Biology subjects

Iraji, A.

Publications and source records attributed to Iraji, A..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

Dynamic fusion of structural and functional connectivity via joint connectivity matrix ICA

The integration of multimodal MRI images, including functional MRI (fMRI) and diffusion MRI (dMRI), depicts a key advancement in neuroimaging, since it offers a more comprehensive and better understanding of brain function and connections. FMRI captures brain functional activity while dMRI reveals structural connectivity via white matter bundles, each of which provide unique yet complementary insights; however combing these two modalities, particularly at a dynamic level, is challenging due to their drastically different data characteristics. This study introduces a novel framework named "dynamic fusion," which extends joint component analysis (cmICA) to integrate static structural connectivity (SC) with dynamic functional connectivity (FC). Our approach gauges the relationship between these joint components across various temporal states, aiming to discover both static and dynamic features of brain connectivity. We applied this approach to fMRI and dMRI data from the same set of control subjects, which we had previously studied to estimate joint parcellation and their structural and functional connections using a static model only, and also included a comparable number of individuals with schizophrenia from the same study. Our results reveal that dynamic fusion not only highlights diverse temporal dynamics in FC but, more importantly, also shows how SC patterns differ across dynamic functional states at different time frames, providing new insights into brain organization. Furthermore, it successfully detects joint structural-functional connectivity differences between individuals with schizophrenia and controls, demonstrating its potential for detecting hidden group differences. Overall, this study establishes our dynamic fusion as a powerful tool for integrating structural and dynamic functional connectivity data, enhancing our understanding of brain connectivity and offering new perspectives for studying neurological disorders.

neuroscience↗

bioRxiv2025

LINKING MULTI-SCALE BRAIN CONNECTIVITY WITH VIGILANCE, WORKING MEMORY, AND BEHAVIOR IN ADOLESCENTS

This study examines how multi-scale intrinsic connectivity networks (ICNs) relate to cognitive and behavioral functions in adolescents, focusing on attention/vigilance, working memory, and behavioral regulation. Leveraging the NeuroMark 2.2 multi-scale ICN template obtained from over 100,000 subjects, we obtained multi-scale ICNs from baseline resting-state fMRI data from the ABCD Study. For this study, we are interested in "the fronto-thalamo-cerebellar (FTC) circuitry" and choose the subdomains of Neuromark 2.2 that cover it: Cerebellar (CB), Subcortical - Extended Thalamic (SC-ET), Higher Cognition - Insular Temporal (HC-IT), and Higher Cognition - Frontal (HC-FR), previously identified as relevant to cognitive and behavioral functions. Employing a multivariate approach combining principal component analysis (PCA) and canonical correlation analysis (CCA), we examined associations between these multi-scale ICNs and cognitive-behavioral outcomes. Our findings revealed significant associations, particularly for one of the estimated canonical components, linking multi-scale ICNs to cognitive and behavioral measures across both discovery and replication sets. This connectivity pattern may serve as a potential marker for attention, working memory, and behavioral regulation, offering new insights into a wide spectrum of neurodevelopmental disorders including Attention-Deficit/Hyperactivity Disorder (ADHD).

neuroscience↗

bioRxiv2025

NeuroMark-HiFi: A Data-Driven Method for Detecting High-Spatial-Frequency Functional Brain Networks

ObjectiveThe Traditional functional neuroimaging approaches typically focus on low-frequency spatial structures, potentially overlooking critical fine-scale connectivity disruptions associated with brain disorders MethodsWe introduce NeuroMark-HiFi, a fully automated algorithm designed to enhance the detection of high-spatial-frequency functional brain network patterns. NeuroMark-HiFi systematically preserves and analyzes fine-grained network variations by integrating reference-informed independent component analysis (ICA), 3D high-frequency spatial filtering, and a frequency-informed ICA decomposition to extract high-frequency functional components with greater precision. ResultsSimulation studies and mathematical evaluations demonstrate that NeuroMark-HiFi significantly improves sensitivity to both individual and group differences driven by small local shifts in spatial patterns of intrinsic connectivity networks (ICNs). Compared to traditional methods, NeuroMark-HiFi revealed additional group differences between individuals with schizophrenia (SZ) and healthy controls (HC), particularly in the visual, sensorimotor, frontal, temporal, and insular networks. ConclusionNeuroMark-HiFi successfully captures biologically meaningful alterations in spatial network patterns that conventional approaches may miss. SignificanceBy improving sensitivity to subtle brain network alterations, NeuroMark-HiFi holds promise for early diagnosis, treatment monitoring, neurodevelopment studies, aging research, and multimodal biomarker discovery, advancing the goals of precision psychiatry and neuroscience.

