 ms; 0.7 mm isotropic resolution) using Multi-Scale Dipole Inversion. Automatic Segmentation of Hippocampal Subfields (ASHS) was applied to high resolution T1- and T2-weighted images for segmentation. The mean magnetic susceptibility and volume of hippocampal subfields was evaluated in 50 data sets, comprising 5 repeat acquisitions on 10 healthy participants (age 32 + or -6 years; 3 female). RESULTS: Averaging over subjects, susceptibility values spanned an 18ppb range over the hippocampus (ranging from -13.3ppb in DG to 4.7ppb in ERC). Susceptibility values in the larger hippocampal subfields showed a consistent pattern of variation across subjects, being generally more positive in ERC and SUB than in CA1 and more positive in CA1 than in DG and TAIL. The standard deviation of subfield susceptibilities over subjects ranged from 8.2ppb in the TAIL to 1.7ppb in CA1, and the average standard deviation across repeated measurements, which ranges from 1.7 to 4 ppb, was less than half of the inter-participant standard deviation in all subfields. Susceptibility values in the smaller subfields (CA2 and CA3) were more variable, but ICC(2,k) values for all subfields were >0.82. CONCLUSION: The reported data characterises the variation and reproducibility of hippocampal subfield susceptibility measurements at 7T.

Background: Transcranial magnetic stimulation (TMS) targeting the left dorsolateral prefrontal cortex (dlPFC) is an established treatment option in major depressive disorder. One of the most common approaches for targeting the dlPFC is the Beam F3 method, which determines the stimulation site (F3Beam) as a function of external cranial measurements. Precise knowledge of the individual stimulation site is essential for imaging-based analyses of TMS effects. However, due to the method's reliance on individual anatomy, retrospective identification of F3Beam targets across cohorts is challenging, limiting the analysis of existing datasets. We developed a scalable method to reconstruct subject-specific F3Beam target locations for e-field simulations based on structural imaging. Methods: High-resolution three-dimensional (3D) T1-weighted MRI was used to generate individual scalp meshes via the ''Simulation of Non-Invasive Brain Stimulation'' (SimNIBS) software. Subject-specific anatomical distances and coordinates of interest were measured geodesically using a Python-based script to reconstruct the individual F3Beam targets. Validation included a retrospective comparison between digital geodesic measurements and manual cranial measurements in 20 patients and a prospective comparison with MR-visible scalp markers in 2 healthy controls. To assess the impact of our targeting algorithm on e-field simulations, volumetric e-field maps based on three potential targets (F3Beam, F3MNI, F3Geo) were generated in SimNIBS and compared using voxel-wise statistics in SPM12. Results: Retrospective analysis revealed a systematic bias towards higher in vivo measurements compared to digital geodesic measurements, though deviations in the final distances determining F3Beam (xBeam and yBeam) were minimal ({Delta}xBeam: 0.11 {+/-} 0.08 cm; {Delta}yBeam: 0.14 {+/-} 0.21 cm). Prospective validation demonstrated that F3Beam coordinates better matched in vivo coil positions than group-template-derived targets (F3MNI). Group-level analysis showed method-dependent clustering of coil positions with corresponding voxel-wise e-field differences. Conclusions: Individualized geodesic measurements may enable accurate, scalable and retrospective identification of Beam F3 targets and coil orientations. This approach may yield more accurate e-field simulations than group-template based targeting and provides a practical method for retrospective analysis of existing TMS treatment cohorts. This could be leveraged to identify response predictors or imaging-based biomarkers of treatment response.

