Search Results (247)

Background/Objectives: Multimodal data fusion is increasingly applied in neuroinformatics to integrate heterogeneous sources of information. However, the optimal strategies for combining modalities with markedly different dimensionality, scale, and noise characteristics remain unclear. To our knowledge, this is among the first systematic and controlled benchmarks explicitly disentangling the effects of fusion strategy and feature scaling within a unified deep learning framework. Methods: Using data from 747 healthy participants from the Human Connectome Project, we evaluated multiple fusion paradigms—including early fusion, attention-based fusion, subspace-based fusion, and graph-based fusion—within a unified and reproducible framework. Importantly, we assessed how different feature scaling techniques (Standard, Min–Max, and Robust scaling) interact with fusion strategies and influence model performance. Biological sex was used as a controlled benchmark task to focus on methodological insights rather than task-specific optimization. Results: Early feature-level fusion consistently achieved the highest classification performance across all evaluated configurations. In particular, direct concatenation of cognitive and neuroimaging features combined with Standard Scaling yielded the best results (AUC–ROC = 0.96 (0.95–0.96)), outperforming unimodal baselines as well as intermediate and late fusion strategies. Conclusions: This systematic benchmark demonstrates that multimodal deep learning performance in neuroscience is driven primarily by the interaction between fusion strategy and feature scaling rather than by architectural complexity alone. By explicitly disentangling the effects of fusion level and preprocessing within a unified framework, this study provides practical methodological guidance for the design, evaluation, and reproducible deployment of multimodal deep learning models in neuroscience. Full article

Background: Glioblastoma is an extremely aggressive brain tumor that diffusely infiltrates white matter and alters large-scale brain connectivity. Most prognostic models focus on localized tumor features and clinical variables, overlooking broader effects on the brain’s structural connectome. This study addressed this limitation by integrating graph-theoretical analysis of preoperative diffusion tensor imaging (DTI)-derived structural connectomes with machine learning (ML) to improve prediction of overall survival (OS) in newly diagnosed glioblastoma. Methods: Preoperative DTI data from 871 glioblastoma patients from the UPenn-GBM and UCSF-PDGM cohorts were processed to construct whole-brain structural connectomes weighted by tract count and quantitative anisotropy (QA). Global and nodal graph-theoretical network metrics were extracted and combined with demographic and clinical information. Ten ML models were trained and validated on 784 patients (90% of the cohort). The three best-performing algorithms were tested on a held-out cohort of 87 patients (10%). Results: Random forest, adaptive boosting, and KStar showed the strongest validation performance. In held-out internal testing, random forest models using degree and QA-weighted strength achieved accuracies of 0.862 and 0.874, with AUROCs of 0.929 and 0.909, for predicting OS beyond one year. Strength and clustering coefficient were key predictors, with over two-thirds of significant nodes localized in the temporal lobe, particularly the parahippocampal, and superior, middle, and inferior temporal gyri. Conclusions: Graph-theoretical quantification of structural brain network disruption combined with ML allows accurate prediction of OS in glioblastoma. These results support a network-based conceptualization of the disease and indicate that connectome-derived metrics may complement established prognostic frameworks. Full article

