Multi-scale community organization of the human structural connectome and its relationship with resting-state functional connectivity

BrainNetCNN: Convolutional neural networks for brain networks; towards predicting neurodevelopment

We propose BrainNetCNN, a convolutional neural network (CNN) framework to predict clinical neurodevelopmental outcomes from brain networks. In contrast to the spatially local convolutions done in traditional image-based CNNs, our BrainNetCNN is composed of novel edge-to-edge, edge-to-node and node-to-graph convolutional filters that leverage the topological locality of structural brain networks. We apply the BrainNetCNN framework to predict cognitive and motor developmental outcome scores from structural brain networks of infants born preterm. Diffusion tensor images (DTI) of preterm infants, acquired between 27 and 46 weeks gestational age, were used to construct a dataset of structural brain connectivity networks. We first demonstrate the predictive capabilities of BrainNetCNN on synthetic phantom networks with simulated injury patterns and added noise. BrainNetCNN outperforms a fully connected neural-network with the same number of model parameters on both phantoms with focal and diffuse injury patterns. We then apply our method to the task of joint prediction of Bayley-III cognitive and motor scores, assessed at 18 months of age, adjusted for prematurity. We show that our BrainNetCNN framework outperforms a variety of other methods on the same data. Furthermore, BrainNetCNN is able to identify an infant's postmenstrual age to within about 2 weeks. Finally, we explore the high-level features learned by BrainNetCNN by visualizing the importance of each connection in the brain with respect to predicting the outcome scores. These findings are then discussed in the context of the anatomy and function of the developing preterm infant brain.

Flow-Based Network Analysis of the Caenorhabditis elegans Connectome

We exploit flow propagation on the directed neuronal network of the nematode C. elegans to reveal dynamically relevant features of its connectome. We find flow-based groupings of neurons at different levels of granularity, which we relate to functional and anatomical constituents of its nervous system. A systematic in silico evaluation of the full set of single and double neuron ablations is used to identify deletions that induce the most severe disruptions of the multi-resolution flow structure. Such ablations are linked to functionally relevant neurons, and suggest potential candidates for further in vivo investigation. In addition, we use the directional patterns of incoming and outgoing network flows at all scales to identify flow profiles for the neurons in the connectome, without pre-imposing a priori categories. The four flow roles identified are linked to signal propagation motivated by biological input-response scenarios.

One of the goals of systems neuroscience is to elucidate the relationship between the structure of neuronal networks and the functional dynamics that they implement. An ideal model organism to study such interactions is the roundworm C. elegans, which not only has a fully mapped connectome, but has also been the object of extensive behavioural, genetic and neurophysiological experiments. Here we present an analysis of the neuronal network of C. elegans from a dynamical flow perspective. Our analysis reveals a multi-scale organisation of the signal flow in the network linked to anatomical and functional features of neurons, as well as identifying different neuronal roles in relation to signal propagation. We use our computational framework to explore biological input-response scenarios as well as exhaustive in silico ablations, which we relate to experimental findings reported in the literature.

Connectome sensitivity or specificity: which is more important?

Connectomes with high sensitivity and high specificity are unattainable with current axonal fiber reconstruction methods, particularly at the macro-scale afforded by magnetic resonance imaging. Tensor-guided deterministic tractography yields sparse connectomes that are incomplete and contain false negatives (FNs), whereas probabilistic methods steered by crossing-fiber models yield dense connectomes, often with low specificity due to false positives (FPs). Densely reconstructed probabilistic connectomes are typically thresholded to improve specificity at the cost of a reduction in sensitivity. What is the optimal tradeoff between connectome sensitivity and specificity? We show empirically and theoretically that specificity is paramount. Our evaluations of the impact of FPs and FNs on empirical connectomes indicate that specificity is at least twice as important as sensitivity when estimating key properties of brain networks, including topological measures of network clustering, network efficiency and network modularity. Our asymptotic analysis of small-world networks with idealized modular structure reveals that as the number of nodes grows, specificity becomes exactly twice as important as sensitivity to the estimation of the clustering coefficient. For the estimation of network efficiency, the relative importance of specificity grows linearly with the number of nodes. The greater importance of specificity is due to FPs occurring more prevalently between network modules rather than within them. These spurious inter-modular connections have a dramatic impact on network topology. We argue that efforts to maximize the sensitivity of connectome reconstruction should be realigned with the need to map brain networks with high specificity.

