ADHD diagnosis guided by functional brain networks combined with domain knowledge

Utilizing functional magnetic resonance imaging (fMRI) to model functional brain networks (FBNs) is increasingly prominent in attention-deficit/hyperactivity disorder (ADHD) research, revealing neural impact and mechanisms through the exploration of activated brain regions. However, current FBNs-based methods face two major challenges. The primary challenge stems from the limitations of existing modeling methods in accurately capturing both regional correlations and long-distance dependencies (LDDs) within the dynamic brain, thereby affecting the diagnostic accuracy of FBNs as biomarkers. Additionally, limited sample size and class imbalance also pose a challenge to the learning performance of the model. To address the issues, we propose an automated diagnostic framework, which integrates modeling, multimodal fusion, and classification into a unified process. It aims to extract representative FBNs and efficiently incorporate domain knowledge to guide ADHD classification. Our work mainly includes three-fold: 1) A multi-head attention-based region-enhancement module (MAREM) is designed to simultaneously capture regional correlations and LDDs across the entire sequence of brain activity, which facilitates the construction of representative FBNs. 2) The multimodal supplementary learning module (MSLM) is proposed to integrate domain knowledge from phenotype data with FBNs from neuroimaging data, achieving information complementarity and alleviating the problems of insufficient medical data and unbalanced sample categories. 3) An ADHD automatic diagnosis framework guided by FBNs and domain knowledge (ADF-FAD) is proposed to help doctors make more accurate decisions, which is applied to the ADHD-200 dataset to confirm its effectiveness. The results indicate that the FBNs extracted by MAREM perform well in modeling and classification. After with MSLM, the model achieves accuracy of 92.4%, 74.4%, and 80% at NYU, PU, and KKI, respectively, demonstrating its ability to effectively capture crucial information related to ADHD diagnosis. Codes are available at https://github.com/zhuimengxuebao/ADF-FAD.

Topological features of functional brain networks and subclinical impulsivity: an investigation in younger and older adults

Impulsive traits (i.e., the tendency to act without forethought regardless of negative outcomes) are frequently found in healthy populations. When exposed to risk factors, individuals may develop debilitating disorders of impulse control (addiction, substance abuse, gambling) characterized by behavioral and cognitive deficits, eventually leading to huge socioeconomic costs. With the far-reaching aim of preventing the onset of impulsive disorders, it is relevant to investigate the topological organization of functional brain networks associated with impulsivity in sub-clinical populations. Taking advantage of the open-source LEMON dataset, we investigated the topological features of resting-state functional brain networks associated with impulsivity in younger (n = 146, age: 20-35) and older (n = 61, age: 59-77) individuals, using a graph-theoretical approach. Specifically, we computed indices of segregation and integration at the level of specific circuits and nodes known to be involved in impulsivity (frontal, limbic, and striatal networks). In younger individuals, results revealed that impulsivity was associated with a more widespread, less clustered and less efficient functional organization, at all levels of analyses and in all selected networks. Conversely, impulsivity in older individuals was associated with reduced integration and increased segregation of striatal regions. Speculatively, such alterations of functional brain networks might underlie behavioral and cognitive abnormalities associated with impulsivity, a working hypothesis worth being tested in future research. Lastly, differences between younger and older individuals might reflect the implementation of age-specific adaptive strategies, possibly accounting for observed differences in behavioral manifestations. Potential interpretations, limitations and implications are discussed.

Exploring timescale-specific functional brain networks and their associations with aging and cognitive performance in a healthy cohort without dementia

INTRODUCTION

Functional brain networks (FBNs) coordinate brain functions and are studied in fMRI using blood-oxygen-level-dependent (BOLD) signal correlations. Previous research links FBN changes to aging and cognitive decline, but various physiological factors influnce BOLD signals. Few studies have investigated the intrinsic components of the BOLD signal in different timescales using signal decomposition. This study aimed to explore differences between intrinsic FBNs and traditional BOLD-FBN, examining their associations with age and cognitive performance in a healthy cohort without dementia.

MATERIALS AND METHODS

A total of 396 healthy participants without dementia (men = 157; women = 239; age range = 20-85 years) were enrolled in this study. The BOLD signal was decomposed into several intrinsic signals with different timescales using ensemble empirical mode decomposition, and FBNs were constructed based on both the BOLD and intrinsic signals. Subsequently, network features-global efficiency and local efficiency values-were estimated to determine their relationship with age and cognitive performance.