bioengineering↗

bioRxiv2025

PET-derived amyloid patterns in gray and white matter across Alzheimer's disease: A high-model-order ICA

INTRODUCTIONAlzheimers Disease (AD) is a neurodegenerative disorder marked by gray matter (GM) changes driven by amyloid-beta (A{beta}) plaques and neurofibrillary tangles. While GM alterations are well documented, spatially distinct patterns of homogeneous A{beta} uptake and white matter (WM) involvement remain underexplored. METHODSWe applied high-order independent component analysis (ICA) to 716 [18F]Florbetapir PET scans, identifying 80 GM and 13 WM networks. Diagnostic and cognitive associations were evaluated via statistical modeling. RESULTSIdentified networks delineated a progression trajectory, with mild cognitive impairment (MCI) profiles in temporoparietal and frontal subdomains more closely aligned with AD than cognitively normal (CN) profiles. GM networks, including the hippocampal-entorhinal complex and precuneus, and WM networks, including the retrolenticular internal capsule, demonstrated robust associations with cognitive performance. DISCUSSIONOur findings highlight the utility of high-order ICA in identifying reproducible A{beta} networks and the contribution of WM networks, such as the posterior corpus callosum, in the early pathological landscape of AD.

neuroscience↗

bioRxiv2025

Topologically Optimized Intrinsic Brain Networks

The estimation of brain networks is instrumental in quantifying and evaluating brain function. Nevertheless, achieving precise estimations of subject-level networks has proven to be a formidable task. In response to this challenge, researchers have developed group-inference frameworks that leverage robust group-level estimations as a common reference point to infer corresponding subject-level networks. Generally, existing approaches either leverage the common reference as a strict, voxel-wise spatial constraint (i.e., strong constraints at the voxel level) or impose no constraints. Here, we propose a targeted approach that harnesses the topological information of group-level networks to encode a high-level representation of spatial properties to be used as constraints, which we refer to as Topologically Optimized Intrinsic Brain Networks (TOIBN). Consequently, our method inherits the significant advantages of constraint-based approaches, such as enhancing estimation efficacy in noisy data or small sample sizes. On the other hand, our method provides a softer constraint than voxel-wise penalties, which can result in the loss of individual variation, increased susceptibility to model biases, and potentially missing important subject-specific information. Our analyses show that the subject maps from our method are less noisy and true to the group networks while promoting subject variability that can be lost from strict constraints. We also find that the topological properties resulting from the TOIBN maps are more expressive of differences between individuals with schizophrenia and controls in the default mode, subcortical, and visual networks.

neuroscience↗

bioRxiv2025

Building Multivariate Molecular Imaging Brain Atlases Using the NeuroMark PET Independent Component Analysis Framework

