Modeling and interpreting mesoscale network dynamics

Disrupted Dynamic Network Attribution Associated with Gait Disorder in Cerebral Small Vessel Disease

Background and Aims: Previous research has focused on static functional connectivity in gait disorders caused by cerebral small vessel disease (CSVD), neglecting dynamic functional connections and network attribution. This study aims to investigate alterations in dynamic functional network connectivity (dFNC) and topological organization variance in CSVD-related gait disorders. Methods: A total of 85 patients with CSVD, including 41 patients with CSVD and gait disorders (CSVD-GD), 44 patients with CSVD and non-gait disorders (CSVD-NGD), and 32 healthy controls (HC), were enrolled in this study. Five networks composed of 10 independent components were selected using independent component analysis. Sliding time window and k-means clustering methods were used for dFNC analysis. The relationship between alterations in the dFNC properties and gait metrics was further assessed. Results: Three reproducible dFNC states were determined (State 1: sparsely connected, State 2: intermediate pattern, and State 3: strongly connected). CSVD-GD showed significantly higher fractional windows (FW) and mean dwell time (MDT) in State 1 compared with CSVD-NGD. Higher local efficiency variance was observed in the CSVD-GD group compared with HC, but no differences were found in the global efficiency comparison. Both the FW and MDT in State 1 were negatively correlated with gait speed and step length, and the relationship between MDT of State 1 and gait speed was mediated by overall cognition, information processing speed, and executive function. Conclusions: Our study uncovered abnormal dFNC indicators and variations in topological organization in CSVD-GD, offering potential early prediction indicators and freshening insights into the underlying pathogenesis of gait disturbances in CSVD.

Machine Learning and Graph Signal Processing Applied to Healthcare: A Review

Signal processing is a very useful field of study in the interpretation of signals in many everyday applications. In the case of applications with time-varying signals, one possibility is to consider them as graphs, so graph theory arises, which extends classical methods to the non-Euclidean domain. In addition, machine learning techniques have been widely used in pattern recognition activities in a wide variety of tasks, including health sciences. The objective of this work is to identify and analyze the papers in the literature that address the use of machine learning applied to graph signal processing in health sciences. A search was performed in four databases (Science Direct, IEEE Xplore, ACM, and MDPI), using search strings to identify papers that are in the scope of this review. Finally, 45 papers were included in the analysis, the first being published in 2015, which indicates an emerging area. Among the gaps found, we can mention the need for better clinical interpretability of the results obtained in the papers, that is not to restrict the results or conclusions simply to performance metrics. In addition, a possible research direction is the use of new transforms. It is also important to make new public datasets available that can be used to train the models.

Development of neonatal connectome dynamics and its prediction for cognitive and language outcomes at age 2

The functional brain connectome is highly dynamic over time. However, how brain connectome dynamics evolves during the third trimester of pregnancy and is associated with later cognitive growth remains unknown. Here, we use resting-state functional Magnetic Resonance Imaging (MRI) data from 39 newborns aged 32 to 42 postmenstrual weeks to investigate the maturation process of connectome dynamics and its role in predicting neurocognitive outcomes at 2 years of age. Neonatal brain dynamics is assessed using a multilayer network model. Network dynamics decreases globally but increases in both modularity and diversity with development. Regionally, module switching decreases with development primarily in the lateral precentral gyrus, medial temporal lobe, and subcortical areas, with a higher growth rate in primary regions than in association regions. Support vector regression reveals that neonatal connectome dynamics is predictive of individual cognitive and language abilities at 2  years of age. Our findings highlight network-level neural substrates underlying early cognitive development.

The emergence of multiscale connectomics-based approaches in stroke recovery

Stroke is a leading cause of adult disability. Understanding stroke damage and recovery requires deciphering changes in complex brain networks across different spatiotemporal scales. While recent developments in brain readout technologies and progress in complex network modeling have revolutionized current understanding of the effects of stroke on brain networks at a macroscale, reorganization of smaller scale brain networks remains incompletely understood. In this review, we use a conceptual framework of graph theory to define brain networks from nano- to macroscales. Highlighting stroke-related brain connectivity studies at multiple scales, we argue that multiscale connectomics-based approaches may provide new routes to better evaluate brain structural and functional remapping after stroke and during recovery.

NBS-SNI, an extension of the network-based statistic: Abnormal functional connections between important structural actors

Elucidating the coupling between the structure and the function of the brain and its development across maturation has attracted a lot of interest in the field of network neuroscience in the last 15 years. Mounting evidence supports the hypothesis that the onset of certain brain disorders is linked with the interplay between the structural architecture of the brain and its functional processes, often accompanied with unusual connectivity features. This paper introduces a method called the network-based statistic-simultaneous node investigation (NBS-SNI) that integrates both representations into a single framework, and identifies connectivity abnormalities in case-control studies. With this method, significance is given to the properties of the nodes, as well as to their connections. This approach builds on the well-established network-based statistic (NBS) proposed in 2010. We uncover and identify the regimes in which NBS-SNI offers a gain in statistical resolution to identify a contrast of interest using synthetic data. We also apply our method on two real case-control studies, one consisting of individuals diagnosed with autism and the other consisting of individuals diagnosed with early psychosis. Using NBS-SNI and node properties such as the closeness centrality and local information dimension, we found hypo- and hyperconnected subnetworks and show that our method can offer a 9 percentage points gain in prediction power over the standard NBS.

Dynamic nonreversibility view of intrinsic brain organization and brain dynamic analysis of repetitive transcranial magnitude stimulation

Intrinsic neural activities are characterized as endless spontaneous fluctuation over multiple time scales. However, how the intrinsic brain organization changes over time under local perturbation remains an open question. By means of statistical physics, we proposed an approach to capture whole-brain dynamics based on estimating time-varying nonreversibility and k-means clustering of dynamic varying nonreversibility patterns. We first used synthetic fMRI to investigate the effects of window parameters on the temporal variability of varying nonreversibility. Second, using real test-retest fMRI data, we examined the reproducibility, reliability, biological, and physiological correlation of the varying nonreversibility substates. Finally, using repetitive transcranial magnetic stimulation-fMRI data, we investigated the modulation effects of repetitive transcranial magnetic stimulation on varying nonreversibility substate dynamics. The results show that: (i) as window length increased, the varying nonreversibility variance decreased, while the sliding step almost did not alter it; (ii) the global high varying nonreversibility states and low varying nonreversibility states were reproducible across multiple datasets and different window lengths; and (iii) there were increased low varying nonreversibility states and decreased high varying nonreversibility states when the left frontal lobe was stimulated, but not the occipital lobe. Taken together, these results provide a thermodynamic equilibrium perspective of intrinsic brain organization and reorganization under local perturbation.

Towards a biologically annotated brain connectome

The brain is a network of interleaved neural circuits. In modern connectomics, brain connectivity is typically encoded as a network of nodes and edges, abstracting away the rich biological detail of local neuronal populations. Yet biological annotations for network nodes - such as gene expression, cytoarchitecture, neurotransmitter receptors or intrinsic dynamics - can be readily measured and overlaid on network models. Here we review how connectomes can be represented and analysed as annotated networks. Annotated connectomes allow us to reconceptualize architectural features of networks and to relate the connection patterns of brain regions to their underlying biology. Emerging work demonstrates that annotated connectomes help to make more veridical models of brain network formation, neural dynamics and disease propagation. Finally, annotations can be used to infer entirely new inter-regional relationships and to construct new types of network that complement existing connectome representations. In summary, biologically annotated connectomes offer a compelling way to study neural wiring in concert with local biological features.