Moderate hypoxia is increasingly recognized as a physiological driver of neuroprotection and neuroregeneration. In this first randomised, double-blind, controlled, four-arm trial, we demonstrate the cognitive and neuroplastic effects of cognitive training under moderate inspiratory hypoxia in humans. Healthy volunteers underwent three weeks of either cognitive or sham training under normobaric hypoxia (12% O2) or normoxia (20% O2) for 3.5 hours daily, six days per week. Participants were assessed at baseline, treatment completion, and one-month follow-up. The primary outcome was change in a broad cognitive composite score. Additional cognitive, blood-based, and neuroimaging outcomes were assessed, including measurement of the presynaptic protein SV2A with [11C]UCB-J positron emission tomography (PET) and neural activity through functional magnetic resonance imaging (fMRI). In total, 126 participants were randomised to hypoxia-cognitive training (H-CT: n=36), hypoxia-sham training (H-ST: n=30), normoxia- cognitive training (N-CT: n=30), or normoxia-sham training (N-ST: n=30). Intention-to-treat analyses showed no effect of H-CT relative to N-ST in the primary outcome at treatment completion (primary endpoint; treatment effect=0.11, 95% CI=[-0.06;0.28], p=0.19), but improvements emerged at follow-up (treatment effect=0.17, 95% CI=[0.01;0.34], p=0.04). N-CT induced transient improvement in the primary outcome at treatment completion (treatment effect=0.20, 95% CI=[0.02;0.38], p=0.03), which rendered non-significant at follow-up. Finally, H-ST showed no significant cognitive change relative to N-ST. Moderate hypoxia was safe and well-tolerated. Cognitive benefits were accompanied by decreased hippocampal presynaptic density measured with [11C]UCB-J PET. In conclusion, three weeks of H-CT can enhance cognition with associated effects on neuroplasticity, although with a delayed onset of effects on cognition.

Schizophrenia is characterized by persistent executive dysfunction and atypical engagement of prefrontal circuits underlying attentional control. However, the neural dynamics of executive processing during ecologically relevant tasks remain underexplored. This study examined frontal theta/beta oscillatory patterns and trial level EEG responsiveness as indices of adaptive cognitive control in schizophrenia compared to healthy controls. Thirty adults with schizophrenia (M age = 39.9, SD = 9.10 years) and a matched healthy control group (N = 30; M age = 32.25, SD = 6.50 years) underwent quantitative EEG during an eyes-open resting state and while performing two executive tasks: an augmented reality visuomotor challenge (LCAR) and a mobile guided daily routine task (B2B). Frontal theta/beta ratios (TBR) at Fz and Cz indexed attentional engagement. Trial level responsiveness was assessed via discrete Stimulus Response Events (SREs). Results: Both LCAR and B2B elicited significant TBR increases relative to eyes open rest at midline frontal sites (p < .001), reflecting elevated executive demand. Compared to healthy controls, participants with schizophrenia exhibited higher baseline TBR and reduced modulation across task segments. In contrast, controls showed stronger SRE linked variability and greater memory gains, indicating more efficient task locked cognitive adaptation. Age related effects were also observed, with participants under 40 years showing higher resting TBR at Fp1 (p =.01). Conclusion: Findings advance understanding of prefrontal theta/beta modulation as a neurophysiological marker of adaptive executive control during complex, ecologically valid tasks. By integrating real-world paradigms with trial level EEG analyses, this study contributes to models of dynamic information processing and cognitive resource allocation in schizophrenia. Keywords: schizophrenia; healthy controls; theta/beta ratio; QEEG; executive function; augmented reality; neural biomarkers

Major depressive disorder (MDD) is a heterogeneous mental disorder characterised by rumination. Mindfulness-based cognitive therapy (MBCT) is an evidence-based treatment developed to target rumination and recurrence risk. Ongoing studies have begun to identify neural changes associated with treatment effects. However, the low-dimensional organisation underlying whole-brain dynamics remains largely unexplored and may provide a more complete characterisation of the neural processes through which MBCT exerts its therapeutic effects in MDD. Here, we investigated functional magnetic resonance imaging (fMRI) of a randomised controlled trial of MBCT with treatment as usual (TAU), or TAU alone, in a group of MDD patients (N=80). We applied a novel framework, complex harmonics decomposition (CHARM), to uncover low-dimensional manifolds in the spacetime domain, capturing local as well as non-local interactions made possible by brain criticality and amplified by the anatomical long-range connectivity. We successfully identified distinct distributed spatiotemporal manifolds across brain states and outperformed traditional dimensionality reduction techniques. During rumination after MBCT we found consistent recruitment of regions involved in bodily and interoceptive processing integrated within the whole-brain across manifolds, changes in latent configurations associated with clinical and behavioural improvements, and greater flexibility within the reduced space. Integration of bodily and interoceptive processing regions within distributed whole-brain manifolds and greater brain flexibility may be associated with reduced 'stickiness' of ruminative thinking patterns following mindfulness training in depression. Our findings highlight the promise of low-dimensional manifolds and long-range interactions arising from critical brain dynamics in understanding how mindfulness targets depressive ruminative processing.