Generative models of the human connectome

The human connectome represents a network map of the brain's wiring diagram and the pattern into which its connections are organized is thought to play an important role in cognitive function. The generative rules that shape the topology of the human connectome remain incompletely understood. Earlier work in model organisms has suggested that wiring rules based on geometric relationships (distance) can account for many but likely not all topological features. Here we systematically explore a family of generative models of the human connectome that yield synthetic networks designed according to different wiring rules combining geometric and a broad range of topological factors. We find that a combination of geometric constraints with a homophilic attachment mechanism can create synthetic networks that closely match many topological characteristics of individual human connectomes, including features that were not included in the optimization of the generative model itself. We use these models to investigate a lifespan dataset and show that, with age, the model parameters undergo progressive changes, suggesting a rebalancing of the generative factors underlying the connectome across the lifespan.

Dynamic fluctuations coincide with periods of high and low modularity in resting-state functional brain networks

We investigate the relationship of resting-state fMRI functional connectivity estimated over long periods of time with time-varying functional connectivity estimated over shorter time intervals. We show that using Pearson's correlation to estimate functional connectivity implies that the range of fluctuations of functional connections over short time-scales is subject to statistical constraints imposed by their connectivity strength over longer scales. We present a method for estimating time-varying functional connectivity that is designed to mitigate this issue and allows us to identify episodes where functional connections are unexpectedly strong or weak. We apply this method to data recorded from N=80 participants, and show that the number of unexpectedly strong/weak connections fluctuates over time, and that these variations coincide with intermittent periods of high and low modularity in time-varying functional connectivity. We also find that during periods of relative quiescence regions associated with default mode network tend to join communities with attentional, control, and primary sensory systems. In contrast, during periods where many connections are unexpectedly strong/weak, default mode regions dissociate and form distinct modules. Finally, we go on to show that, while all functional connections can at times manifest stronger (more positively correlated) or weaker (more negatively correlated) than expected, a small number of connections, mostly within the visual and somatomotor networks, do so a disproportional number of times. Our statistical approach allows the detection of functional connections that fluctuate more or less than expected based on their long-time averages and may be of use in future studies characterizing the spatio-temporal patterns of time-varying functional connectivity.

A Functional Cartography of Cognitive Systems

One of the most remarkable features of the human brain is its ability to adapt rapidly and efficiently to external task demands. Novel and non-routine tasks, for example, are implemented faster than structural connections can be formed. The neural underpinnings of these dynamics are far from understood. Here we develop and apply novel methods in network science to quantify how patterns of functional connectivity between brain regions reconfigure as human subjects perform 64 different tasks. By applying dynamic community detection algorithms, we identify groups of brain regions that form putative functional communities, and we uncover changes in these groups across the 64-task battery. We summarize these reconfiguration patterns by quantifying the probability that two brain regions engage in the same network community (or putative functional module) across tasks. These tools enable us to demonstrate that classically defined cognitive systems-including visual, sensorimotor, auditory, default mode, fronto-parietal, cingulo-opercular and salience systems-engage dynamically in cohesive network communities across tasks. We define the network role that a cognitive system plays in these dynamics along the following two dimensions: (i) stability vs. flexibility and (ii) connected vs. isolated. The role of each system is therefore summarized by how stably that system is recruited over the 64 tasks, and how consistently that system interacts with other systems. Using this cartography, classically defined cognitive systems can be categorized as ephemeral integrators, stable loners, and anything in between. Our results provide a new conceptual framework for understanding the dynamic integration and recruitment of cognitive systems in enabling behavioral adaptability across both task and rest conditions. This work has important implications for understanding cognitive network reconfiguration during different task sets and its relationship to cognitive effort, individual variation in cognitive performance, and fatigue.