Heterogeneity of dynamic synergetic configurations of salience network in children with autism spectrum disorder

Atypical functional connectivity (FC) patterns have been identified in autism spectrum disorders (ASD), especially within salience network (SN) and between SN and default mode network (DMN) and central executive network (CEN). But whether the dynamic configuration of intra-SN and inter-SN (SN with DMN and CEN) FC in ASD is also heterogeneous remains unknown. Based on the resting-state functional magnetic resonance imaging data from 105 ASD and 102 typically-developing controls (TC), we calculated the time-varying FC of intra-SN and inter-SN (SN with DMN and CEN). Then, the joint recurrence features for the time-varying FC were calculated to assess how the SN dynamically recruits different configurations of network segregation and integration in ASD, that is, synergies, from the dynamical systems perspective. We analyzed the differences in synergetic patterns between ASD subtypes obtained by k-means clustering algorithm based on the synergy of SN and TC, and investigated the relationships between synergy of SN and severity of clinical symptoms of ASD for ASD subtypes. Two ASD subtypes were revealed, where the synergy of SN in ASD subtype 1 has lower stability and periodicity compared to the TC, and ASD subtype 2 exhibits the opposite alteration. Synergy of SN for ASD subtype 1 and 2 was found to predict the severity of communication impairments and restricted and repetitive behaviors in ASD, respectively. These results suggest the existence of subtypes with distinct patterns of the synergy of SN in ASD, and provide insight into the complex pathophysiological mechanism of clinical manifestations of ASD.

Connectivity-based Meta-Bands: A new approach for automatic frequency band identification in connectivity analyses

The majority of electroencephalographic (EEG) and magnetoencephalographic (MEG) studies filter and analyse neural signals in specific frequency ranges, known as "canonical" frequency bands. However, this segmentation, is not exempt from limitations, mainly due to the lack of adaptation to the neural idiosyncrasies of each individual. In this study, we introduce a new data-driven method to automatically identify frequency ranges based on the topological similarity of the frequency-dependent functional neural network. The resting-state neural activity of 195 cognitively healthy subjects from three different databases (MEG: 123 subjects; EEG1: 27 subjects; EEG2: 45 subjects) was analysed. In a first step, MEG and EEG signals were filtered with a narrow-band filter bank (1 Hz bandwidth) from 1 to 70 Hz with a 0.5 Hz step. Next, the connectivity in each of these filtered signals was estimated using the orthogonalized version of the amplitude envelope correlation to obtain the frequency-dependent functional neural network. Finally, a community detection algorithm was used to identify communities in the frequency domain showing a similar network topology. We have called this approach the "Connectivity-based Meta-Bands" (CMB) algorithm. Additionally, two types of synthetic signals were used to configure the hyper-parameters of the CMB algorithm. We observed that the classical approaches to band segmentation are partially aligned with the underlying network topologies at group level for the MEG signals, but they are missing individual idiosyncrasies that may be biasing previous studies, as revealed by our methodology. On the other hand, the sensitivity of EEG signals to reflect this underlying frequency-dependent network structure is limited, revealing a simpler frequency parcellation, not aligned with that defined by the "canonical" frequency bands. To the best of our knowledge, this is the first study that proposes an unsupervised band segmentation method based on the topological similarity of functional neural network across frequencies. This methodology fully accounts for subject-specific patterns, providing more robust and personalized analyses, and paving the way for new studies focused on exploring the frequency-dependent structure of brain connectivity.

Increased flexibility of brain dynamics in patients with multiple sclerosis

Patients with multiple sclerosis consistently show widespread changes in functional connectivity. Yet, alterations are heterogeneous across studies, underscoring the complexity of functional reorganization in multiple sclerosis. Here, we aim to provide new insights by applying a time-resolved graph-analytical framework to identify a clinically relevant pattern of dynamic functional connectivity reconfigurations in multiple sclerosis. Resting-state data from 75 patients with multiple sclerosis (N = 75, female:male ratio of 3:2, median age: 42.0 ± 11.0 years, median disease duration: 6 ± 11.4 years) and 75 age- and sex-matched controls (N = 75, female:male ratio of 3:2, median age: 40.2 ± 11.8 years) were analysed using multilayer community detection. Local, resting-state functional system and global levels of dynamic functional connectivity reconfiguration were characterized using graph-theoretical measures including flexibility, promiscuity, cohesion, disjointedness and entropy. Moreover, we quantified hypo- and hyper-flexibility of brain regions and derived the flexibility reorganization index as a summary measure of whole-brain reorganization. Lastly, we explored the relationship between clinical disability and altered functional dynamics. Significant increases in global flexibility (t = 2.38, P_FDR = 0.024), promiscuity (t = 1.94, P_FDR = 0.038), entropy (t = 2.17, P_FDR = 0.027) and cohesion (t = 2.45, P_FDR = 0.024) were observed in patients and were driven by pericentral, limbic and subcortical regions. Importantly, these graph metrics were correlated with clinical disability such that greater reconfiguration dynamics tracked greater disability. Moreover, patients demonstrate a systematic shift in flexibility from sensorimotor areas to transmodal areas, with the most pronounced increases located in regions with generally low dynamics in controls. Together, these findings reveal a hyperflexible reorganization of brain activity in multiple sclerosis that clusters in pericentral, subcortical and limbic areas. This functional reorganization was linked to clinical disability, providing new evidence that alterations of multilayer temporal dynamics play a role in the manifestation of multiple sclerosis.

Supervising the Decoder of Variational Autoencoders to Improve Scientific Utility

Probabilistic generative models are attractive for scientific modeling because their inferred parameters can be used to generate hypotheses and design experiments. This requires that the learned model provides an accurate representation of the input data and yields a latent space that effectively predicts outcomes relevant to the scientific question. Supervised Variational Autoencoders (SVAEs) have previously been used for this purpose, as a carefully designed decoder can be used as an interpretable generative model of the data, while the supervised objective ensures a predictive latent representation. Unfortunately, the supervised objective forces the encoder to learn a biased approximation to the generative posterior distribution, which renders the generative parameters unreliable when used in scientific models. This issue has remained undetected as reconstruction losses commonly used to evaluate model performance do not detect bias in the encoder. We address this previously-unreported issue by developing a second-order supervision framework (SOS-VAE) that updates the decoder parameters, rather than the encoder, to induce a predictive latent representation. This ensures that the encoder maintains a reliable posterior approximation and the decoder parameters can be effectively interpreted. We extend this technique to allow the user to trade-off the bias in the generative parameters for improved predictive performance, acting as an intermediate option between SVAEs and our new SOS-VAE. We also use this methodology to address missing data issues that often arise when combining recordings from multiple scientific experiments. We demonstrate the effectiveness of these developments using synthetic data and electrophysiological recordings with an emphasis on how our learned representations can be used to design scientific experiments.

Multi-modal and multi-subject modular organization of human brain networks

The human brain is a complex network of anatomically interconnected brain areas. Spontaneous neural activity is constrained by this architecture, giving rise to patterns of statistical dependencies between the activity of remote neural elements. The non-trivial relationship between structural and functional connectivity poses many unsolved challenges about cognition, disease, development, learning and aging. While numerous studies have focused on statistical relationships between edge weights in anatomical and functional networks, less is known about dependencies between their modules and communities. In this work, we investigate and characterize the relationship between anatomical and functional modular organization of the human brain, developing a novel multi-layer framework that expands the classical concept of multi-layer modularity. By simultaneously mapping anatomical and functional networks estimated from different subjects into communities, this approach allows us to carry out a multi-subject and multi-modal analysis of the brain's modular organization. Here, we investigate the relationship between anatomical and functional modules during resting state, finding unique and shared structures. The proposed framework constitutes a methodological advance in the context of multi-layer network analysis and paves the way to further investigate the relationship between structural and functional network organization in clinical cohorts, during cognitively demanding tasks, and in developmental or lifespan studies.