Background: The recognition of autism spectrum disorder (ASD) has been a challenge due to the heterogeneity in symptoms and complex variations in brain function. Resting-state functional magnetic resonance imaging (rs-fMRI) has become instrumental in studying these disorders by accessing underlying abnormal neural activity and connectivity. Recently, deep learning approaches have shifted the analysis of brain networks by capturing spatiotemporal information from fMRI sequences. Nonetheless, most existing studies are limited by relying on a single representational scale, typically restricting analysis to either voxel-level spatiotemporal patterns or static connectivity matrices. Additionally, the dynamic reconfiguration of functional coupling and its variations across different anatomical parcellations are often ignored, which obscures neurobiologically meaningful dynamics. Methods: In this regard, we propose a multi-atlas dynamic connectivity transformer fused with 4D spatiotemporal modeling for ASD recognition (MADCT-4D). Specifically, the framework comprises two complementary branches. The 4D spatiotemporal branch encodes raw rs-fMRI volumes to learn hierarchical representations of evolving neural activity, while the dynamic-connectivity branch models time-resolved functional connectivity sequences constructed from multiple atlases, enabling the network to capture dynamic reconfiguration at the connectome level under different parcellation granularities. Moreover, we perform late fusion by combining the branch-specific decision scores with a learnable gate, allowing the model to adaptively weight voxel-level dynamics and multi-atlas connectivity evidence for each subject. Results: Extensive experiments on the publicly available ABIDE dataset demonstrate that the proposed method achieves 90.2% accuracy for ASD recognition, outperforming multiple competitive baselines. Conclusions: The proposed framework yields interpretable biomarkers based on learned dynamic connectivity patterns that are consistent with altered functional coupling in ASD. Full article

Volume electron microscopy (vEM) provides nanometer-scale 3D imaging, yet its axial (z) resolution is often much lower than the in-plane ($xy$) resolution, yielding anisotropic volumes that hinder segmentation and connectomic reconstruction. We present a two-stage cross-axial super-resolution framework for isotropic reconstruction that combines a conditional diffusion model and domain-specific self-supervised pretraining of a vision transformer (ViT). First, the student–teacher self-distillation paradigm of DINOv3 is adopted to learn representations from large sets of high-resolution $xy$ sections, capturing vEM-specific texture statistics and ultrastructural patterns. Second, a conditional diffusion denoiser is trained with supervised anisotropic degradation simulated by z-downsampling, while a perceptual loss based on frozen ViT feature distances constrains generated slices to match real-section distributions. These constraints recover axial high-frequency details and reduce hallucinated textures and inter-slice drift, improving cross-slice consistency. Experiments on two public vEM datasets show improved fidelity, perceptual quality, and membrane-boundary continuity over interpolation and learning-based baselines. Full article

Background: Disability progression in multiple sclerosis (MS) is increasingly recognized as a consequence of large-scale brain network disruption rather than isolated regional damage. Although diffusion tensor imaging (DTI) is the reference method for assessing structural connectivity, its limited availability restricts widespread clinical application. There is therefore a critical need for alternative approaches capable of capturing network-level alterations using routinely acquired MRI data. Objective: This study aimed to determine whether synthetic structural connectivity matrices derived from standard regional volumetric MRI can capture clinically meaningful network alterations in MS and predict subsequent functional progression, particularly upper limb decline. Methods: Regional brain volumetry was obtained from routine T1-weighted MRI using an automated, clinically approved volumetric pipeline. Synthetic structural connectivity matrices were generated by integrating principles of structural covariance, distance-dependent connectivity, and disease-specific vulnerability patterns. Graph-theoretical network metrics were extracted to characterize global and regional topology. Machine learning models including logistic regression, support vector machines, random forests, and gradient boosting were trained to predict clinical progression defined by worsening on the 9-Hole Peg Test. Dimensionality reduction was performed using principal component analysis, and model performance was evaluated using balanced accuracy, AUC-ROC, and resampling-based validation. Feature importance analyses were conducted to identify network vulnerability patterns. Results: Synthetic connectivity networks exhibited biologically plausible properties, including preserved but attenuated small-world organization. Global efficiency showed a strong inverse correlation with disability severity (EDSS). Patients with clinical progression demonstrated marked reductions in network integration and segregation, alongside increased characteristic path length. Machine learning models achieved robust prediction of upper limb functional decline, with ensemble-based methods performing best (balanced accuracy > 80%, AUC-ROC up to 0.85). A limited subset of connections accounted for a disproportionate share of predictive power, predominantly involving frontoparietal associative networks, thalamocortical pathways, and inter-hemispheric connections. In a longitudinal subset, network-level alterations preceded measurable clinical deterioration by several months. Conclusions: Synthetic structural connectivity derived from routine volumetric MRI captures clinically relevant network-level disruption in multiple sclerosis and enables accurate prediction of functional progression. By bridging network neuroscience with widely accessible imaging data, this framework provides a pragmatic alternative for connectomic analysis when diffusion imaging is unavailable and supports a network-based understanding of disease evolution in MS. Full article