Multimodal description of whole brain connectivity: A comparison of resting state MEG, fMRI, and DWI

Structural and functional connectivity (SC and FC) have received much attention over the last decade, as they offer unique insight into the coordination of brain functioning. They are often assessed independently with three imaging modalities: SC using diffusion‐weighted imaging (DWI), FC using functional magnetic resonance imaging (fMRI), and magnetoencephalography/electroencephalography (MEG/EEG). DWI provides information about white matter organization, allowing the reconstruction of fiber bundles. fMRI uses blood‐oxygenation level‐dependent (BOLD) contrast to indirectly map neuronal activation. MEG and EEG are direct measures of neuronal activity, as they are sensitive to the synchronous inputs in pyramidal neurons. Seminal studies have targeted either the electrophysiological substrate of BOLD or the anatomical basis of FC. However, multimodal comparisons have been scarcely performed, and the relation between SC, fMRI‐FC, and MEG‐FC is still unclear. Here we present a systematic comparison of SC, resting state fMRI‐FC, and MEG‐FC between cortical regions, by evaluating their similarities at three different scales: global network, node, and hub distribution. We obtained strong similarities between the three modalities, especially for the following pairwise combinations: SC and fMRI‐FC; SC and MEG‐FC at theta, alpha, beta and gamma bands; and fMRI‐FC and MEG‐FC in alpha and beta. Furthermore, highest node similarity was found for regions of the default mode network and primary motor cortex, which also presented the highest hubness score. Distance was partially responsible for these similarities since it biased all three connectivity estimates, but not the unique contributor, since similarities remained after controlling for distance. Hum Brain Mapp 37:20–34, 2016. © 2015 The Authors Human Brain Mapping Published by Wiley Periodicals, Inc.

Analysing Local Sparseness in the Macaque Brain Network

Understanding the network structure of long distance pathways in the brain is a necessary step towards developing an insight into the brain's function, organization and evolution. Dense global subnetworks of these pathways have often been studied, primarily due to their functional implications. Instead we study sparse local subnetworks of the pathways to establish the role of a brain area in enabling shortest path communication between its non-adjacent topological neighbours. We propose a novel metric to measure the topological communication load on a vertex due to its immediate neighbourhood, and show that in terms of distribution of this local communication load, a network of Macaque long distance pathways is substantially different from other real world networks and random graph models. Macaque network contains the entire range of local subnetworks, from star-like networks to clique-like networks, while other networks tend to contain a relatively small range of subnetworks. Further, sparse local subnetworks in the Macaque network are not only found across topographical super-areas, e.g., lobes, but also within a super-area, arguing that there is conservation of even relatively short-distance pathways. To establish the communication role of a vertex we borrow the concept of brokerage from social science, and present the different types of brokerage roles that brain areas play, highlighting that not only the thalamus, but also cingulate gyrus and insula often act as "relays" for areas in the neocortex. These and other analysis of communication load and roles of the sparse subnetworks of the Macaque brain provide new insights into the organisation of its pathways.

Modular Brain Networks

The development of new technologies for mapping structural and functional brain connectivity has led to the creation of comprehensive network maps of neuronal circuits and systems. The architecture of these brain networks can be examined and analyzed with a large variety of graph theory tools. Methods for detecting modules, or network communities, are of particular interest because they uncover major building blocks or subnetworks that are particularly densely connected, often corresponding to specialized functional components. A large number of methods for community detection have become available and are now widely applied in network neuroscience. This article first surveys a number of these methods, with an emphasis on their advantages and shortcomings; then it summarizes major findings on the existence of modules in both structural and functional brain networks and briefly considers their potential functional roles in brain evolution, wiring minimization, and the emergence of functional specialization and complex dynamics.