RESULTS

The findings revealed that the global efficiency of traditional BOLD-FBN correlated significantly with age, with specific intrinsic FBNs contributing to these correlations. Moreover, local efficiency analysis demonstrated that intrinsic FBNs were more meaningful than traditional BOLD-FBN in identifying brain regions related to age and cognitive performance.

CONCLUSIONS

These results underscore the importance of exploring timescales of BOLD signals when constructing FBN and highlight the relevance of specific intrinsic FBNs to aging and cognitive performance. Consequently, this decomposition-based FBN-building approach may offer valuable insights for future fMRI studies.

Altered functional brain networks in problematic smartphone and social media use: resting-state fMRI study

Nowadays, the limitless availability to the World Wide Web can lead to general Internet misuse and dependence. Currently, smartphone and social media use belong to the most prevalent Internet-related behavioral addiction forms. However, the neurobiological background of these Internet-related behavioral addictions is not sufficiently explored. In this study, these addiction forms were assessed with self-reported questionnaires. Resting-state functional magnetic resonance imaging was acquired for all participants (n = 59, 29 males) to examine functional brain networks. The resting-state networks that were discovered using independent component analysis were analyzed to estimate within network differences. Significant negative associations with social media addiction and smartphone addiction were found in the language network, the lateral visual networks, the auditory network, the sensorimotor network, the executive network and the frontoparietal network. These results suggest that problematic smartphone and social media use are associated with sensory processing and higher cognitive functioning.

Autism spectrum disorder diagnosis based on deep unrolling-based spatial constraint representation

Accurate diagnosis of autism spectrum disorder (ASD) is crucial for effective treatment and prognosis. Functional brain networks (FBNs) constructed from functional magnetic resonance imaging (fMRI) have become a popular tool for ASD diagnosis. However, existing model-driven approaches used to construct FBNs lack the ability to capture potential non-linear relationships between data and labels. Moreover, most existing studies treat the FBNs construction and disease classification as separate steps, leading to large inter-subject variability in the estimated FBNs and reducing the statistical power of subsequent group comparison. To address these limitations, we propose a new approach to FBNs construction called the deep unrolling-based spatial constraint representation (DUSCR) model and integrate it with a convolutional classifier to create an end-to-end framework for ASD recognition. Specifically, the model spatial constraint representation (SCR) is solved using a proximal gradient descent algorithm, and we unroll it into deep networks using the deep unrolling algorithm. Classification is then performed using a convolutional prototype learning model. We evaluated the effectiveness of the proposed method on the ABIDE I dataset and observed a significant improvement in model performance and classification accuracy. The resting state fMRI images are preprocessed into time series data and 3D coordinates of each region of interest. The data are fed into the DUSCR model, a model for building functional brain networks using deep learning instead of traditional models, that we propose, and then the outputs are fed into the convolutional classifier with prototype learning to determine whether the patient has ASD disease.

MRgFUS of the nucleus ventralis intermedius in essential tremor modulates functional connectivity within the classical tremor network and beyond

BACKGROUND

Magnetic resonance-guided focused ultrasound (MRgFUS) of the thalamic ventral intermediate nucleus is an incisionless lesional treatment for essential tremor.

OBJECTIVE

To examine relationships between tremor severity and functional connectivity in patients with essential tremor and to assess long-term changes in the tremor network after sonication of the ventral intermediate nucleus.

METHODS

Twenty-one patients with essential tremor (70.33 ± 11.32 years) were included in the final analysis and underwent resting state functional magnetic resonance imaging at 3 T before and 6 months after treatment. Tremor severity (Fahn-Tolosa-Marin Clinical Rating Scale) was evaluated and functional connectivity was investigated using independent component analysis.

RESULTS

MRgFUS of the thalamic ventral intermediate nucleus reduced contralateral tremor effectively. Multiple regression analysis revealed exclusively negative correlations between FC and tremor severity, notably in the right cerebellar lobe VI and the left cerebellar lobe VIIIa (cerebellar network), in the left occipital fusiform gyrus (lateral visual network), the anterior division of the left superior temporal gyrus (fronto-parieto-temporal network), and in the posterior division of the left parahippocampal gyrus and the bilateral lingual gyri (default mode network). Six months after treatment, increased functional connectivity was observed in almost all tremor-associated clusters, except the cluster localized in the left cerebellum.