Three-dimensional (3D) ex vivo imaging of cleared tissue from intact brains from animal models, human brain surgical specimens, and large postmortem human and non-human primate brain specimens is essential for understanding physiological neural connectivity and pathological alterations underlying neurological and neuropsychiatric disorders. Contemporary light-sheet microscopy enables rapid, high-resolution imaging of large, cleared samples but is limited by the orthogonal arrangement of illumination and detection optics, which constrains specimen size. Light-sheet theta microscopy (LSTM) overcomes this limitation by employing two oblique illumination paths while maintaining a perpendicular detection geometry. Here, we report the development of a next-generation, fully integrated and user-friendly LSTM system that enables uniform subcellular-resolution imaging (with subcellular resolution determined by the lateral performance of the system) throughout large specimens without constraining lateral (XY) dimensions. The system provides a seamless workflow encompassing image acquisition, data storage, pre- and post-processing, enhancement and quantitative analysis. Performance is demonstrated by high-resolution 3D imaging of intact mouse brains and human brain samples, including complete downstream analyses such as digital neuron tracing, vascular reconstruction and design-based stereological analysis. This enhanced and accessible LSTM implementation enables rapid quantitative mapping of molecular and cellular features in very large biological specimens. Full article

Electrophysiology, mechanobiology, and the study of soft matter within cells demonstrate increasing amounts of evidence that neuronal signaling arises from interactions between membrane potential, force, and phase. Herein, we have attempted to collect and organize the evidence for each of these areas of study into an approximate structure called the electromechanical connectome: a three-way state–space (membrane potentials, nanoscale mechanical forces, and cytoplasmic rheology, including phase-separated liquid–liquid droplets) where membrane potentials, nanoscale mechanical forces, and cytoplasmic rheology, and phase-separated liquid–liquid droplets are likely to influence one another, influencing synaptic processing, plasticity and network stability. We will also attempt to illustrate the following: how changes in electrostatic fields can be used to alter the arrangement of lipids, hydration, and dielectric microdomains, and the contact geometry between organelles and activity dependent transcription; how mechanical dynamics associated with spines, axons, and the active zone of synapses may be used to modify the energy landscape of channels, the docking and priming of vesicles, and the transport of cytoskeletons; and how viscosity corridors, along with phase-separated micro-reactors, can be used to regulate the kinetics of signaling, molecular trafficking and metabolic processes in local environments. With these connections in mind, we will propose a multiphysical attractor model in which cognition is the result of navigating through metastable manifolds, while neurodegenerative disease may be a result of the progressive loss of electromechanical coherence, phase boundary control and energetic flexibility. Finally, we will present testable hypotheses and use AI-enabled digital twin methods to potentially quantify the early deformation of manifolds and provide precision biomarkers and therapeutic options. Full article