Integrated Analysis and Visualization of Group Differences in Structural and Functional Brain Connectivity: Applications in Typical Ageing and Schizophrenia

Structural and functional brain connectivity are increasingly used to identify and analyze group differences in studies of brain disease. This study presents methods to analyze uni- and bi-modal brain connectivity and evaluate their ability to identify differences. Novel visualizations of significantly different connections comparing multiple metrics are presented. On the global level, “bi-modal comparison plots” show the distribution of uni- and bi-modal group differences and the relationship between structure and function. Differences between brain lobes are visualized using “worm plots”. Group differences in connections are examined with an existing visualization, the “connectogram”. These visualizations were evaluated in two proof-of-concept studies: (1) middle-aged versus elderly subjects; and (2) patients with schizophrenia versus controls. Each included two measures derived from diffusion weighted images and two from functional magnetic resonance images. The structural measures were minimum cost path between two anatomical regions according to the “Statistical Analysis of Minimum cost path based Structural Connectivity” method and the average fractional anisotropy along the fiber. The functional measures were Pearson’s correlation and partial correlation of mean regional time series. The relationship between structure and function was similar in both studies. Uni-modal group differences varied greatly between connectivity types. Group differences were identified in both studies globally, within brain lobes and between regions. In the aging study, minimum cost path was highly effective in identifying group differences on all levels; fractional anisotropy and mean correlation showed smaller differences on the brain lobe and regional levels. In the schizophrenia study, minimum cost path and fractional anisotropy showed differences on the global level and within brain lobes; mean correlation showed small differences on the lobe level. Only fractional anisotropy and mean correlation showed regional differences. The presented visualizations were helpful in comparing and evaluating connectivity measures on multiple levels in both studies.

Faster unfolding of communities: speeding up the Louvain algorithm

Many complex networks exhibit a modular structure of densely connected groups of nodes. Usually, such a modular structure is uncovered by the optimization of some quality function. Although flawed, modularity remains one of the most popular quality functions. The Louvain algorithm was originally developed for optimizing modularity, but has been applied to a variety of methods. As such, speeding up the Louvain algorithm enables the analysis of larger graphs in a shorter time for various methods. We here suggest to consider moving nodes to a random neighbor community, instead of the best neighbor community. Although incredibly simple, it reduces the theoretical runtime complexity from O(m) to O(nlog〈k〉) in networks with a clear community structure. In benchmark networks, it speeds up the algorithm roughly 2-3 times, while in some real networks it even reaches 10 times faster runtimes. This improvement is due to two factors: (1) a random neighbor is likely to be in a "good" community and (2) random neighbors are likely to be hubs, helping the convergence. Finally, the performance gain only slightly diminishes the quality, especially for modularity, thus providing a good quality-performance ratio. However, these gains are less pronounced, or even disappear, for some other measures such as significance or surprise.

A novel brain partition highlights the modular skeleton shared by structure and function

Elucidating the intricate relationship between brain structure and function, both in healthy and pathological conditions, is a key challenge for modern neuroscience. Recent progress in neuroimaging has helped advance our understanding of this important issue, with diffusion images providing information about structural connectivity (SC) and functional magnetic resonance imaging shedding light on resting state functional connectivity (rsFC). Here, we adopt a systems approach, relying on modular hierarchical clustering, to study together SC and rsFC datasets gathered independently from healthy human subjects. Our novel approach allows us to find a common skeleton shared by structure and function from which a new, optimal, brain partition can be extracted. We describe the emerging common structure-function modules (SFMs) in detail and compare them with commonly employed anatomical or functional parcellations. Our results underline the strong correspondence between brain structure and resting-state dynamics as well as the emerging coherent organization of the human brain.