Theoretical foundations of studying criticality in the brain

Criticality is hypothesized as a physical mechanism underlying efficient transitions between cortical states and remarkable information-processing capacities in the brain. While considerable evidence generally supports this hypothesis, nonnegligible controversies persist regarding the ubiquity of criticality in neural dynamics and its role in information processing. Validity issues frequently arise during identifying potential brain criticality from empirical data. Moreover, the functional benefits implied by brain criticality are frequently misconceived or unduly generalized. These problems stem from the nontriviality and immaturity of the physical theories that analytically derive brain criticality and the statistic techniques that estimate brain criticality from empirical data. To help solve these problems, we present a systematic review and reformulate the foundations of studying brain criticality, that is, ordinary criticality (OC), quasi-criticality (qC), self-organized criticality (SOC), and self-organized quasi-criticality (SOqC), using the terminology of neuroscience. We offer accessible explanations of the physical theories and statistical techniques of brain criticality, providing step-by-step derivations to characterize neural dynamics as a physical system with avalanches. We summarize error-prone details and existing limitations in brain criticality analysis and suggest possible solutions. Moreover, we present a forward-looking perspective on how optimizing the foundations of studying brain criticality can deepen our understanding of various neuroscience questions.

Intra- and inter-regional dynamics in cortical-striatal-tegmental networks

It is increasingly recognized that networks of brain areas work together to accomplish computational goals. However, functional connectivity networks are not often compared between different behavioral states and across different frequencies of electrical oscillatory signals. In addition, connectivity is always defined as the strength of signal relatedness between two atlas-based anatomical locations. Here, we performed an exploratory analysis using data collectected from high density arrays in the prefrontal cortex (PFC), striatum (STR), and ventral tegmental area (VTA) of male rats. These areas have all been implicated in a wide range of different tasks and computations including various types of memory as well as reward valuation, habit formation and execution, and skill learning. Novel intra-regional clustering analyses identified patterns of spatially restricted, temporally coherent, and frequency specific signals that were reproducible across days and were modulated by behavioral states. Multiple clusters were identified within each anatomical region, indicating a mesoscopic scale of organization. Generalized eigendecomposition (GED) was used to dimension-reduce each cluster to a single component time series. Dense inter-cluster connectivity was modulated by behavioral state, with connectivity becoming reduced when the animals were exposed to a novel object, compared to a baseline condition. Behavior-modulated connectivity changes were seen across the spectrum, with delta, theta, and gamma all being modulated. These results demonstrate the brain's ability to reorganize functionally at both the intra- and inter-regional levels during different behavioral states.

Alterations in Connectome Dynamics in Autism Spectrum Disorder: A Harmonized Mega- and Meta-analysis Study Using the Autism Brain Imaging Data Exchange Dataset

BACKGROUND

Neuroimaging studies have reported functional connectome aberrancies in autism spectrum disorder (ASD). However, the time-varying patterns of connectome topology in individuals with ASD and the connection between these patterns and gene expression profiles remain unknown.

METHODS

To investigate case-control differences in dynamic connectome topology, we conducted mega- and meta-analyses of resting-state functional magnetic resonance imaging data of 939 participants (440 patients with ASD and 499 healthy control subjects, all males) from 18 independent sites, selected from the Autism Brain Imaging Data Exchange (ABIDE) dataset. Functional data were preprocessed and analyzed using harmonized protocols, and brain module dynamics was assessed using a multilayer network model. We further leveraged postmortem brain-wide gene expression data to identify transcriptomic signatures associated with ASD-related alterations in brain dynamics.

RESULTS

Compared with healthy control participants, individuals with ASD exhibited a higher global mean and lower standard deviation of whole-brain module dynamics, indicating an unstable and less regionally differentiated pattern. More specifically, individuals with ASD showed higher module switching, primarily in the medial prefrontal cortex, posterior cingulate gyrus, and angular gyrus, and lower switching in the visual regions. These alterations in brain dynamics were predictive of social impairments in individuals with ASD and were linked with expression profiles of genes primarily involved in the regulation of neurotransmitter transport and secretion as well as with previously identified autism-related genes.

CONCLUSIONS

This study is the first to identify consistent alterations in brain network dynamics in ASD and the transcriptomic signatures related to those alterations, furthering insights into the biological basis behind this disorder.

NeuMapper: A scalable computational framework for multiscale exploration of the brain's dynamical organization

For better translational outcomes, researchers and clinicians alike demand novel tools to distill complex neuroimaging data into simple yet behaviorally relevant representations at the single-participant level. Recently, the Mapper approach from topological data analysis (TDA) has been successfully applied on noninvasive human neuroimaging data to characterize the entire dynamical landscape of whole-brain configurations at the individual level without requiring any spatiotemporal averaging at the outset. Despite promising results, initial applications of Mapper to neuroimaging data were constrained by (1) the need for dimensionality reduction and (2) lack of a biologically grounded heuristic for efficiently exploring the vast parameter space. Here, we present a novel computational framework for Mapper-designed specifically for neuroimaging data-that removes limitations and reduces computational costs associated with dimensionality reduction and parameter exploration. We also introduce new meta-analytic approaches to better anchor Mapper-generated representations to neuroanatomy and behavior. Our new NeuMapper framework was developed and validated using multiple fMRI datasets where participants engaged in continuous multitask experiments that mimic "ongoing" cognition. Looking forward, we hope our framework will help researchers push the boundaries of psychiatric neuroimaging toward generating insights at the single-participant level across consortium-size datasets.

Multiframe Evolving Dynamic Functional Connectivity (EVOdFNC): A Method for Constructing and Investigating Functional Brain Motifs

The study of brain network connectivity as a time-varying property began relatively recently and, to date, has remained primarily concerned with capturing a handful of discrete static states that characterize connectivity as measured on a timescale shorter than that of the full scan. Capturing group-level representations of temporally evolving patterns of connectivity is a challenging and important next step in fully leveraging the information available in large resting state functional magnetic resonance imaging (rs-fMRI) studies. We introduce a flexible, extensible data-driven framework for the stable identification of group-level multiframe (movie-style) dynamic functional network connectivity (dFNC) states. Our approach employs uniform manifold approximation and embedding (UMAP) to produce a continuity-preserving planar embedding of high-dimensional time-varying measurements of whole-brain functional network connectivity. Planar linear exemplars summarizing dominant dynamic trends across the population are computed from local linear approximations to the two-dimensional 2D embedded trajectories. A high-dimensional representation of each 2D exemplar segment is obtained by averaging the dFNC observations corresponding to the n planar nearest neighbors of τ evenly spaced points along the 2D line segment representation (where n is the UMAP number-of-neighbors parameter and τ is the temporal duration of trajectory segments being approximated). Each of the 2D exemplars thus “lifts” to a multiframe high-dimensional dFNC trajectory of length τ. The collection of high-dimensional temporally evolving dFNC representations (EVOdFNCs) derived in this manner are employed as dynamic basis objects with which to characterize observed high-dimensional dFNC trajectories, which are then expressed as weighted combination of these basis objects. Our approach yields new insights into anomalous patterns of fluidly varying whole-brain connectivity that are significantly associated with schizophrenia as a broad diagnosis as well as with certain symptoms of this serious disorder. Importantly, we show that relative to conventional hidden Markov modeling with single-frame unvarying dFNC summary states, EVOdFNCs are more sensitive to positive symptoms of schizophrenia including hallucinations and delusions, suggesting that a more dynamic characterization is needed to help illuminate such a complex brain disorder.