Purpose: {beta}-hydroxybutyrate (BHB), a ketone body and alternative cerebral energy substrate, can be measured in vivo using J-difference edited proton magnetic resonance spectroscopy (1H-MRS). Oral ketone supplementation with substrates such as the ketone monoester (R)-3-hydroxybutyl-(R)-3-hydroxybutyrate (KME) and 1,3-butanediol (BD) have gained attention as a mechanism to elevate circulating BHB and induce ketosis without dietary restrictions. Elevated brain ketone availability is of growing therapeutic interest as a strategy to support neuronal energetics in conditions such as epilepsy, neurodegenerative disease, and alcohol use disorder (AUD). However, both pathways introduce BD into the bloodstream, which crosses the blood-brain barrier. Critically, BD exhibits a spectral signature that closely resembles the prominent BHB peak in JDE-MR spectroscopic imaging (MRSI), identified in a pilot AUD study. Methods: Two separate JDE-MRSI acquisitions tailored for BHB and BD editing were implemented, exploiting frequency separation between the BHB (4.14ppm) and BD (3.95ppm) coupling partners of the observed 1.2ppm resonance to independently quantify each metabolite. Results: Brain BD concentrations (0.25-0.58mM) were comparable to or exceeded corresponding BHB concentrations (0.20-0.27mM) in all volunteers after consumption of a single dose of the KME, indicating that BD constitutes a major fraction of the signal conventionally attributed to BHB. Combined BHB+BD concentrations (~0.45-0.85mM) were consistent with brain BHB values reported in prior studies employing similar doses of the KME, indicating that those measurements likely reflect a combined BHB+BD signal. Conclusions: Separate quantification of the two metabolites is important for interpreting brain ketone studies and for understanding the full pharmacology of KME supplementation.

Compared to traditional gross volumetrics, surface- based models provide greater spatial precision for understanding brain alterations related to developmental, neurological, and psychiatric disorders. Large-scale brain initiatives are combining data from around the world to discover and improve illness- related brain markers. Here, we present a toolkit for 3D brain geometry analysis aimed at addressing key challenges facing large- scale neuroimaging studies. Our framework incorporates scalable methods for multisite data integration, site-specific confound correction, accelerated statistical modeling, interpretable machine learning, and interactive results visualization. The toolkit was tested on data from 21 independently collected study samples participating in the ENIGMA Bipolar Disorder Working Group (N = 3,373). Compared to traditional volume features, we show how subcortical shape measures can be combined across study sites to capture spatially complex differences between diagnostic groups and associations with common treatments. Statistical modeling was accelerated using the Fast and Efficient Mixed- Effects Algorithm (FEMA) and achieved a 16-fold reduction in computation time compared to traditional approaches. Machine learning models showed shape features may provide greater predictive performance over traditional volumes for both diagnostic and treatment prediction tasks, with interpretable weight maps providing insights into the local features driving model performance.

Traumatic brain injury produces widespread axonal damage can be assessed histologically using amyloid precursor protein (APP) immunohistochemistry, which labels injured axonal profiles at cellular resolution [1, 2]. However, quantification of APP pathology remains a major bottleneck: annotation is manual, time-consuming, spatially localized, and variable across raters, limiting scalability and reproducibility. This limitation is particularly important in studies that use histology as a reference for neuroimaging or other tissue-level measurements, where cellular APP pathology must be quantified in a spatial form that can be aligned with imaging abnormalities. Here, we introduce PIGMENT, an annotation-efficient deep-learning framework for automated segmentation and quantification of APP-positive pathology in porcine white matter histology. PIGMENT uses a compact SegFormer-B0 architecture trained on 525 expert-annotated 512 x 512-pixel tiles from four APP-stained sections across three pigs. Because APP-positive profiles are sparse, fragmented, stain-variable, and morphologically diverse, PIGMENT combines limited expert labels with APP-specific augmentation designed to model variation in APP-positive intensity, size, continuity, fragmentation, and local tissue context. We evaluated PIGMENT using an instance-level detection rate that measures whether discrete APP-positive components are localized. Across held-out APP-stained data, PIGMENT achieved a mean instance-level detection rate of 0.86. Across the configurations tested, the highest mean detection rate was achieved by a training set that included sections from different animals, suggesting that annotation diversity may be an important factor under limited-label conditions. By extending limited high-confidence expert annotations into whole-section APP burden maps, PIGMENT provides a scalable framework for characterizing the extent and spatial distribution of traumatic axonal injury. These maps may support future studies that align histological injury burden with imaging-derived measures.