Using Pareto optimality to explore the topology and dynamics of the human connectome

Graph theory has provided a key mathematical framework to analyse the architecture of human brain networks. This architecture embodies an inherently complex relationship between connection topology, the spatial arrangement of network elements, and the resulting network cost and functional performance. An exploration of these interacting factors and driving forces may reveal salient network features that are critically important for shaping and constraining the brain's topological organization and its evolvability. Several studies have pointed to an economic balance between network cost and network efficiency with networks organized in an ‘economical’ small-world favouring high communication efficiency at a low wiring cost. In this study, we define and explore a network morphospace in order to characterize different aspects of communication efficiency in human brain networks. Using a multi-objective evolutionary approach that approximates a Pareto-optimal set within the morphospace, we investigate the capacity of anatomical brain networks to evolve towards topologies that exhibit optimal information processing features while preserving network cost. This approach allows us to investigate network topologies that emerge under specific selection pressures, thus providing some insight into the selectional forces that may have shaped the network architecture of existing human brains.

A network convergence zone in the hippocampus

The hippocampal formation is a key structure for memory function in the brain. The functional anatomy of the brain suggests that the hippocampus may be a convergence zone, as it receives polysensory input from distributed association areas throughout the neocortex. However, recent quantitative graph-theoretic analyses of the static large-scale connectome have failed to demonstrate the centrality of the hippocampus; in the context of the whole brain, the hippocampus is not among the most connected or reachable nodes. Here we show that when communication dynamics are taken into account, the hippocampus is a key hub in the connectome. Using a novel computational model, we demonstrate that large-scale brain network topology is organized to funnel and concentrate information flow in the hippocampus, supporting the long-standing hypothesis that this region acts as a critical convergence zone. Our results indicate that the functional capacity of the hippocampus is shaped by its embedding in the large-scale connectome.

Structural connectivity abnormality in children with acute mild traumatic brain injury using graph theoretical analysis

The traumatic biomechanical forces associated with mild traumatic brain injury (mTBI) typically impart diffuse, as opposed to focal, brain injury potentially disrupting the structural connectivity between neural networks. Graph theoretical analysis using diffusion tensor imaging was used to assess injury-related differences in structural connectivity between 23 children (age 11-16 years) with mTBI and 20 age-matched children with isolated orthopedic injuries (OI) scanned within 96 h postinjury. The distribution of hub regions and the associations between alterations in regional network measures and symptom burden, as assessed by the postconcussion symptom scale score (PCSS), were also examined. In comparison to the OI group, the mTBI group was found to have significantly higher small-worldness (P < 0.0001), higher normalized clustering coefficients (P < 0.0001), higher normalized characteristic path length (P = 0.007), higher modularity (P = 0.0005), and lower global efficiency (P < 0.0001). A series of hub regions in the mTBI group were found to have significant alterations in regional network measures including nodal degree, nodal clustering coefficient, and nodal between-ness centrality. Correlation analysis showed that PCSS total score acquired at the time of imaging was significantly associated with the nodal degree of two hubs, the superior frontal gyrus at orbital section and the middle frontal gyrus. These findings provide new evidence of acute white matter alteration at both global and regional network level following mTBI in children furthering our understanding of underlying mechanisms of acute neurological insult associated with mTBI.

Frustrated hierarchical synchronization and emergent complexity in the human connectome network

The spontaneous emergence of coherent behavior through synchronization plays a key role in neural function, and its anomalies often lie at the basis of pathologies. Here we employ a parsimonious (mesoscopic) approach to study analytically and computationally the synchronization (Kuramoto) dynamics on the actual human-brain connectome network. We elucidate the existence of a so-far-uncovered intermediate phase, placed between the standard synchronous and asynchronous phases, i.e. between order and disorder. This novel phase stems from the hierarchical modular organization of the connectome. Where one would expect a hierarchical synchronization process, we show that the interplay between structural bottlenecks and quenched intrinsic frequency heterogeneities at many different scales, gives rise to frustrated synchronization, metastability, and chimera-like states, resulting in a very rich and complex phenomenology. We uncover the origin of the dynamic freezing behind these features by using spectral graph theory and discuss how the emerging complex synchronization patterns relate to the need for the brain to access –in a robust though flexible way– a large variety of functional attractors and dynamical repertoires without ad hoc fine-tuning to a critical point.