Schizophrenia induces abnormal frequency-dependent patterns of dynamic brain network reconfiguration during an auditory oddball task

OBJECTIVE

Schizophrenia is a psychiatric disorder that has been shown to disturb the dynamic top-down processing of sensory information. Various imaging techniques have revealed abnormalities in brain activity associated with this disorder, both locally and between cerebral regions. However, there is increasing interest in investigating dynamic network response to novel and relevant events at the network level during an attention-demanding task with high-temporal-resolution techniques. The aim of the work was: (i) to test the capacity of a novel algorithm to detect recurrent brain meta-states from auditory oddball task recordings; and (ii) to evaluate how the dynamic activation and behavior of the aforementioned meta-states were altered in schizophrenia, since it has been shown to impair top-down processing of sensory information.

APPROACH

A novel unsupervised method for the detection of brain meta-states based on recurrence plots and community detection algorithms, previously tested on resting-state data, was used on auditory oddball task recordings. Brain meta-states and several properties related to their activation during target trials in the task were extracted from electroencephalography (EEG) data from patients with schizophrenia and cognitively healthy controls.

MAIN RESULTS

The methodology successfully detected meta-states during an auditory oddball task, and they appeared to show both frequency-dependent time-locked and non-time-locked activity with respect to the stimulus onset. Moreover, patients with schizophrenia displayed higher network diversity, and showed more sluggish meta-state transitions, reflected in increased dwell times, less complex meta-state sequences, decreased meta-state space speed, and abnormal ratio of negative meta-state correlations.

SIGNIFICANCE

Abnormal cognition in schizophrenia is also reflected in decreased brain flexibility at the dynamic network level, which may hamper top-down processing, possibly indicating impaired decision-making linked to dysfunctional predictive coding. Moreover, the results showed the ability of the methodology to find meaningful and task-relevant changes in dynamic connectivity and pathology-related group differences.

Network controllability mediates the relationship between rigid structure and flexible dynamics

Precisely how the anatomical structure of the brain supports a wide range of complex functions remains a question of marked importance in both basic and clinical neuroscience. Progress has been hampered by the lack of theoretical frameworks explaining how a structural network of relatively rigid interareal connections can produce a diverse repertoire of functional neural dynamics. Here, we address this gap by positing that the brain's structural network architecture determines the set of accessible functional connectivity patterns according to predictions of network control theory. In a large developmental cohort of 823 youths aged 8 to 23 years, we found that the flexibility of a brain region's functional connectivity was positively correlated with the proportion of its structural links extending to different cognitive systems. Notably, this relationship was mediated by nodes' boundary controllability, suggesting that a region's strategic location on the boundaries of modules may underpin the capacity to integrate information across different cognitive processes. Broadly, our study provides a mechanistic framework that illustrates how temporal flexibility observed in functional networks may be mediated by the controllability of the underlying structural connectivity.

Altered temporal reachability highlights the role of sensory perception systems in major depressive disorder

BACKGROUND

The latest studies have considered the time-dependent structures in dynamic brain networks. However, the effect of periphery structures on the temporal flow of information remains unexplored in patients with major depressive disorder (MDD). In this work, we aimed to explore the pattern of interactions between brain regions in MDD across space and time.

METHODS

We concentrated on the temporal reachability of nodes in temporal brain networks derived from the resting-state functional magnetic resonance imaging (rs-fMRI) of 55 MDD patients and 62 sex-, age-matched healthy controls. Specifically, temporal connectedness and temporal efficiency (TEF) were estimated based on the length of temporal paths between node pairs. Subsequently, the temporal clustering coefficient (TCC) and temporal distance were jointly employed to explore the patterns in which a node's periphery structure affects its reachability.

RESULTS

Significantly higher TEF and lower TCC were found in temporal brain networks in MDD. Besides, significant between-group differences of nodal TCC were detected in regions of sensory perception systems. Considering the temporal paths that begin or end at these regions, MDD patients showed several altered temporal distances.

CONCLUSION

Our results showed that the temporal reachability of specific brain regions in MDD could be affected as their periphery structures evolve, which may explain the dysfunction of sensory perception systems in the spatiotemporal domain.

Functional Complex Networks Based on Operational Architectonics: Application on Electroencephalography-Brain-computer Interface for Imagined Speech

A new method for analyzing brain complex dynamics and states is presented. This method constructs functional brain graphs and is comprised of two pylons: (a) Operational architectonics (OA) concept of brain and mind functioning. (b) Network neuroscience. In particular, the algorithm utilizes OA framework for a non-parametric segmentation of EEGs, which leads to the identification of change points, namely abrupt jumps in EEG amplitude, called Rapid Transition Processes (RTPs). Subsequently, the time coordinates of RTPs are used for the generation of undirected weighted complex networks fulfilling a scale-free topology criterion, from which various network metrics of brain connectivity are estimated. These metrics form feature vectors, which can be used in machine learning algorithms for classification and/or prediction. The method is tested in classification problems on an EEG-based BCI data set, acquired from individuals during imagery pronunciation tasks of various words/vowels. The classification results, based on a Naïve Bayes classifier, show that the overall accuracies were found to be above chance level in all tested cases. This method was also compared with other state-of-the-art computational approaches commonly used for functional network generation, exhibiting competitive performance. The method can be useful to neuroscientists wishing to enhance their repository of brain research algorithms.

Imaging the temporal dynamics of brain states with highly sampled fMRI

The spontaneous dynamics of the brain modulate its function from moment to moment, shaping neural computation and cognition. Functional MRI (fMRI), while classically used as a tool for spatial localization, is increasingly being used to identify the temporal dynamics of brain activity. fMRI analyses focused on the temporal domain have revealed important new information about the dynamics underlying states such as arousal, attention, and sleep. Dense temporal sampling – either by using fast fMRI acquisition, or multiple repeated scan sessions within individuals – can further enrich the information present in these studies. This review focuses on recent developments in using fMRI to identify dynamics across brain states, particularly vigilance and sleep states, and the potential for highly temporally sampled fMRI to answer these questions.

Altered Brain Functional Network in Subtypes of Parkinson's Disease: A Dynamic Perspective

Background: Parkinson's disease (PD) is a highly heterogeneous disease, especially in the clinical characteristics and prognosis. The PD is divided into two subgroups: tremor-dominant phenotype and non-tremor-dominant phenotype. Previous studies reported abnormal changes between the two PD phenotypes by using the static functional connectivity analysis. However, the dynamic properties of brain networks between the two PD phenotypes are not yet clear. Therefore, we aimed to uncover the dynamic functional network connectivity (dFNC) between the two PD phenotypes at the subnetwork level, focusing on the temporal properties of dFNC and the variability of network efficiency.

Methods: We investigated the resting-state functional MRI (fMRI) data from 29 tremor-dominant PD patients (PDTD), 25 non-tremor-dominant PD patients (PDNTD), and 20 healthy controls (HCs). Sliding window approach, k-means clustering, independent component analysis (ICA), and graph theory analysis were applied to analyze the dFNC. Furthermore, the relationship between alterations in the dynamic properties and clinical features was assessed.

Results: The dFNC analyses identified four reoccurring states, one of them showing sparse connections (state I). PDTD patients stayed longer time in state I and showed increased FNC between BG and vSMN in state IV. Both PD phenotypes exhibited higher FNC between dSMN and FPN in state II and state III compared with the controls. PDNTD patients showed decreased FNC between BG and FPN but increased FNC in the bilateral FPN compared with both PDTD patients and controls. In addition, PDNTD patients exhibited greater variability in global network efficiency. Tremor scores were positively correlated with dwell time in state I along with increased FNC between BG and vSMN in state IV.

Conclusions: This study explores the dFNC between the PDTD and PDNTD patients, which offers new evidence on the abnormal time-varying brain functional connectivity and their network destruction of the two PD phenotypes, and may help better understand the neural substrates underlying different types of PD.