Molecular imaging analyses using positron emission tomography (PET) data often rely on macro-anatomical regions of interest (ROI), which may not align with chemo-architectural boundaries and obscure functional distinctions. While methods such as independent component analysis (ICA) have been useful to address this limitation, the fully data-driven nature can make it challenging to compare results across studies. Here, we introduce the NeuroMark PET approach, utilizing spatially constrained independent component analysis to define overlapping regions that may reflect the brains molecular architecture. We first generate an ICA template for the PET radiotracer florbetapir (FBP), targeting amyloid-{beta} (A{beta}) accumulation in the brain, using blind ICA on large datasets to identify replicable independent components. Only components that targeted A{beta} were included in this study, defined as A{beta} networks (A{beta}Ns), by omitting components targeting myelin or other non-A{beta} targets. Next, we use the A{beta}Ns as priors for spatially constrained ICA, resulting in a fully automated ICA pipeline called NeuroMark PET. This NeuroMark pipeline, including its A{beta}Ns, was validated against a standard neuroanatomical PET atlas, using data from the Alzheimers Disease Neuroimaging Initiative (ADNI). The study included 296 cognitively normal participants with FBP PET scans and 173 with florbetaben (FBB) PET scans, an analogue radiotracer also targeting A{beta} accumulation. Our results show that NeuroMark PET captures biologically meaningful, participant-specific features, such as subject specific loading values, consistent across individuals, and also shows higher sensitivity and power for detecting age-related changes compared to traditional atlas-based ROIs. Using this framework, we also highlight some of the advantages of using ICA analysis for PET data. In this study, an A{beta}N consists of weighted voxels and forms a pattern throughout the entire brain. For example, components may have weighted values at every voxel and can overlap with one another, enabling the separation of artifacts which may coincide with the A{beta}Ns of interest. In addition, this approach allows for the differentiation, separating white matter components, which may overlap in complex ways with the A{beta}Ns, mainly residing in the neighboring gray matter. Results also showed that the most age associated A{beta}N (representing the cognitive control network, CC1) exhibited a stronger association with age compared with macro-anatomical regions of interest. This may suggest that each NeuroMark FBP A{beta}N represents a spatial network following chemo-architectural uptake with greater biological relevance compared with anatomical ROIs. In summary, the proposed NeuroMark PET approach offers a fully automated framework, providing accurate and reproducible brain A{beta}Ns. This approach enhances our ability to investigate the molecular underpinnings of brain function and pathology, offering an alternative to traditional ROI-based analyses.

neuroscience↗

bioRxiv2025

Identifying Neurobiological Psychosis Biotypes Using Multi-Scale Functional Network Connectivity and its Latent Independent Subspace

This study aims to identify Psychosis Imaging Neurosubtypes (PINs)-- homogeneous subgroups of individuals with psychosis characterized by distinct neurobiology derived from imaging features. Specifically, we utilized resting-state fMRI data from 2103 B-SNIP 1&2 participants (1127 with psychosis, 350 relatives, 626 controls) to compute subject-specific multiscale functional network connectivity (msFNC). We then derived a low-dimensional neurobiological subspace, termed Latent Network Connectivity (LNC), which captured system-wide interconnected multiscale information across three components (cognitive-related, typical, psychosis-related). Projections of psychosis participants msFNC onto this subspace revealed three PINs through unsupervised learning, each with distinct cognitive, clinical, and connectivity profiles, spanning all DSM diagnoses (Schizophrenia, Bipolar, Schizoaffective). PIN-1, the most cognitively impaired, showed Cerebellar-Subcortical and Visual-Sensorimotor hypoconnectivity, alongside Visual-Subcortical hyperconnectivity. Most cognitively preserved PIN-2 showed Visual-Subcortical, Subcortical-Sensorimotor, and Subcortical-Higher Cognition hypoconnectivity. PIN-3 exhibited intermediate cognitive function, showing Cerebellar-Subcortical hypoconnectivity alongside Cerebellar-Sensorimotor and Subcortical-Sensorimotor hyperconnectivity. Notably, 55% of relatives aligned with the same neurosubtype as their affected family members--a significantly higher rate than random chance (p-valueRelatives-to-PIN-1 < 0.001, p-valueRelatives-to-PIN-2 < 0.05, p-valueRelatives-to-PIN-3 < 0.001) compared to a non-significant 37% DSM-based classification, supporting a biological basis of these neurosubtypes. Cognitive performance reliably aligns with distinct brain connectivity patterns, which are also evident in relatives, supporting their construct validity. Our PINs differed from original B-SNIP Biotypes, which were determined from electrophysiological, cognitive, and oculomotor data. These findings underscore the limitations of DSM-based classifications in capturing the biological complexity of psychotic disorders and highlight the potential of imaging-based neurosubtypes to enhance our understanding of the psychosis spectrum.

neuroscience↗

bioRxiv2025

Deciphering Multiway Multiscale Brain Network Connectivity:Insights from Birth to 6 Months