Background: Stroke is a leading cause of mortality and long-term disability, and aphasia is among its most common and debilitating sequelae. Diffusion tensor imaging (DTI) and advanced diffusion imaging techniques enable the assessment of white matter integrity and provide clinically relevant measures in post-stroke aphasia. Methods: We conducted a comprehensive review of studies applying DTI or advanced diffusion imaging to investigate structural connectivity in adults with post-stroke aphasia (PSA). PubMed, CENTRAL, Ovid MEDLINE, and Embase were searched, and eligible studies were synthesized according to their diagnostic, prognostic, or therapeutic focus. Results: Ninety-five studies were included. Of these, 59 were classified as diagnostic, 17 as prognostic, and 19 as therapeutic. Most studies employed conventional DTI (n = 77), while a growing body of research utilized advanced diffusion models, including CSD, DSI, and DKI (n = 18). Conclusions: This comprehensive synthesis demonstrates the evolution of diffusion imaging in PSA research. While conventional DTI has provided foundational insights, advanced diffusion methods offer superior characterization of complex fiber architecture and improved clinical–anatomical correlation. Diffusion-derived markers of dorsal and ventral language pathways were consistently associated with language performance, while connectome-level analyses highlighted the importance of preserved global network architecture for recovery. Continued efforts are needed to translate diffusion imaging findings into clinical applicable biomarkers to guide personalized aphasia rehabilitation, with greater use of advanced methods. Full article

Background: Deep brain stimulation (DBS) has traditionally followed diagnosis-driven, nucleus-centered targeting paradigms. Increasing evidence supports a circuit-based framework in which clinical outcomes depend on modulation of symptom-relevant networks rather than diagnostic labels alone. This approach is particularly relevant in mixed movement disorder phenotypes such as dystonic tremor, where the most disabling symptom may not align with the conventional surgical target. Methods: We report a clinically illustrative single case treated using a symptom-oriented, connectome-informed DBS strategy. Clinical phenotype, tremor severity, functional impairment, prior medical and botulinum toxin treatments, and longitudinal outcomes were systematically reviewed. DBS target selection prioritized the dominant, treatment-refractory symptom rather than the underlying dystonia diagnosis. Surgical planning incorporated high-resolution MRI with patient-specific thalamic segmentation using Brainlab Brain Elements^®, followed by postoperative lead localization and volume of tissue activated visualization with the SureTune™ platform. Results: A 54-year-old left-handed woman with long-standing cervical dystonia developed a severe, markedly asymmetric dystonic tremor predominantly affecting the left upper limb, resulting in profound functional disability. Instead of conventional bilateral globus pallidus internus DBS, unilateral right ventral intermediate nucleus (VIM) DBS was selected to engage tremor-related cerebellothalamic circuits. Rapid and marked improvement was observed, with tremor severity reduced to mild levels within 15 days after stimulation onset. At 6-month follow-up, overall tremor severity improved from 49 to 13 points on the Fahn–Tolosa–Marin Tremor Rating Scale, corresponding to a 73.5% reduction. This improvement was associated with restoration of legible handwriting, independent feeding and drinking, and recovery of bimanual fine motor function. Clinical benefit remained stable throughout follow-up, without stimulation-related adverse effects. Conclusions: This case illustrates the feasibility of a symptom-oriented, connectome-informed DBS strategy in selected patients with dystonic tremor. When symptom expression and network involvement are markedly asymmetric, selective unilateral modulation of the tremor-dominant circuit may achieve meaningful and durable functional improvement. Further studies are needed to assess the generalizability of this approach. Full article

Research has demonstrated that axonal signaling processes are influenced by both static structural factors and dynamic metabolic and electro-dynamic processes. Imaging, computational modeling and research in molecular neuroscience have demonstrated that multiple processes contribute to axonal signal processing, including periodic rearrangement of cytoskeletal structures and membrane structures, and redistribution of ion channel clusters and organelles (such as mitochondria), which occur rapidly and transiently to modify excitability. The dynamics of energy production and distribution also vary between regions of the axon and at different time points during signal generation and transmission. Additionally, myelin-associated glia may temporarily modulate their metabolic and structural contributions to axonal conduction. Advanced AI-based techniques for mapping and simulating ultrastructure and the use of closed-loop perturbation experiments demonstrate that axons can generate multiple distinct electromechanical states, and therefore potentially influence both the timing of signals generated by the axon, the routing of signals to branches of the axon, and the robustness of signal propagation. While the existence of these adaptive microstates appears well established, there are many aspects of their influence on circuit level function that are poorly understood. In summary, these data support the concept that axonal conduction represents a continuum of reversible and state-dependent configurations generated by integrated interactions among molecular, structural and energetic processes. Therefore, this review will attempt to synthesize the available literature into a unified conceptual framework and identify areas of uncertainty that may direct future research into the adaptive processes underlying axonal computation. Full article