The modulation of brain network integration and arousal during exploration

There is growing interest in how neuromodulators shape brain networks. Recent neuroimaging studies provide evidence that brainstem arousal systems, such as the locus coeruleus-norepinephrine system (LC-NE), influence functional connectivity and brain network topology, suggesting they have a role in flexibly reconfiguring brain networks in order to adapt behavior and cognition to environmental demands. To date, however, the relationship between brainstem arousal systems and functional connectivity has not been assessed within the context of a task with an established relationship between arousal and behavior, with most prior studies relying on incidental variations in arousal or pharmacological manipulation and static brain networks constructed over long periods of time. These factors have likely contributed to a heterogeneity of effects across studies. To address these issues, we took advantage of the association between LC-NE-linked arousal and exploration to probe the relationships between exploratory choice, arousal—as measured indirectly via pupil diameter—and brain network dynamics. Exploration in a bandit task was associated with a shift toward fewer, more weakly connected modules that were more segregated in terms of connectivity and topology but more integrated with respect to the diversity of cognitive systems represented in each module. Functional connectivity strength decreased, and changes in connectivity were correlated with changes in pupil diameter, in line with the hypothesis that brainstem arousal systems influence the dynamic reorganization of brain networks. More broadly, we argue that carefully aligning dynamic network analyses with task designs can increase the temporal resolution at which behaviorally- and cognitively-relevant modulations can be identified, and offer these results as a proof of concept of this approach.

Brain Connectivity Studies on Structure-Function Relationships: A Short Survey with an Emphasis on Machine Learning

This short survey reviews the recent literature on the relationship between the brain structure and its functional dynamics. Imaging techniques such as diffusion tensor imaging (DTI) make it possible to reconstruct axonal fiber tracks and describe the structural connectivity (SC) between brain regions. By measuring fluctuations in neuronal activity, functional magnetic resonance imaging (fMRI) provides insights into the dynamics within this structural network. One key for a better understanding of brain mechanisms is to investigate how these fast dynamics emerge on a relatively stable structural backbone. So far, computational simulations and methods from graph theory have been mainly used for modeling this relationship. Machine learning techniques have already been established in neuroimaging for identifying functionally independent brain networks and classifying pathological brain states. This survey focuses on methods from machine learning, which contribute to our understanding of functional interactions between brain regions and their relation to the underlying anatomical substrate.

Modeling dynamic characteristics of brain functional connectivity networks using resting-state functional MRI

Dynamic network analysis using resting-state functional magnetic resonance imaging (rs-fMRI) provides a great insight into fundamentally dynamic characteristics of human brains, thus providing an efficient solution to automated brain disease identification. Previous studies usually pay less attention to evolution of global network structures over time in each brain's rs-fMRI time series, and also treat network-based feature extraction and classifier training as two separate tasks. To address these issues, we propose a temporal dynamics learning (TDL) method for network-based brain disease identification using rs-fMRI time-series data, through which network feature extraction and classifier training are integrated into the unified framework. Specifically, we first partition rs-fMRI time series into a sequence of segments using overlapping sliding windows, and then construct longitudinally ordered functional connectivity networks. To model the global temporal evolution patterns of these successive networks, we introduce a group-fused Lasso regularizer in our TDL framework, while the specific network architecture is induced by an ℓ₁-norm regularizer. Besides, we develop an efficient optimization algorithm to solve the proposed objective function via the Alternating Direction Method of Multipliers (ADMM). Compared with previous studies, the proposed TDL model can not only explicitly model the evolving connectivity patterns of global networks over time, but also capture unique characteristics of each network defined at each segment. We evaluate our TDL on three real autism spectrum disorder (ASD) datasets with rs-fMRI data, achieving superior results in ASD identification compared with several state-of-the-art methods.

Time-Varying Functional Connectivity Decreases as a Function of Acute Nicotine Abstinence

BACKGROUND

The nicotine withdrawal syndrome (NWS) includes affective and cognitive disruptions whose incidence and severity vary across time during acute abstinence. However, most network-level neuroimaging uses static measures of resting-state functional connectivity and assumes time-invariance and is thus unable to capture dynamic brain-behavior relationships. Recent advances in resting-state functional connectivity signal processing allow characterization of time-varying functional connectivity (TVFC), which characterizes network communication between networks that reconfigure over the course of data collection. Therefore, TVFC may more fully describe network dysfunction related to the NWS.

METHODS

To isolate alterations in the frequency and diversity of communication across network boundaries during acute nicotine abstinence, we scanned 25 cigarette smokers in the nicotine-sated and abstinent states and applied a previously validated method to characterize TVFC at a network and a nodal level within the brain.

RESULTS

During abstinence, we found brain-wide decreases in the frequency of interactions between network nodes in different modular communities (i.e., temporal flexibility). In addition, within a subset of the networks examined, the variability of these interactions across community boundaries (i.e., spatiotemporal diversity) also decreased. Finally, within 2 of these networks, the decrease in spatiotemporal diversity was significantly related to NWS clinical symptoms.

CONCLUSIONS

Using multiple measures of TVFC in a within-subjects design, we characterized a novel set of changes in network communication and linked these changes to specific behavioral symptoms of the NWS. These reductions in TVFC provide a meso-scale network description of the relative inflexibility of specific large-scale brain networks during acute abstinence.

Abnormal meta-state activation of dynamic brain networks across the Alzheimer spectrum

The characterization of the distinct dynamic functional connectivity (dFC) patterns that activate in the brain during rest can help to understand the underlying time-varying network organization. The presence and behavior of these patterns (known as meta-states) have been widely studied by means of functional magnetic resonance imaging (fMRI). However, modalities with high-temporal resolution, such as electroencephalography (EEG), enable the characterization of fast temporally evolving meta-state sequences. Mild cognitive impairment (MCI) and dementia due to Alzheimer's disease (AD) have been shown to disrupt spatially localized activation and dFC between different brain regions, but not much is known about how they affect meta-state network topologies and their network dynamics. The main hypothesis of the study was that MCI and dementia due to AD alter normal meta-state sequences by inducing a loss of structure in their patterns and a reduction of their dynamics. Moreover, we expected that patients with MCI would display more flexible behavior compared to patients with dementia due to AD. Thus, the aim of the current study was twofold: (i) to find repeating, distinctly organized network patterns (meta-states) in neural activity; and (ii) to extract information about meta-state fluctuations and how they are influenced by MCI and dementia due to AD. To accomplish these goals, we present a novel methodology to characterize dynamic meta-states and their temporal fluctuations by capturing aspects based on both their discrete activation and the continuous evolution of their individual strength. These properties were extracted from 60-s resting-state EEG recordings from 67 patients with MCI due to AD, 50 patients with dementia due to AD, and 43 cognitively healthy controls. First, the instantaneous amplitude correlation (IAC) was used to estimate instantaneous functional connectivity with a high temporal resolution. We then extracted meta-states by means of graph community detection based on recurrence plots (RPs), both at the individual- and group-level. Subsequently, a diverse set of properties of the continuous and discrete fluctuation patterns of the meta-states was extracted and analyzed. The main novelty of the methodology lies in the usage of Louvain GJA community detection to extract meta-states from IAC-derived RPs and the extended analysis of their discrete and continuous activation. Our findings showed that distinct dynamic functional connectivity meta-states can be found on the EEG time-scale, and that these were not affected by the oscillatory slowing induced by MCI or dementia due to AD. However, both conditions displayed a loss of meta-state modularity, coupled with shorter dwell times and higher complexity of the meta-state sequences. Furthermore, we found evidence that meta-state sequencing is not entirely random; it shows an underlying structure that is partially lost in MCI and dementia due to AD. These results show evidence that AD progression is associated with alterations in meta-state switching, and a degradation of dynamic brain flexibility.