Non-suicidal self-injury (NSSI) among adolescents is a prevalent mental health problem and an important indicator of potential suicide risk. Early objective identification and neural mechanism analysis are therefore crucial for clinical screening and intervention. Traditional assessments mainly rely on self-report scales and clinical interviews, which are vulnerable to subjective bias, clinical experience, and missed diagnosis. Electroencephalography (EEG), with its non-invasive, low-cost, and high-temporal-resolution characteristics, provides a promising physiological basis for identifying NSSI-related neural abnormalities. However, EEG-based intelligent recognition of adolescent NSSI remains limited, and existing studies often emphasize classification performance while lacking systematic neurophysiological interpretation. To address these issues, this study proposes CGA-NSSI, a lightweight deep learning framework for adolescent NSSI recognition. The model integrates a one-dimensional convolutional neural network, bidirectional gated recurrent unit, and multi-head self-attention mechanism to extract local spatiotemporal EEG features, model long-range temporal dependencies, and focus on key pathology-related time segments and channels. A standardized preprocessing pipeline, together with Mixup augmentation and Focal Loss, is further used to alleviate sample imbalance and improve robustness in small clinical EEG datasets. Experiments on a real-world adolescent clinical EEG dataset show that CGA-NSSI can effectively identify NSSI-related EEG patterns under imbalanced sample conditions. Interpretability and functional connectivity analyses further reveal prefrontal-centered cross-regional network reorganization, excessive static functional coupling, reduced dynamic connectivity fluctuations, and increased abnormal state occupancy. These findings suggest that CGA-NSSI not only improves objective NSSI recognition but also provides neurophysiological evidence for understanding adolescent self-injury.

IntroductionChemogenetic tools such as Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) provide a powerful means to causally manipulate defined neuronal populations in vivo. While chemogenetic fMRI studies have consistently demonstrated robust hemodynamic responses following circuit perturbation, considerably less is known about the accompanying metabolic consequences. Functional magnetic resonance spectroscopy (fMRS) offers the potential to probe these neurochemical processes, yet the relationship between hemodynamic and metabolic responses remains poorly understood. Here, we combined chemogenetics, pharmacological fMRI (ph-fMRI), and proton magnetic resonance spectroscopy (1H-MRS/fMRS) at 7 T to investigate the temporal evolution of metabolic and hemodynamic responses in the rat motor cortex. MethodsFemale Fischer rats received viral injections in the motor cortex to express either a pan-neuronal hM3D(Gq) DREADD construct (hSyn-hM3Dq) or an interneuron-targeted construct (hDlx-hM3Dq). Ph-fMRI, fMRS, and 1H-MRS measurements were performed before, during, and following systemic administration of clozapine-N-oxide (CNO, 1 mg/kg). Functional MRS was acquired during the acute response phase (0-60 min post-injection), while conventional 1H-MRS measurements were obtained at a delayed time point (70 min post-injection). ResultsChemogenetic modulation produced robust and opposing hemodynamic responses. Pan-neuronal activation elicited focal positive BOLD responses (+3.5 {+/-} 1.5%), whereas interneuron-targeted activation generated significant negative BOLD responses (-3.3 {+/-} 0.8%). In contrast, acute fMRS measurements revealed no significant changes in Glx or GABA concentrations during the first hour following CNO administration, despite the presence of strong hemodynamic effects. However, delayed metabolic alterations were detected 70 min after CNO administration. Animals expressing the pan-neuronal construct exhibited significant increases in GABA (+14.4%) and total choline compounds (+57.8%), whereas interneuron-targeted animals displayed reductions in several metabolites, including Glx (-15.6%), total NAA (-16.9%), glucose (-25.9%), and total creatine (-25.4%). ConclusionChemogenetic perturbation of cortical circuits produced robust hemodynamic responses but more subtle and temporally complex metabolic effects. The absence of detectable acute changes in Glx and GABA despite strong BOLD responses, together with the emergence of delayed neurochemical alterations, highlights the challenges of interpreting metabolic signals in relation to circuit activity.