CONCLUSIONS

Our findings suggest that tremor-related activity in essential tremor extends beyond the classical cerebellar network, additionally involving areas related to visual processing. Functional restoration of network activity after sonication of the ventral intermediate nucleus is observed within the classical tremor network (cerebellum) and notably also in visual processing areas.

Age related changes and sex related differences of functional brain networks in childhood: A high-density EEG study

OBJECTIVE

The aim of this study was to explore functional network age-related changes and sex-related differences during the early lifespan with a high-density resting state electroencephalography (rs-EEG).

METHODS

We analyzed two data sets of high-density rs-EEG in healthy children and adolescents. We recorded a 64-channel EEG and calculated functional connectomes in 27 participants aged 5-18 years. To validate our results, we used publicly available data and calculated functional connectomes in another 86 participants aged 6-18 years from a 128-channel rs-EEG. We were primarily interested in alpha frequency band, but we also analyzed theta and beta frequency bands.

RESULTS

We observed age-related increase of characteristic path, clustering coefficient and interhemispheric strength in the alpha frequency band of both data sets and in the beta frequency band of the larger validation data set. Age-related increase of global efficiency was seen in the theta band of the validation data set and in the alpha band of the test data set. Increase in small worldness was observed only in the alpha frequency band of the test data set. We also observed an increase of individual peak alpha frequency with age in both data sets. Sex-related differences were only observed in the beta frequency band of the larger validation data set, with females having higher values than same aged males.

CONCLUSIONS

Functional brain networks show indices of higher segregation, but also increasing global integration with maturation. Age-related changes are most prominent in the alpha frequency band.

SIGNIFICANCE

To the best of our knowledge, our study was the first to analyze maturation related changes and sex-related differences of functional brain networks with a high-density EEG and to compare functional connectomes generated from two diverse high-density EEG data sets. Understanding the age-related changes and sex-related differences of functional brain networks in healthy children and adolescents is crucial for identifying network abnormalities in different neurologic and psychiatric conditions, with the aim to identify possible markers for prognosis and treatment.

Central nodes of canine functional brain networks are concentrated in the cingulate gyrus

Compared to the field of human fMRI, knowledge about functional networks in dogs is scarce. In this paper, we present the first anatomically-defined ROI (region of interest) based functional network map of the companion dog brain. We scanned 33 awake dogs in a "task-free condition". Our trained subjects, similarly to humans, remain willingly motionless during scanning. Our goal is to provide a reference map with a current best estimate for the organisation of the cerebral cortex as measured by functional connectivity. The findings extend a previous spatial ICA (independent component analysis) study (Szabo et al. in Sci Rep 9(1):1.25. https://doi.org/10.1038/s41598-019-51752-2 , 2019), with the current study including (1) more subjects and (2) improved scanning protocol to avoid asymmetric lateral distortions. In dogs, similarly to humans (Sacca et al. in J Neurosci Methods. https://doi.org/10.1016/j.jneumeth.2021.109084 , 2021), ageing resulted in increasing framewise displacement (i.e. head motion) in the scanner. Despite the inherently different approaches between model-free ICA and model-based ROI, the resulting functional networks show a remarkable similarity. However, in the present study, we did not detect a designated auditory network. Instead, we identified two highly connected, lateralised multi-region networks extending to non-homotropic regions (Sylvian L, Sylvian R), including the respective auditory regions, together with the associative and sensorimotor cortices and the insular cortex. The attention and control networks were not split into two fully separated, dedicated networks. Overall, in dogs, fronto-parietal networks and hubs were less dominant than in humans, with the cingulate gyrus playing a central role. The current manuscript provides the first attempt to map whole-brain functional networks in dogs via a model-based approach.