Dynamic community detection reveals transient reorganization of functional brain networks across a female menstrual cycle

Sex steroid hormones have been shown to alter regional brain activity, but the extent to which they modulate connectivity within and between large-scale functional brain networks over time has yet to be characterized. Here, we applied dynamic community detection techniques to data from a highly sampled female with 30 consecutive days of brain imaging and venipuncture measurements to characterize changes in resting-state community structure across the menstrual cycle. Four stable functional communities were identified, consisting of nodes from visual, default mode, frontal control, and somatomotor networks. Limbic, subcortical, and attention networks exhibited higher than expected levels of nodal flexibility, a hallmark of between-network integration and transient functional reorganization. The most striking reorganization occurred in a default mode subnetwork localized to regions of the prefrontal cortex, coincident with peaks in serum levels of estradiol, luteinizing hormone, and follicle stimulating hormone. Nodes from these regions exhibited strong intranetwork increases in functional connectivity, leading to a split in the stable default mode core community and the transient formation of a new functional community. Probing the spatiotemporal basis of human brain–hormone interactions with dynamic community detection suggests that hormonal changes during the menstrual cycle result in temporary, localized patterns of brain network reorganization.

Sex steroid hormones influence the central nervous system across multiple spatiotemporal scales. Estrogen and progesterone concentrations rise and fall throughout the menstrual cycle, but it remains poorly understood whether day-to-day fluctuations in hormones shape human brain dynamics. Here, we assessed the structure and stability of resting-state brain network connectivity in concordance with serum hormone levels from a female who underwent fMRI and venipuncture for 30 consecutive days. Our results reveal that while network structure is largely stable over the course of a menstrual cycle, temporary reorganization of several large-scale functional brain networks occurs during the ovulatory window. In particular, a default mode subnetwork exhibits increased connectivity with itself and with nodes belonging to the temporoparietal and limbic networks, providing novel perspective into brain-hormone interactions.

Detecting Dynamic Community Structure in Functional Brain Networks Across Individuals: A Multilayer Approach

OBJECTIVE

We present a unified statistical framework for characterizing community structure of brain functional networks that captures variation across individuals and evolution over time. Existing methods for community detection focus only on single-subject analysis of dynamic networks; while recent extensions to multiple-subjects analysis are limited to static networks.

METHOD

To overcome these limitations, we propose a multi-subject, Markov-switching stochastic block model (MSS-SBM) to identify state-related changes in brain community organization over a group of individuals. We first formulate a multilayer extension of SBM to describe the time-dependent, multi-subject brain networks. We develop a novel procedure for fitting the multilayer SBM that builds on multislice modularity maximization which can uncover a common community partition of all layers (subjects) simultaneously. By augmenting with a dynamic Markov switching process, our proposed method is able to capture a set of distinct, recurring temporal states with respect to inter-community interactions over subjects and the change points between them.

RESULTS

Simulation shows accurate community recovery and tracking of dynamic community regimes over multilayer networks by the MSS-SBM. Application to task fMRI reveals meaningful non-assortative brain community motifs, e.g., core-periphery structure at the group level, that are associated with language comprehension and motor functions suggesting their putative role in complex information integration. Our approach detected dynamic reconfiguration of modular connectivity elicited by varying task demands and identified unique profiles of intra and inter-community connectivity across different task conditions.

CONCLUSION

The proposed multilayer network representation provides a principled way of detecting synchronous, dynamic modularity in brain networks across subjects.

Semi-automatic Extraction of Functional Dynamic Networks Describing Patient's Epileptic Seizures

Intracranial electroencephalography (EEG) studies using stereotactic EEG (SEEG) have shown that during seizures, epileptic activity spreads across several anatomical regions from the seizure onset zone toward remote brain areas. A full and objective characterization of this patient-specific time-varying network is crucial for optimal surgical treatment. Functional connectivity (FC) analysis of SEEG signals recorded during seizures enables to describe the statistical relations between all pairs of recorded signals. However, extracting meaningful information from those large datasets is time consuming and requires high expertise. In the present study, we first propose a novel method named Brain-wide Time-varying Network Decomposition (BTND) to characterize the dynamic epileptogenic networks activated during seizures in individual patients recorded with SEEG electrodes. The method provides a number of pathological FC subgraphs with their temporal course of activation. The method can be applied to several seizures of the patient to extract reproducible subgraphs. Second, we compare the activated subgraphs obtained by the BTND method with visual interpretation of SEEG signals recorded in 27 seizures from nine different patients. As a whole, we found that activated subgraphs corresponded to brain regions involved during the course of the seizures and their time course was highly consistent with classical visual interpretation. We believe that the proposed method can complement the visual analysis of SEEG signals recorded during seizures by highlighting and characterizing the most significant parts of epileptic networks with their activation dynamics.

Topographic gradients of intrinsic dynamics across neocortex

The intrinsic dynamics of neuronal populations are shaped by both microscale attributes and macroscale connectome architecture. Here we comprehensively characterize the rich temporal patterns of neural activity throughout the human brain. Applying massive temporal feature extraction to regional haemodynamic activity, we systematically estimate over 6000 statistical properties of individual brain regions' time-series across the neocortex. We identify two robust spatial gradients of intrinsic dynamics, one spanning a ventromedial-dorsolateral axis and dominated by measures of signal autocorrelation, and the other spanning a unimodal-transmodal axis and dominated by measures of dynamic range. These gradients reflect spatial patterns of gene expression, intracortical myelin and cortical thickness, as well as structural and functional network embedding. Importantly, these gradients are correlated with patterns of meta-analytic functional activation, differentiating cognitive versus affective processing and sensory versus higher-order cognitive processing. Altogether, these findings demonstrate a link between microscale and macroscale architecture, intrinsic dynamics, and cognition.

Dynamic Brain Connectivity in Resting State Functional MR Imaging

Dynamic functional connectivity adds another dimension to resting-state functional MR imaging analysis. In recent years, dynamic functional connectivity has been increasingly used in resting-state functional MR imaging, and several studies have demonstrated that dynamic functional connectivity patterns correlate with different physiologic and pathologic brain states. In fact, evidence suggests that dynamic functional connectivity is a more sensitive marker than static functional connectivity; therefore, it might be a promising tool to add to clinical functional neuroimaging. This article provides a broad overview of dynamic functional connectivity and reviews its general principles, techniques, and potential clinical applications.

Mesoscale diffusion magnetic resonance imaging of the ex vivo human hippocampus

Mesoscale diffusion magnetic resonance imaging (MRI) endeavors to bridge the gap between macroscopic white matter tractography and microscopic studies investigating the cytoarchitecture of human brain tissue. To ensure a robust measurement of diffusion at the mesoscale, acquisition parameters were arrayed to investigate their effects on scalar indices (mean, radial, axial diffusivity, and fractional anisotropy) and streamlines (i.e., graphical representation of axonal tracts) in hippocampal layers. A mesoscale resolution afforded segementation of the pyramidal cell layer (CA1-4), the dentate gyrus, as well as stratum moleculare, radiatum, and oriens. Using ex vivo samples, surgically excised from patients with intractable epilepsy (n = 3), we found that shorter diffusion times (23.7 ms) with a b-value of 4,000 s/mm2 were advantageous at the mesoscale, providing a compromise between mean diffusivity and fractional anisotropy measurements. Spatial resolution and sample orientation exerted a major effect on tractography, whereas the number of diffusion gradient encoding directions minimally affected scalar indices and streamline density. A sample temperature of 15°C provided a compromise between increasing signal-to-noise ratio and increasing the diffusion properties of the tissue. Optimization of the acquisition afforded a system's view of intra- and extra-hippocampal connections. Tractography reflected histological boundaries of hippocampal layers. Individual layer connectivity was visualized, as well as streamlines emanating from individual sub-fields. The perforant path, subiculum and angular bundle demonstrated extra-hippocampal connections. Histology of the samples confirmed individual cell layers corresponding to ROIs defined on MR images. We anticipate that this ex vivo mesoscale imaging will yield novel insights into human hippocampal connectivity.