The thalamic reticular nucleus (TRN) is a thin, inhibitory shell surrounding the thalamus. It regulates the thalamocortical information flow, and thereby plays a central role in attention, task switching, and the sleep-wake cycle. Despite its importance, the TRN remains poorly studied in the human brain. This is largely because its small size and deep anatomical location limit its visibility with conventional non-invasive neuroimaging techniques. Here, we assessed whether the human TRN can be reliably visualised and segmented in vivo using ultra-high field (UHF) magnetic resonance imaging (MRI) at 7 Tesla. High resolution (0.35 mm isotropic) partial-brain T2* and T1 scans were acquired from healthy individuals, followed by manual delineation of the TRN. These in vivo segmentations were compared with TRN estimates obtained from two high-quality postmortem datasets serving as an anatomical reference. In vivo segmentations of TRN volume and thickness closely matched measurements derived from the postmortem reference datasets, and quantitative comparisons showed high consistency in TRN shape and location across individuals while also capturing meaningful inter-individual variability. Using these segmentations, we constructed a publicly available probabilistic atlas of the human TRN. This atlas provides a new resource for incorporating TRN anatomy into functional, structural, and clinical neuroimaging studies. Our findings demonstrate that the human TRN can be robustly mapped in vivo at 7T and establish a foundation for future investigations into its structure and function.

Dementia is a progressive neurological condition characterized by cognitive decline and structural brain changes that evolve. Longitudinal modeling of these changes is important for improving disease monitoring, identifying progression patterns, and supporting early risk stratification. This study developed an explainable longitudinal machine-learning framework for dementia progression, using cognitive and Magnetic Resonance Imaging (MRI)-derived biomarkers from the Open Access Series of Imaging Studies (OASIS-2) longitudinal dataset. The dataset included 150 subjects and 373 repeated observations classified as Non-demented, Demented, or Converted. Current-visit features, previous-visit features, and slope-based temporal features were constructed from Mini-Mental State Examination, Clinical Dementia Rating, normalized whole-brain volume, estimated total intracranial volume, atlas scaling factor, Age, and MRI delay. Baseline models were compared with a longitudinal gradient-boosted model, using patient-level splitting to reduce data leakage across repeated visits. The proposed longiGradient Gradient boosting model achieved the best held-out test performance, with an accuracy of 88.16%, a macro F1-score of 0.776, and a weighted F1-score of 0.860. The model showed strong classification performance for Demented and Non-demented individuals, while converted cases remained more difficult to identify. A regularized gradient boosting model was also evaluated as an overfitting sensitivity analysis; although it reduced the perfect training fit, it did not improve held-out test performance. Feature importance, permutation importance, and SHapley Additive exPlanations identified Clinical Dementia Rating as the dominant predictor, with slope-based Clinical Dementia Rating providing additional longitudinal information. These findings suggest that combining cognitive measures, MRI-derived biomarkers, and temporal feature engineering can improve dementia progression modeling, although external validation in larger longitudinal cohorts is needed.