Converging evidence suggests that understanding the human brain requires more than just examining pairwise functional brain interactions. The human brain is a complex, nonlinear system, and focusing solely on linear pairwise functional connectivity often overlooks important nonlinear and higher-order relationships. Infancy is a critical period marked by significant brain development that could contribute to future learning, health, and life success. Exploring higher-order functional relationships in the brain can provide insight into brain function and development. To the best of our knowledge, there is no existing research on multiway, multiscale brain network interactions in infants. In this study, we comprehensively investigate the interactions among brain intrinsic connectivity networks (ICNs), including both pairwise (pair-FNC) and triple relationships (tri-FNC). We focused on an infant dataset collected between birth and six months, a critical period for brain maturation. Our results revealed significant hierarchical, multiway, multiscale brain functional network interactions in the infant brain. These findings suggest that tri-FNC provide additional insights beyond what pairwise interactions reveal during early brain development. The tri-FNC predominantly involve the default mode, sensorimotor, visual, limbic, language, salience, and central executive domains. Notably, these triplet networks align with the classical triple network model of the human brain, which includes the default mode network, the salience network, and the central executive network. This suggests that the brain network system might already be initially established during the first six months of infancy. Interestingly, tri-FNC in the default mode and salience domains showed significantly stronger nonlinear interactions with age compared to pair-FNC. We also found that pair-FNC were less effective at detecting these networks. The present study suggests that exploring tri-FNC can offer additional insights beyond pair-FNC by capturing higher-order nonlinear interactions, potentially yielding more reliable biomarkers to characterize developmental trajectories.

neuroscience↗

bioRxiv2024

Evidence for white matter intrinsic connectivity networks at rest and during a task: a large-scale study and templates

Understanding white matter (WM) functional connectivity is crucial for unraveling brain function and dysfunction. In this study, we present a novel WM intrinsic connectivity network (ICN) template derived from over 100,000 fMRI scans, identifying 97 robust WM ICNs using spatially constrained independent component analysis (scICA). This WM template, combined with a previously identified gray matter (GM) ICN template from the same dataset, was applied to analyze a resting-state fMRI (rs-fMRI) dataset from the Bipolar-Schizophrenia Network on Intermediate Phenotypes 2 (BSNIP2; 590 subjects) and a task-based fMRI dataset from the MIND Clinical Imaging Consortium (MCIC; 75 subjects). Our analysis highlights distinct spatial maps for WM and GM ICNs, with WM ICNs showing higher frequency profiles. Modular structure within WM ICNs and interactions between WM and GM modules were identified. Task-based fMRI revealed event-related BOLD signals in WM ICNs, particularly within the corticospinal tract, lateralized to finger movement. Notable differences in static functional network connectivity (sFNC) matrices were observed between controls (HC) and schizophrenia (SZ) subjects in both WM and GM networks. This open-source WM NeuroMark template and automated pipeline offer a powerful tool for advancing WM connectivity research across diverse datasets.

neuroscience↗

bioRxiv2024

st-DenseViT: A Weakly Supervised Spatiotemporal Vision Transformer for Dense Prediction of Dynamic Brain Networks

ObjectiveModeling dynamic neuronal activity within brain networks enables the precise tracking of rapid temporal fluctuations across different brain regions. However, current approaches in computational neuroscience fall short of capturing and representing the spatiotemporal dynamics within each brain network. We developed a novel weakly supervised spatiotemporal dense prediction model capable of generating personalized 4D dynamic brain networks from fMRI data, providing a more granular representation of brain activity over time. MethodsWe developed a model that leverages the vision transformer (ViT) as its backbone, jointly encoding spatial and temporal information from fMRI inputs using two different configurations: space-time and sequential encoders. The model generates 4D brain network maps that evolve over time, capturing dynamic changes in both spatial and temporal dimensions. In the absence of ground-truth data, we used spatially constrained windowed independent component analysis (ICA) components derived from fMRI data as weak supervision to guide the training process. The model was evaluated using large-scale resting-state fMRI datasets, and statistical analyses were conducted to assess the effectiveness of the generated dynamic maps using various metrics. ResultsOur model effectively produced 4D brain maps that captured both inter-subject and temporal variations, offering a dynamic representation of evolving brain networks. Notably, the model demonstrated the ability to produce smooth maps from noisy priors, effectively denoising the resulting brain dynamics. Additionally, statistically significant differences were observed in the temporally averaged brain maps, as well as in the summation of absolute temporal gradient maps, between patients with schizophrenia and healthy controls. For example, within the Default Mode Network (DMN), significant differences emerged in the temporally averaged space-time configurations, particularly in the thalamus, where healthy controls exhibited higher activity levels compared to subjects with schizophrenia. These findings highlight the models potential for differentiating between clinical populations. ConclusionThe proposed spatiotemporal dense prediction model offers an effective approach for generating dynamic brain maps by capturing significant spatiotemporal variations in brain activity. Leveraging weak supervision through ICA components enables the model to learn dynamic patterns without direct ground-truth data, making it a robust and efficient tool for brain mapping. SignificanceThis work presents an important new approach for dynamic brain mapping, potentially opening up new opportunities for studying brain dynamics within specific networks. By framing the problem as a spatiotemporal dense prediction task in computer vision, we leverage the spatiotemporal ViT architecture combined with weakly supervised learning techniques to efficiently and effectively estimate these maps.