Abnormal Properties of Cortical Functional Brain Network in Major Depressive Disorder: Graph Theory Analysis Based on Electroencephalography-Source Estimates

Studies of scalp electroencephalography (EEG) had shown altered topological organization of functional brain networks in patients with major depressive disorder (MDD). However, most previous EEG-based network analyses were performed at sensor level, while the interpretation of obtained results was not straightforward due to volume conduction effect. To reduce the impact of this defect, the whole cortical functional brain networks of MDD patients were studied during resting state based on EEG-source estimates in this paper. First, scalp EEG signals were recorded from 19 patients with MDD and 20 normal controls under resting eyes-closed state, and cortical neural signals were estimated by using sLORETA method. Then, the correntropy coefficient of wavelet packet coefficients was performed to calculate functional connectivity (FC) matrices in four different frequency bands: δ, θ, α, β, respectively. Afterwards, topological properties of brain networks were analyzed by graph theory approaches. The results showed that the global FC strength of MDD patients was significantly higher than that of healthy subjects in α band. Also, it was found that MDD patients have abnormally increased clustering coefficient and local efficiency in both α and β bands compared to normal people. Furthermore, patients with MDD exhibited increased nodal clustering coefficients in the left lingual gryus and left precuneus in α band. In addition, β band global clustering coefficient was positively correlated with the scores of depression severity. Therefore, the findings indicated the cortical functional brain networks in MDD patients were disruptions, which suggested it would be one of potential causes of depression.

The pro-inflammatory factors contribute to the EEG microstate abnormalities in patients with major depressive disorder

Pro-inflammatory factors may be associated with abnormalities in functional brain networks, which may be a mechanism in the pathogenesis of major depressive disorder (MDD). Electroencephalogram (EEG) microstates reflect the functioning of brain networks. However, the relationship between pro-inflammatory factors and the microstate abnormalities in patients with MDD is poorly understood. 24 MDD patients and 24 age-and sex-matched healthy controls (HC) were recruited. Montgomery-Asberg Depression Rating Scale(MADRS) were assessed. Serum (interleukin- 2(IL- 2), tumor necrosis factor-α (TNF-α) and hs-C-reactive protein (CRP)and EEG data were collected. K-means clustering was performed to characterize different microstates. For each microstate, duration, occurrence and coverage were estimated. Four microstates (e.g. A, B, C, D) were characterized, MDD group showed lower duration, occurrence and coverage of microstate B and microstate D, while higher duration of microstate A and microstate C and levels of IL-2, TNF-α, hs-CRP than HC group. The duration, occurrence and coverage of microstate D were negatively correlated with levels of pro-inflammatory factors (IL-2, TNF- α and hs- CRP) (all P < 0.05). Serum pro-inflammatory induced the abnormalities of microstate D. Together, these findings add to the understanding of the pathophysiology of MDD and point to pro-inflammatory factors contribute to EEG microstate abnormalities in patients with MDD.

Collaborative learning of graph generation, clustering and classification for brain networks diagnosis

PURPOSE

Accurate diagnosis of autism spectrum disorder (ASD) plays a key role in improving the condition and quality of life for patients. In this study, we mainly focus on ASD diagnosis with functional brain networks (FBNs). The major challenge for brain networks modeling is the high dimensional connectivity in brain networks and limited number of subjects, which hinders the classification capability of graph convolutional networks (GCNs).

METHOD

To alleviate the influence of the limited data and high dimensional connectivity, we introduce a unified three-stage graph learning framework for brain network classification, involving multi-graph clustering, graph generation and graph classification. The framework combining Graph Generation, Clustering and Classification Networks (GraphCGC-Net) enhances the critical connections by multi-graph clustering (MGC) with a supervision scheme, and generates realistic brain networks by simultaneously preserving the global consistent distribution and local topology properties.

RESULTS

To demonstrate the effectiveness of our approach, we evaluate the performance of the proposed method on the Autism Brain Imaging Data Exchange (ABIDE) dataset and conduct extensive experiments on the ASD classification problem. Our proposed method achieves an average accuracy of 70.45% and an AUC of 72.76% on ABIDE. Compared with the traditional GCN model, the proposed GraphCGC-Net obtains 9.3%, and 10.64% improvement in terms of accuracy and AUC metrics, respectively.

CONCLUSION

The comprehensive experiments demonstrate that our GraphCGC-Net is effective for graph classification in brain disorders diagnosis. Moreover, we find that MGC can generate biologically meaningful subnetworks, which is highly consistent with the previous neuroimaging-derived biomarker evidence of ASD. More importantly, the promising results suggest that applying generative adversarial networks (GANs) in brain networks to improve the classification performance is worth further investigation.