Signal diffusion along connectome gradients and inter-hub routing differentially contribute to dynamic human brain function

Human cognition is dynamic, alternating over time between externally-focused states and more abstract, often self-generated, patterns of thought. Although cognitive neuroscience has documented how networks anchor particular modes of brain function, mechanisms that describe transitions between distinct functional states remain poorly understood. Here, we examined how time-varying changes in brain function emerge within the constraints imposed by macroscale structural network organization. Studying a large cohort of healthy adults (n = 326), we capitalized on manifold learning techniques that identify low dimensional representations of structural connectome organization and we decomposed neurophysiological activity into distinct functional states and their transition patterns using Hidden Markov Models. Structural connectome organization predicted dynamic transitions anchored in sensorimotor systems and those between sensorimotor and transmodal states. Connectome topology analyses revealed that transitions involving sensorimotor states traversed short and intermediary distances and adhered strongly to communication mechanisms of network diffusion. Conversely, transitions between transmodal states involved spatially distributed hubs and increasingly engaged long-range routing. These findings establish that the structure of the cortex is optimized to allow neural states the freedom to vary between distinct modes of processing, and so provides a key insight into the neural mechanisms that give rise to the flexibility of human cognition.

Modeling time-varying brain networks with a self-tuning optimized Kalman filter

Brain networks are complex dynamical systems in which directed interactions between different areas evolve at the sub-second scale of sensory, cognitive and motor processes. Due to the highly non-stationary nature of neural signals and their unknown noise components, however, modeling dynamic brain networks has remained one of the major challenges in contemporary neuroscience. Here, we present a new algorithm based on an innovative formulation of the Kalman filter that is optimized for tracking rapidly evolving patterns of directed functional connectivity under unknown noise conditions. The Self-Tuning Optimized Kalman filter (STOK) is a novel adaptive filter that embeds a self-tuning memory decay and a recursive regularization to guarantee high network tracking accuracy, temporal precision and robustness to noise. To validate the proposed algorithm, we performed an extensive comparison against the classical Kalman filter, in both realistic surrogate networks and real electroencephalography (EEG) data. In both simulations and real data, we show that the STOK filter estimates time-frequency patterns of directed connectivity with significantly superior performance. The advantages of the STOK filter were even clearer in real EEG data, where the algorithm recovered latent structures of dynamic connectivity from epicranial EEG recordings in rats and human visual evoked potentials, in excellent agreement with known physiology. These results establish the STOK filter as a powerful tool for modeling dynamic network structures in biological systems, with the potential to yield new insights into the rapid evolution of network states from which brain functions emerge.

During normal behavior, brains transition between functional network states several times per second. This allows humans to quickly read a sentence, and a frog to catch a fly. Understanding these fast network dynamics is fundamental to understanding how brains work, but up to now it has proven very difficult to model fast brain dynamics for various methodological reasons. To overcome these difficulties, we designed a new Kalman filter (STOK) by innovating on previous solutions from control theory and state-space modelling. We show that STOK accurately models fast network changes in simulations and real neural data, making it an essential new tool for modelling fast brain networks in the time and frequency domain.

Learning in brain-computer interface control evidenced by joint decomposition of brain and behavior

OBJECTIVE

Motor imagery-based brain-computer interfaces (BCIs) use an individual's ability to volitionally modulate localized brain activity, often as a therapy for motor dysfunction or to probe causal relations between brain activity and behavior. However, many individuals cannot learn to successfully modulate their brain activity, greatly limiting the efficacy of BCI for therapy and for basic scientific inquiry. Formal experiments designed to probe the nature of BCI learning have offered initial evidence that coherent activity across spatially distributed and functionally diverse cognitive systems is a hallmark of individuals who can successfully learn to control the BCI. However, little is known about how these distributed networks interact through time to support learning.

APPROACH

Here, we address this gap in knowledge by constructing and applying a multimodal network approach to decipher brain-behavior relations in motor imagery-based brain-computer interface learning using magnetoencephalography. Specifically, we employ a minimally constrained matrix decomposition method - non-negative matrix factorization - to simultaneously identify regularized, covarying subgraphs of functional connectivity, to assess their similarity to task performance, and to detect their time-varying expression.

MAIN RESULTS

We find that learning is marked by diffuse brain-behavior relations: good learners displayed many subgraphs whose temporal expression tracked performance. Individuals also displayed marked variation in the spatial properties of subgraphs such as the connectivity between the frontal lobe and the rest of the brain, and in the temporal properties of subgraphs such as the stage of learning at which they reached maximum expression. From these observations, we posit a conceptual model in which certain subgraphs support learning by modulating brain activity in sensors near regions important for sustaining attention. To test this model, we use tools that stipulate regional dynamics on a networked system (network control theory), and find that good learners display a single subgraph whose temporal expression tracked performance and whose architecture supports easy modulation of sensors located near brain regions important for attention.

SIGNIFICANCE

The nature of our contribution to the neuroscience of BCI learning is therefore both computational and theoretical; we first use a minimally-constrained, individual specific method of identifying mesoscale structure in dynamic brain activity to show how global connectivity and interactions between distributed networks supports BCI learning, and then we use a formal network model of control to lend theoretical support to the hypothesis that these identified subgraphs are well suited to modulate attention.

Model-based whole-brain effective connectivity to study distributed cognition in health and disease

Neuroimaging techniques are now widely used to study human cognition. The functional associations between brain areas have become a standard proxy to describe how cognitive processes are distributed across the brain network. Among the many analysis tools available, dynamic models of brain activity have been developed to overcome the limitations of original connectivity measures such as functional connectivity. This goes in line with the many efforts devoted to the assessment of directional interactions between brain areas from the observed neuroimaging activity. This opinion article provides an overview of our model-based whole-brain effective connectivity to analyze fMRI data, while discussing the pros and cons of our approach with respect to other established approaches. Our framework relies on the multivariate Ornstein-Uhlenbeck (MOU) process and is thus referred to as MOU-EC. Once tuned, the model provides a directed connectivity estimate that reflects the dynamical state of BOLD activity, which can be used to explore cognition. We illustrate this approach using two applications on task-evoked fMRI data. First, as a connectivity measure, MOU-EC can be used to extract biomarkers for task-specific brain coordination, understood as the patterns of areas exchanging information. The multivariate nature of connectivity measures raises several challenges for whole-brain analysis, for which machine-learning tools present some advantages over statistical testing. Second, we show how to interpret changes in MOU-EC connections in a collective and model-based manner, bridging with network analysis. Our framework provides a comprehensive set of tools that open exciting perspectives to study distributed cognition, as well as neuropathologies.