neuroscience↗

bioRxiv2024

Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

The human brain undergoes remarkable development during the first six postnatal months, a period of dramatic structural and functional change critical for understanding neurodevelopmental trajectories. Previous infant resting-state functional MRI (rsfMRI) studies identified intrinsic connectivity networks (ICNs) but reported widely varying network numbers and organization, limiting cross-study comparisons. A recent analysis of over 100,000 subjects spanning from adolescents to adults generated a 105-network multi-scale template that greatly enhanced replicability, but the presence of these canonical ICNs in infants has not been investigated. We analysed resting-state scans from infants aged 0-6 months using two complementary approaches: burst independent component analysis (burstICA), a fully data-driven, model-order-agnostic method, and the reference-informed NeuroMark framework. burstICA successfully recovered the full set of canonical ICNs directly from infant data, and simulation-based validation confirmed that the recovered networks were biologically specific rather than methodological biases. The NeuroMark framework enhanced spatial correspondence with the reference template and maintained distinct, reproducible network profiles across scans, demonstrating high reliability for cross-study and cross-age comparisons. Networks critical for higher-order cognition, such as the default mode and salience networks, displayed adult-like topography comparable to that of primary sensory networks even within the first 6 months after birth. Subcortical networks also exhibited precise spatial organization, underscoring early maturation of deep brain structures. Functional network connectivity analyses revealed close similarity to adult profiles and clear antagonistic patterns between sensory systems and higher-order networks, indicating that foundational aspects of mature brain organization are already emerging in early infancy. Together, these findings establish a methodological foundation for early network mapping in infancy and lay critical groundwork for longitudinal and translational studies of brain development.

neuroscience↗

bioRxiv2024

Spontaneous Brain Dynamics Associated With Acceleration Of Longterm Functional Connectome In Postnatal Development

The first six postnatal months are a critical period for brain development, marked by rapid changes in functional neural circuits. However, long-term changes in neonatal functional connectome lacks an interpretive imaging indicator for the future development due to the non-linearity characteristics. In this study, we introduce an approach to extract intrinsic brain states from short-term brain dynamics to study the long-term (longitudinal) development. We found a high association (r=0.460) between the co-activated pattern of specific brain state and the acceleration pattern of non-linear development of static functional connectome. The fractional occupancy, self-sustaining probability of this short-term state share the similar age tendency with the long-term change rate within the majority of the function connectome. These findings suggest that short-term brain dynamics could serve as potential biomarkers for predicting the long-term development of functional connectome.

neuroscience↗

bioRxiv2024

Normalized Dynamic Time Warping Increases Sensitivity In Differentiating Functional Network Connectivity In Schizophrenia

Our study advances the application of dynamic time warping (DTW) as a functional connectivity measure by introducing a normalization technique which enhances the detection of schizophrenia effects in comparison to both standard DTW and traditional correlation methods. By rigorously examining the statistical validity of DTW and our proposed normalized DTW measure, we show that it effectively captures interdependencies between fMRI signals beyond linear correlation, offering a more robust, complementary and informative approach to functional connectivity analysis. Through comprehensive evaluations, we demonstrate that normalized DTW is more sensitive to differences in functional brain network connections between schizophrenia and controls, highlighting its potential to provide deeper insight into clinical research. Clinical RelevanceThis study enhances our understanding of the functional specificity of schizophrenia by emphasizing the importance of nonlinear relationships through the introduction of a normalization technique for the DTW metric.