Attention Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder that is found in both children and adults. While this disorder often continues in adulthood, diagnosis can be challenging, particularly in females. Unlike males, who are often diagnosed with ADHD due to their externalizing behaviors (i.e., impulsive nature), most females show inattentive symptoms (i.e., in focusing, disorganization), which makes this disorder hard to detect. This paper proposes a machine learning approach to detect ADHD among females. The Wids Datathon 2025 provides three datasets: categorical data, quantitative data, and function connectomes. It contains information on 1213 participants who are seeking to take a test to detect ADHD. Categorical data includes 10 attributes, quantitative data has 19 attributes, and functional connectomes contain 19,901 attributes which are relevant to studying the participants’ overall condition. By combining both XGBoost and Random Forest, an accuracy of 79.42% was achieved. The results show that machine learning algorithms can help in improving ADHD detection in females, leading to better diagnoses in future. Full article

The National Institutes of Health (NIH) Toolbox cognition battery and Trail Making Tests (TMT) are widely used to quantify cognitive aging and to detect early cognitive vulnerability in Alzheimer’s disease and related conditions. However, these tests are often treated as interchangeable markers of global cognition, despite likely differences in their dependence on specific brain systems, limiting interpretability across studies and clinical contexts. To address this gap, we examined associations between four commonly used cognitive measures—fluid cognition, crystallized cognition, TMT-A, and TMT-B—and multimodal MRI metrics in 725 healthy volunteers aged 36 to 100 years from the Human Connectome Project–Aging. Voxel-wise diffusion MRI and vertex-wise cortical thickness and volume analyses were adjusted for age, sex, and years of education. Higher crystallized and fluid cognition scores and faster TMT-A/B completion times were generally associated with greater white matter integrity. TMT-B showed the most extensive diffusion and cortical associations, involving major projection, commissural, and association pathways and frontoparietal and temporo-occipital cortices. TMT-A and crystallized cognition demonstrated intermediate, overlapping patterns, whereas fluid cognition showed only focal brainstem and limited cortical correlates. These findings demonstrate systematic differences in the neuroanatomical substrates underlying commonly used cognitive tests and provide normative structure–cognition reference maps that can improve test selection, mechanistic interpretation, and sensitivity to brain health in studies of aging, vascular risk, and preclinical neurodegenerative disease. Full article

Functional magnetic resonance imaging (fMRI) is a valuable tool for presurgical brain mapping, traditionally implemented with task-based paradigms (tb-fMRI) that measure blood oxygenation level-dependent (BOLD) signal changes during controlled motor or cognitive tasks. Tb-fMRI is a well-established tool for non-invasive localization of cortical eloquent areas, yet its dependence on patient cooperation and intact cognition limits use in individuals with aphasia, cognitive impairment, or in pediatric and other vulnerable populations. Resting-state fMRI (rs-fMRI) provides a task-free alternative by leveraging spontaneous low-frequency BOLD fluctuations to delineate intrinsic functional networks, including motor and language systems that show good spatial concordance with tb-fMRI and with direct cortical stimulation. This narrative review outlines the methodological foundations of tb-fMRI and rs-fMRI, comparing acquisition protocols, preprocessing and denoising pipelines, analytic approaches, and validation strategies relevant to presurgical planning. Particular emphasis is given to the technical and physiological foundations of BOLD imaging, statistical modeling, and the influence of motion, noise, and standardization on data reliability. Emerging evidence indicates that rs-fMRI can reliably expand mapping to patients with limited task compliance and may serve as a robust complementary modality in complex clinical contexts, though its methodological heterogeneity and absence of unified practice guidelines currently constrain widespread adoption. Future advances in harmonized preprocessing, multicenter validation, and integration with connectomics and machine learning frameworks are likely to be critical for translating rs-fMRI into routine, reliable presurgical workflows. Full article