Early childhood developmental functional connectivity of autistic brains with non-negative matrix factorization

Autism spectrum disorder (ASD) is associated with altered patterns of over- and under-connectivity of neural circuits. Age-related changes in neural connectivities remain unclear for autistic children as compared with normal children. In this study, a parts-based network-decomposition technique, known as non-negative matrix factorization (NMF), was applied to identify a set of possible abnormal connectivity patterns in brains affected by ASD, using resting-state electroencephalographic (EEG) data. Age-related changes in connectivities in both inter- and intra-hemispheric areas were studied in a total of 256 children (3-6 years old), both with and without ASD. The findings included the following: (1) the brains of children affected by ASD were characterized by a general trend toward long-range under-connectivity, particularly in interhemispheric connections, combined with short-range over-connectivity; (2) long-range connections were often associated with slower rhythms (δ and θ), whereas synchronization of short-range networks tended to be associated with faster frequencies (α and β); and (3) the α-band specific patterns of interhemispheric connections in ASD could be the most prominent during early childhood neurodevelopment. Therefore, NMF would be useful for further exploring the early childhood developmental functional connectivity of children aged 3-6 with ASD as well as with typical development. Additionally, long-range interhemispheric alterations in connectivity may represent a potential biomarker for the identification of ASD.

More Flexible Integration of Functional Systems After Musical Training in Young Adults

Musical training, because it involves the interaction and integration of diverse functional systems, is an excellent model to investigate training-induced brain plasticity. The human brain functions in a network architecture in which dynamic modules and subgraphs are considered to enable efficient information communication. However, it remains largely unknown how the dynamic integration of functional systems changes with musical training, which may provide new insight into musical training-induced brain plasticity and further the use of music therapy for neuropsychiatric disease and brain injury. Here, 29 healthy young adult novices who received 24 weeks of piano training, and another 27 novices without any intervention were scanned at three time points-before and after musical training and 12 weeks after training. We used nonnegative matrix factorization to identify a set of subgraphs and their corresponding time-dependent coefficients from a concatenated functional network of all the subjects in sliding time windows. The energy and entropy of the time-dependent coefficients were computed to quantify the subgraph's dynamic changes in expression. The training group showed a significantly increased energy of the time-dependent coefficients of 3 subgraphs after training. Furthermore, one of the subgraphs, comprised of primary functional systems and cingulo-opercular task control and salience systems, showed significantly changed entropy in the training group after training. Our results suggest that the integration of functional systems undergoes increased flexibility in fine-scale dynamics after musical training, which reveals how brain functional systems engage in musical performance. The efficacy of musical training induced brain plasticity may provide new therapeutic strategies for brain injury and neuropsychiatric disorders.

Linking Structure and Function in Macroscale Brain Networks

Structure-function relationships are a fundamental principle of many naturally occurring systems. However, network neuroscience research suggests that there is an imperfect link between structural connectivity and functional connectivity in the brain. Here, we synthesize the current state of knowledge linking structure and function in macroscale brain networks and discuss the different types of models used to assess this relationship. We argue that current models do not include the requisite biological detail to completely predict function. Structural network reconstructions enriched with local molecular and cellular metadata, in concert with more nuanced representations of functions and properties, hold great potential for a truly multiscale understanding of the structure-function relationship.

Investigating learning-related neural circuitry with chronic in vivo optical imaging

Fundamental aspects of brain function, including development, plasticity, learning, and memory, can take place over time scales of days to years. Chronic in vivo imaging of neural activity with cellular resolution is a powerful method for tracking the long-term activity of neural circuits. We review recent advances in our understanding of neural circuit function from diverse brain regions that have been enabled by chronic in vivo cellular imaging. Insight into the neural basis of learning and decision-making, in particular, benefit from the ability to acquire longitudinal data from genetically identified neuronal populations, deep brain areas, and subcellular structures. We propose that combining chronic imaging with further experimental and computational innovations will advance our understanding of the neural circuit mechanisms of brain function.

Modeling functional resting-state brain networks through neural message passing on the human connectome

In this work, we propose a natural model for information flow in the brain through a neural message-passing dynamics on a structural network of macroscopic regions, such as the human connectome (HC). In our model, each brain region is assumed to have a binary behavior (active or not), the strengths of interactions among them are encoded in the anatomical connectivity matrix defined by the HC, and the dynamics of the system is defined by the Belief Propagation (BP) algorithm, working near the critical point of the network. We show that in the absence of direct external stimuli the BP algorithm converges to a spatial map of activations that is similar to the Default Mode Network (DMN) of the brain, which has been defined from the analysis of functional MRI data. Moreover, we use Susceptibility Propagation (SP) to compute the matrix of long-range correlations between the different regions and show that the modules defined by a clustering of this matrix resemble several Resting State Networks (RSN) determined experimentally. Both results suggest that the functional DMN and RSNs can be seen as simple consequences of the anatomical structure of the brain and a neural message-passing dynamics between macroscopic regions. With the new model, we explore predictions on how functional maps change when the anatomical brain network suffers structural alterations, like in Alzheimer's disease and in lesions of the Corpus Callosum. The implications and novel interpretations suggested by the model, as well as the role of criticality, are discussed.

Using Low-Dimensional Manifolds to Map Relationships Between Dynamic Brain Networks

As the field of dynamic brain networks continues to expand, new methods are needed to allow for optimal handling and understanding of this explosion in data. We propose here a novel approach that embeds dynamic brain networks onto a two-dimensional (2D) manifold based on similarities and differences in network organization. Each brain network is represented as a single point on the low dimensional manifold with networks of similar topology being located in close proximity. The rich spatio-temporal information has great potential for visualization, analysis, and interpretation of dynamic brain networks. The fact that each network is represented by a single point makes it possible to switch between the low-dimensional space and the full connectivity of any given brain network. Thus, networks in a specific region of the low-dimensional space can be examined to identify network features, such as the location of brain network hubs or the interconnectivity between brain circuits. In this proof-of-concept manuscript, we show that these low dimensional manifolds contain meaningful information, as they were able to successfully discriminate between cognitive tasks and study populations. This work provides evidence that embedding dynamic brain networks onto low dimensional manifolds has the potential to help us better visualize and understand dynamic brain networks with the hope of gaining a deeper understanding of normal and abnormal brain dynamics.

Stability of spontaneous, correlated activity in mouse auditory cortex

Neural systems can be modeled as complex networks in which neural elements are represented as nodes linked to one another through structural or functional connections. The resulting network can be analyzed using mathematical tools from network science and graph theory to quantify the system’s topological organization and to better understand its function. Here, we used two-photon calcium imaging to record spontaneous activity from the same set of cells in mouse auditory cortex over the course of several weeks. We reconstruct functional networks in which cells are linked to one another by edges weighted according to the correlation of their fluorescence traces. We show that the networks exhibit modular structure across multiple topological scales and that these multi-scale modules unfold as part of a hierarchy. We also show that, on average, network architecture becomes increasingly dissimilar over time, with similarity decaying monotonically with the distance (in time) between sessions. Finally, we show that a small fraction of cells maintain strongly-correlated activity over multiple days, forming a stable temporal core surrounded by a fluctuating and variable periphery. Our work indicates a framework for studying spontaneous activity measured by two-photon calcium imaging using computational methods and graphical models from network science. The methods are flexible and easily extended to additional datasets, opening the possibility of studying cellular level network organization of neural systems and how that organization is modulated by stimuli or altered in models of disease.

Neurons coordinate their activity with one another, forming networks that help support adaptive, flexible behavior. Still, little is known about the organization of these networks at the cellular scale and their stability over time. Here, we reconstruct networks from calcium imaging data recorded in mouse primary auditory cortex. We show that these networks exhibit spatially constrained, hierarchical modular structure, which may facilitate specialized information processing. However, we show that connection weights and modular structure are also variable over time, changing on a timescale of days and adopting novel network configurations. Despite this, a small subset of neurons maintain their connections to one another and preserve their modular organization across time, forming a stable temporal core surrounded by a flexible periphery. These findings represent a conceptual bridge linking network analyses of macroscale and cellular-level neuroimaging data. They also represent a complementary approach to existing circuits- and systems-based interrogation of nervous system function, opening the door for deeper and more targeted analysis in the future.