Multi-scale integration and predictability in resting state brain activity

The human brain displays heterogeneous organization in both structure and function. Here we develop a method to characterize brain regions and networks in terms of information-theoretic measures. We look at how these measures scale when larger spatial regions as well as larger connectome sub-networks are considered. This framework is applied to human brain fMRI recordings of resting-state activity and DSI-inferred structural connectivity. We find that strong functional coupling across large spatial distances distinguishes functional hubs from unimodal low-level areas, and that this long-range functional coupling correlates with structural long-range efficiency on the connectome. We also find a set of connectome regions that are both internally integrated and coupled to the rest of the brain, and which resemble previously reported resting-state networks. Finally, we argue that information-theoretic measures are useful for characterizing the functional organization of the brain at multiple scales.

Classification of autism spectrum disorder using supervised learning of brain connectivity measures extracted from synchrostates

OBJECTIVE

The paper investigates the presence of autism using the functional brain connectivity measures derived from electro-encephalogram (EEG) of children during face perception tasks.

APPROACH

Phase synchronized patterns from 128-channel EEG signals are obtained for typical children and children with autism spectrum disorder (ASD). The phase synchronized states or synchrostates temporally switch amongst themselves as an underlying process for the completion of a particular cognitive task. We used 12 subjects in each group (ASD and typical) for analyzing their EEG while processing fearful, happy and neutral faces. The minimal and maximally occurring synchrostates for each subject are chosen for extraction of brain connectivity features, which are used for classification between these two groups of subjects. Among different supervised learning techniques, we here explored the discriminant analysis and support vector machine both with polynomial kernels for the classification task.

MAIN RESULTS

The leave one out cross-validation of the classification algorithm gives 94.7% accuracy as the best performance with corresponding sensitivity and specificity values as 85.7% and 100% respectively.

SIGNIFICANCE

The proposed method gives high classification accuracies and outperforms other contemporary research results. The effectiveness of the proposed method for classification of autistic and typical children suggests the possibility of using it on a larger population to validate it for clinical practice.

Understanding Structural-Functional Relationships in the Human Brain

Relating the brain’s structural connectivity (SC) to its functional connectivity (FC) is a fundamental goal in neuroscience because it is capable of aiding our understanding of how the relatively fixed SC architecture underlies human cognition and diverse behaviors. With the aid of current noninvasive imaging technologies (e.g., structural MRI, diffusion MRI, and functional MRI) and graph theory methods, researchers have modeled the human brain as a complex network of interacting neuronal elements and characterized the underlying structural and functional connectivity patterns that support diverse cognitive functions. Specifically, research has demonstrated a tight SC-FC coupling, not only in interregional connectivity strength but also in network topologic organizations, such as community, rich-club, and motifs. Moreover, this SC-FC coupling exhibits significant changes in normal development and neuropsychiatric disorders, such as schizophrenia and epilepsy. This review summarizes recent progress regarding the SC-FC relationship of the human brain and emphasizes the important role of large-scale brain networks in the understanding of structural-functional associations. Future research directions related to this topic are also proposed.

Contributions and challenges for network models in cognitive neuroscience

The confluence of new approaches in recording patterns of brain connectivity and quantitative analytic tools from network science has opened new avenues toward understanding the organization and function of brain networks. Descriptive network models of brain structural and functional connectivity have made several important contributions; for example, in the mapping of putative network hubs and network communities. Building on the importance of anatomical and functional interactions, network models have provided insight into the basic structures and mechanisms that enable integrative neural processes. Network models have also been instrumental in understanding the role of structural brain networks in generating spatially and temporally organized brain activity. Despite these contributions, network models are subject to limitations in methodology and interpretation, and they face many challenges as brain connectivity data sets continue to increase in detail and complexity.