Low-frequency rTMS treatment alters the topographical organization of functional brain networks in schizophrenia patients with auditory verbal hallucination

Auditory verbal hallucinations (AVH) are an important characteristic of schizophrenia. Repeated transcranial magnetic stimulation (rTMS) has been evidence to be effective in treating AVH. We evaluated the topological properties of resting-state functional brain networks in schizophrenia patients with AVH (n = 32) who received 1-Hz rTMS treatment and matched healthy controls (n = 33). The results showed that the psychotic symptoms and certain neurocognitive performances in patients were improved by rTMS treatment. Furthermore, the pretreatment patients showed abnormal global topological metrics compared with the controls, including lower global efficiency (E_glob, represents the relative quality of information transmission between all nodes in the network) and higher characteristic path length (L_p, characterizes the mean shortest distance between any two nodes in the network). The pretreament patients also showed decreased local topological metrics relative to the controls, including the nodal shortest path (NL_p, quantifies the mean distance between the given node and the other nodes in the network) and nodal efficiency (N_e, measures the information interchange among the neighbor nodes when one node is removed), mainly located in the prefrontal cortex, occipital cortex, and subcortical regions. While the abnormal global and local topological patterns were normalized in patients after rTMS treatment. The multiple linear regression analysis indicated that the baseline topological metrics could be associated with the clinical responses after treatment in the patient group. The results suggested that the topological organization of the functional brain network was globally and regionally altered in schizophrenia patients with AVH after rTMS treatment and may be a potential therapeutic effect for AVH in schizophrenia.

Evidence from theta-burst stimulation that age-related de-differentiation of the hippocampal network is functional for episodic memory

Episodic memory is supported by hippocampal interactions with a distributed network. Aging is associated with memory decline and network de-differentiation. However, the role of de-differentiation in memory decline has not been directly tested. We reasoned that hippocampal network-targeted stimulation could test these theories, as age-related changes in the network response to stimulation would indicate network reorganization, and corresponding changes in memory would suggest that this reorganization is functional. We compared effects of stimulation on fMRI connectivity and memory in younger versus older adults. Theta-burst network-targeted stimulation of left lateral parietal cortex selectively increased hippocampal network connectivity and modulated memory in younger adults. In contrast, stimulation in older adults increased connectivity throughout the brain, without network selectivity, and did not influence memory. These findings provide evidence that network responses to stimulation are de-differentiated in aging and suggest that age-related de-differentiation is relevant for memory. This manuscript is part of the Special Issue entitled "Cognitive Neuroscience of Healthy and Pathological Aging" edited by Drs. M. N. Rajah, S. Belleville, and R. Cabeza. This article is part of the Virtual Special Issue titled COGNITIVE NEUROSCIENCE OF HEALTHY AND PATHOLOGICAL AGING. The full issue can be found on ScienceDirect at https://www.sciencedirect.com/journal/neurobiology-of-aging/special-issue/105379XPWJP.

Multiple Functional Brain Networks Related to Pain Perception Revealed by fMRI

The rise of functional magnetic resonance imaging (fMRI) has led to a deeper understanding of cortical processing of pain. Central to these advances has been the identification and analysis of "functional networks", often derived from groups of pre-selected pain regions. In this study our main objective was to identify functional brain networks related to pain perception by examining whole-brain activation, avoiding the need for a priori selection of regions. We applied a data-driven technique-Constrained Principal Component Analysis for fMRI (fMRI-CPCA)-that identifies networks without assuming their anatomical or temporal properties. Open-source fMRI data collected during a thermal pain task (33 healthy participants) were subjected to fMRI-CPCA for network extraction, and networks were associated with pain perception by modelling subjective pain ratings as a function of network activation intensities. Three functional networks emerged: a sensorimotor response network, a salience-mediated attention network, and the default-mode network. Together, these networks constituted a brain state that explained variability in pain perception, both within and between individuals, demonstrating the potential of data-driven, whole-brain functional network techniques for the analysis of pain imaging data.