Purpose: To investigate sex differences in cerebral blood flow through densely parcellated cortical and subcortical regions using explainable artificial intelligence methods and identify neurobiologically interpretable perfusion biomarkers. Methods: High-resolution pseudo-continuous arterial spin labelling (1.875 mm x 1.875 mm x 3 mm) and structural MRI data were curated from 215 healthy young adults (150 females, 95 males; age 18-30 years) from the publicly available I See your Brains (ISYB) dataset. Cerebral blood flow was quantified using atlas-based regional analysis with the Brainnetome Atlas (246 regions) and optimized registration procedures. Sex classification employed diverse machine learning paradigms including linear classifiers, ensemble methods, and kernel-based approaches for regional CBF features, with deep convolutional neural networks (CNN) applied to whole-brain 3D imaging data. Model interpretability was achieved using SHapley Additive exPlanations (SHAP), computed over an ensemble of 500 logistic regression models (100 iterations x 5-fold cross-validation). Regions appearing among the top 20% of discriminative features more than 289 times were considered statistically significant using binomial testing. GradCAM was used to obtain class-specific attribution maps from the CNN model. Results: Perfusion-based features demonstrated superior sex classification performance compared to structural morphometry. Regional CBF analysis using logistic regression achieved 91 +/- 2% balanced accuracy and 0.95 +/- 0.05 ROC-AUC, substantially outperforming morphometric features (85 +/- 8% balanced accuracy, 0.88 +/- 0.06 ROC-AUC). Deep learning classification of 3D CBF maps achieved a performance of 92 +/- 5% balanced accuracy, 0.92 +/- 0.05 ROC-AUC. SHAP analysis identified 30 statistically significant aggregation-agnostic CBF-based biomarker regions using regional CBF, predominantly involving frontoparietal control networks (27%) and default mode networks (17%). Grad-CAM revealed that the 3D CNN model primarily focused on regions within the frontal lobe. Morphometry-based analysis identified 28 discriminative regions with markedly different anatomical distribution (r = 0.21) emphasizing visual (32%) and default mode (14%) networks. Conclusion: Cerebral blood flow patterns provide highly sensitive and biologically interpretable markers of sex differences in young adult brain. The identification of robust perfusion biomarkers through explainable AI demonstrates the clinical potential of ASL imaging for precision medicine applications in neuroscience. We establish a methodological framework for investigating sex-specific brain physiology using non-invasive neuroimaging.

Longitudinal dementia progression prediction is essential for clinical decision-making. However, models often degrade on external cohorts due to systemic missingness -- where certain biomarkers available during training are completely absent at test time -- compounded by distribution shifts and patient-specific variability. Here, we propose Progression-aware Feature Fusion with Test-Time Adaptation (ProFuse-TTA), a two-stage hierarchical Transformer for longitudinal dementia prediction. Stage 1 learns per-biomarker temporal representations from irregular observations without imputation. Stage 2 fuses them via cross-feature attention, with simulated modality dropout during training for robustness to systemic missingness. At inference, a lightweight test-time adaptation module performs per-individual calibration. We trained on ADNI and evaluated on three external cohorts comprising 2,316 participants and 13,205 timepoints, with controlled modality ablation experiments isolating the effect of systemic missingness. We compared against six baselines, four from a recent benchmark study and two new baselines including one built on a tabular foundation model. ProFuse-TTA achieved the best cross-dataset performance in 8 of 9 settings across clinical diagnosis, MMSE, and hippocampal volume prediction, and ranked first in 14 of 15 ablation scenarios. The model maintained superior performance across varying input lengths and prediction horizons up to 6 years. Pretrained ADNI models are available at XXX.

A central challenge in neuroscience is understanding how the human brain is organised to support optimal functioning and adaptability. One approach to characterise complex brain dynamics is by artificially perturbing whole-brain models. Here, we asked whether whole-brain organisation under perturbation in major depressive disorder (MDD) changes after intervention with psilocybin and escitalopram. First, we built whole-brain models of pre- and post-treatment resting-state functional magnetic resonance imaging (fMRI) and obtained an initial generative effective connectivity (GEC) matrix for each individual. Then, we employed systematic and local artificial perturbations across intensities, re-optimised each model to create a response GEC (GECr), and assessed the extent of brain reorganisation by quantifying the brain network reconfiguration index (NRI). Our results showed that the global brain NRI increases with psilocybin and decreases with escitalopram. Across sessions and interventions, higher global NRI was related with localised perturbations in brain areas orchestrating the brain's hierarchical dynamics. Traditional approaches complemented our investigation. Our findings suggest distinct neural changes following each treatment for MDD. The increase in brain reorganisation under perturbation following psilocybin is consistent with greater brain flexibility and changeability, whereas the decrease following escitalopram suggests more stabilised brain dynamics. Overall, perturbation-induced brain NRI may represent a useful approach for uncovering neural changes following different interventions for depression.