We introduce the Onco-Hem Connectome (OHC), a patient similarity network (PSN) designed to organize real-world hemato-oncology inpatients by exploratory phenotypes with potential clinical utility. Background: Polypharmacy and drug–drug interactions (DDIs) are pervasive in hemato-oncology and vary with comorbidity and treatment intensity. Methods: We retrospectively analyzed a 2023 single-center cohort of 298 patients (1158 hospital episodes). Standardized feature vectors combined demographics, comorbidity (Charlson, Elixhauser), comorbidity polypharmacy score (CPS), aggregate DDI severity score (ADSS), diagnoses, and drug exposures. Cosine similarity defined edges (threshold ≥ 0.6) to build an undirected PSN; communities were detected with modularity-based clustering and profiled by drugs, diagnosis codes, and canonical chemotherapy regimens. Results: The OHC comprised 295 nodes and 4179 edges (density 0.096, modularity Q = 0.433), yielding five communities. Communities differed in comorbidity burden (Kruskal–Wallis ${\epsilon }^2$: Charlson 0.428, Elixhauser 0.650, age 0.125, all FDR-adjusted p < 0.001) but not in utilization (LOS, episodes) after FDR (${\epsilon }^2$ ≈ 0.006–0.010). Drug enrichment (e.g., enoxaparin $\Delta$ = +0.13 in Community 2; vinblastine $\Delta$ = +0.09 in Community 3) and principal diagnoses (e.g., C90.0 23%, C91.1 15%, C83.3 15% in Community 1) supported distinct clinical phenotypes. Robustness analyses showed block-equalized features preserved communities (ARI 0.946; NMI 0.941). Community drug signatures and regimen signals aligned with diagnosis patterns, reflecting the integration of resource-use variables in the feature design. Conclusions: The Onco-Hem Connectome yields interpretable, phenotype-level insights that can inform supportive care bundles, DDI-aware prescribing, and stewardship, and it provides a foundation for phenotype-specific risk models (e.g., prolonged stay, infection, high-DDI episodes) in hemato-oncology. Full article

Magnetic resonance imaging (MRI) super-resolution (SR) enables high-resolution reconstruction from low-resolution acquisitions, reducing scan time and easing hardware demands. However, most deep learning-based SR models are large and computationally heavy, limiting deployment in clinical workstations, real-time pipelines, and resource-restricted platforms such as low-field and portable MRI. We introduce CHARMS, a lightweight convolutional–Transformer hybrid with attention regularization optimized for MRI SR. CHARMS employs a Reverse Residual Attention Fusion backbone for hierarchical local feature extraction, Pixel–Channel and Enhanced Spatial Attention for fine-grained feature calibration, and a Multi-Depthwise Dilated Transformer Attention block for efficient long-range dependency modeling. Novel attention regularization suppresses redundant activations, stabilizes training, and enhances generalization across contrasts and field strengths. Across IXI, Human Connectome Project Young Adult, and paired 3T/7T datasets, CHARMS (~1.9M parameters; ~30 GFLOPs for 256 × 256) surpasses leading lightweight and hybrid baselines (EDSR, PAN, W2AMSN-S, and FMEN) by 0.1–0.6 dB PSNR and up to 1% SSIM at ×2/×4 upscaling, while reducing inference time ~40%. Cross-field fine-tuning yields 7T-like reconstructions from 3T inputs with ~6 dB PSNR and 0.12 SSIM gains over native 3T. With near-real-time performance (~11 ms/slice, ~1.6–1.9 s per 3D volume on RTX 4090), CHARMS offers a compelling fidelity–efficiency balance for clinical workflows, accelerated protocols, and portable MRI. Full article