neuroscience↗

bioRxiv2024

Spatial Development of Brain Networks During The First Six Postnatal Months

The initial months of life constitute a crucial period for human development. A comprehensive understanding of this early phase is essential for unraveling the origins of neurodevelopmental disorders and promoting infant brain health. This study uniquely focuses on the spatial development of intrinsic brain connectivity networks during infancy, which has been less explored compared to functional connectivity. We utilized independent component analysis on resting-state fMRI data from 74 infants to assess how the spatial organization of infant brain networks evolves between birth and six months. Our findings reveal significant changes in spatial characteristics, including an a notable rise in the network-averaged spatial similarity across age, reflecting how closely each participant-specific spatial map aligns with the group-level map for each network. We also observed a marked reduction in the network engagement range by age, representing the extent of voxel intensity range fluctuation within each network. This suggests a continuing process of consolidation, where voxel contributions to the network become more uniform, as indicated by the narrowing of intensity values. The network strength, calculated as the average of all the voxel intensities in the network, indicating the degree of involvement to the specific functional network, increased across age in several networks, such as frontal-mPFC, primary, and secondary visual networks. The network size, along with the network center of mass, illustrating spatial distribution alterations of brain networks by age, varied across different networks. For instance, both metrics increased across age in the secondary visual network but decreased in the temporal network. Additionally, we examined the networks in relation to their linear versus non-linear developmental trajectories across all spatial characteristics, providing a deeper understanding of how these patterns evolve during early infancy. These findings contribute to early brain development understanding and offer insights into potential markers of consolidation and spatial reorganization in large-scale brain networks during infancy.

neuroscience↗

bioRxiv2024

Addressing inconsistency in functional neuroimaging: A replicable data-driven multi-scale functional atlas for canonical brain networks

The advent of multiple neuroimaging methodologies has greatly aided in the conceptualization of large-scale functional brain networks in the field of cognitive neuroscience. However, there is inconsistency across studies in both nomenclature and the functional entities being described. There is a need for a unifying framework that standardizes terminology across studies while also bringing analyses and results into the same reference space. Here we present a whole-brain atlas of canonical functional brain networks derived from more than 100,000 resting-state fMRI datasets. These data-driven functional networks are highly replicable across datasets and capture information from multiple spatial scales. We have organized, labeled, and described the networks with terms familiar to the fields of cognitive and affective neuroscience in order to optimize their utility in future neuroimaging analyses and enhance the accessibility of new findings. The benefits of this atlas are not limited to future template-based or reference-guided analyses, but also extend to other data-driven neuroimaging approaches across modalities, such as those using blind independent component analysis (ICA). Future studies utilizing this atlas will contribute to greater harmonization and standardization in functional neuroimaging research.

neuroscience↗

bioRxiv2024

Unraveling the Neural Landscape of Mental Disorders using Double Functional Independent Primitives (dFIPs)

Mental illnesses extract a high personal and societal cost, and thus explorations of the links between mental illness and functional connectivity in the brain are critical. Investigating major mental illnesses, believed to arise from disruptions in sophisticated neural connections, allows us to comprehend how these neural network disruptions may be linked to altered cognition, emotional regulation, and social interactions. Although neuroimaging has opened new avenues to explore neural alterations linked to mental illnesses, the field still requires precise and sensitive methodologies to inspect these neural substrates of various psychological disorders. In this study, we employ a hierarchical methodology to derive double functionally independent primitives (dFIPs) from resting state functional magnetic resonance neuroimaging data (rs-fMRI). These dFIPs encapsulate canonical overlapping patterns of functional network connectivity (FNC) within the brain. Our investigation focuses on the examination of how combinations of these dFIPs relate to different mental disorder diagnoses. The central aim is to unravel the complex patterns of FNC that correspond to the diverse manifestations of mental illnesses. To achieve this objective, we used a large brain imaging dataset from multiple sites, comprising 5805 total individuals diagnosed with schizophrenia (SCZ), autism spectrum disorder (ASD), bipolar disorder (BPD), major depressive disorder (MDD), and controls. The key revelations of our study unveil distinct patterns associated with each mental disorder through the combination of dFIPs. Notably, certain individual dFIPs exhibit disorder-specific characteristics, while others demonstrate commonalities across disorders. This approach offers a novel, data-driven synthesis of intricate neuroimaging data, thereby illuminating the functional changes intertwined with various mental illnesses. Our results show distinct signatures associated with psychiatric disorders, revealing unique connectivity patterns such as heightened cerebellar connectivity in SCZ and sensory domain hyperconnectivity in ASD, both contrasted with reduced cerebellar-subcortical connectivity. Utilizing the dFIP concept, we pinpoint specific functional connections that differentiate healthy controls from individuals with mental illness, underscoring its utility in identifying neurobiological markers. In summary, our findings delineate how dFIPs serve as unique fingerprints for different mental disorders.

neuroscience↗

bioRxiv2024

SpaDE: Semantic Locality Preserving Biclusteringfor Neuroimaging Data

The most discriminative and revealing patterns in the neuroimaging population are often confined to smaller subdivisions of the samples and features. Especially in neuropsychiatric conditions, symptoms are expressed within micro subgroups of individuals and may only underly a subset of neurological mechanisms. As such, running a whole-population analysis yields suboptimal outcomes leading to reduced specificity and interpretability. Biclustering is a potential solution since subject heterogeneity makes one-dimensional clustering less effective in this realm. Yet, high dimensional sparse input space and semantically incoherent grouping of attributes make post hoc analysis challenging. Therefore, we propose a deep neural network called semantic locality preserving auto decoder (SpaDE), for unsupervised feature learning and biclustering. SpaDE produces coherent subgroups of subjects and neural features preserving semantic locality and enhancing neurobiological interpretability. Also, it regularizes for sparsity to improve representation learning. We employ SpaDE on human brain connectome collected from schizophrenia (SZ) and healthy control (HC) subjects. The model outperforms several state-of-the-art biclustering methods. Our method extracts modular neural communities showing significant (HC/SZ) group differences in distinct brain networks including visual, sensorimotor, and subcortical. Moreover, these biclustered connectivity substructures exhibit substantial relations with various cognitive measures such as attention, working memory, and visual learning.

neuroscience↗

bioRxiv2024

Multi-modal Spatial-modality Attentive Fusion for Studying Neurodegenerative Disorders

Multi-modal learning has emerged as a powerful technique that leverages diverse data sources to enhance learning and decision-making processes. Adapting this approach to analyzing data collected from different biological domains is intuitive, especially for studying neuropsychiatric disorders. A complex neuropsychiatric disorder like schizophrenia (SZ) can affect multiple aspects of the brain and biologies. These biological sources each present distinct yet correlated expressions of subjects underlying physiological processes. Joint learning from these data sources can improve our understanding of the disorder. However, combining these biological sources is challenging for several reasons: (i) observations are domains-specific, leading to data being represented in dissimilar subspaces, and (ii) fused data is often noisy and high-dimensional, making it challenging to identify relevant information. To address these challenges, we propose a multi-modal artificial intelligence (AI) model with a novel fusion module inspired by a bottleneck attention module (BAM). We use deep neural networks (DNN) to learn latent space representations of the input streams. Next, we introduce a two-dimensional (spatio-modality) attention module to regulate the intermediate fusion for SZ classification. We implement spatial attention via a dilated convolutional neural network that creates large receptive fields for extracting significant contextual patterns. The resulting joint learning framework maximizes complementarity allowing us to explore the correspondence among the modalities. We test our model on a multi-modal imaging-genetic dataset and achieve an SZ prediction accuracy of 94.10% (P < 0.0001), outperforming state-of-the-art unimodal and multi-modal models for the task. Moreover, the model provides inherent interpretability that helps identify concepts significant for the neural networks decision and explains the underlying physiopathology of the disorder. Results also show that functional connectivity among subcortical, sensorimotor, and cognitive control domains plays an important role in characterizing SZ. Analysis of the spatio-modality attention scores suggests that structural components like the supplementary motor area, caudate, and insula play a significant role in SZ. Biclustering the attention scores discover a multi-modal cluster that includes genes CSMD1, ATK3, MOB4, and HSPE1, all of which have been identified as relevant to schizophrenia. In summary, feature attribution appears to be especially useful for probing the transient and confined but decisive patterns of complex disorders, and it shows promise for extensive applicability in future studies.

neuroscience↗