Pubertal timing and functional neurodevelopmental alterations independently mediate the effect of family conflict on adolescent psychopathology

This study tested the hypothesis that early life adversity (ELA) heightens psychopathology risk by concurrently altering pubertal and neurodevelopmental timing, and associated gene transcription signatures. Analyses focused on threat- (family conflict/neighbourhood crime) and deprivation-related ELAs (parental inattentiveness/unmet material needs), using longitudinal data from 1514 biologically unrelated youths in the Adolescent Brain and Cognitive Development (ABCD) study. Typical developmental changes in white matter microstructure corresponded to widespread BOLD signal variability (BOLDsv) increases (linked to cell communication and biosynthesis genes) and region-specific task-related BOLDsv increases/decreases (linked to signal transduction, immune and external environmental response genes). Increasing resting-state (RS), but decreasing task-related BOLDsv predicted normative functional network segregation. Family conflict was the strongest concurrent and prospective contributor to psychopathology, while material deprivation constituted an additive risk factor. ELA-linked psychopathology was predicted by higher Time 1 threat-evoked BOLDSV (associated with axonal development, myelination, cell differentiation and signal transduction genes), reduced Time 2 RS BOLDsv (associated with cell metabolism and attention genes) and greater Time 1 to Time 2 control/attention network segregation. Earlier pubertal timing and neurodevelopmental alterations independently mediated ELA effects on psychopathology. Our results underscore the differential roles of the immediate and wider external environment(s) in concurrent and longer-term ELA consequences.

Separate neural networks of implicit emotional processing between pictures and words: A coordinate-based meta-analysis of brain imaging studies

Both pictures and words are frequently employed as experimental stimuli to investigate the neurocognitive mechanisms of emotional processing. However, it remains unclear whether emotional picture processing and emotional word processing share neural underpinnings. To address this issue, we focus on neuroimaging studies examining the implicit processing of affective words and pictures, which require participants to meet cognitive task demands under the implicit influence of emotional pictorial or verbal stimuli. A coordinate-based activation likelihood estimation meta-analysis was conducted on these studies, which revealed no common activation maximum between the picture and word conditions. Specifically, implicit negative picture processing (35 experiments, 393 foci, and 932 subjects) engages the bilateral amygdala, left hippocampus, fusiform gyri, and right insula, which are mainly located in the subcortical network and visual network associated with bottom-up emotional responses. In contrast, implicit negative word processing (34 experiments, 316 foci, and 799 subjects) engages the default mode network and fronto-parietal network including the ventrolateral prefrontal cortex, dorsolateral prefrontal cortex, and dorsomedial prefrontal cortex, indicating the involvement of top-down semantic processing and emotion regulation. Our findings indicate that affective pictures (that intrinsically have an affective valence) and affective words (that inherit the affective valence from their object) modulate implicit emotional processing in different ways, and therefore recruit distinct brain systems.

Functional brain networks underlying automatic and controlled handwriting in Chinese

This study aimed to identify the functional brain networks underlying the distinctions between automatic and controlled handwriting in Chinese. Network-based analysis was applied to functional magnetic resonance imaging data collected while adult participants performed a copying task under automatic and speed-controlled conditions. We found significant differences between automatic and speed-controlled handwriting in functional connectivity within and between the frontoparietal network, default mode network, dorsal attention network, somatomotor network and visual network; these differences reflect the variations in general attentional control and task-relevant visuomotor operations. However, no differences in brain activation were detected between the two handwriting conditions, suggesting that the reorganization of functional networks, rather than the modulation of local brain activation, underlies the dissociations between automatic and controlled handwriting in Chinese. Our findings illustrate the brain basis of handwriting automaticity, shedding new light on how handwriting automaticity may be disrupted in individuals with neurological disorders.

GAT-LI: a graph attention network based learning and interpreting method for functional brain network classification

BACKGROUND

Autism spectrum disorders (ASD) imply a spectrum of symptoms rather than a single phenotype. ASD could affect brain connectivity at different degree based on the severity of the symptom. Given their excellent learning capability, graph neural networks (GNN) methods have recently been used to uncover functional connectivity patterns and biological mechanisms in neuropsychiatric disorders, such as ASD. However, there remain challenges to develop an accurate GNN learning model and understand how specific decisions of these graph models are made in brain network analysis.

RESULTS

In this paper, we propose a graph attention network based learning and interpreting method, namely GAT-LI, which learns to classify functional brain networks of ASD individuals versus healthy controls (HC), and interprets the learned graph model with feature importance. Specifically, GAT-LI includes a graph learning stage and an interpreting stage. First, in the graph learning stage, a new graph attention network model, namely GAT2, uses graph attention layers to learn the node representation, and a novel attention pooling layer to obtain the graph representation for functional brain network classification. We experimentally compared GAT2 model's performance on the ABIDE I database from 1035 subjects against the classification performances of other well-known models, and the results showed that the GAT2 model achieved the best classification performance. We experimentally compared the influence of different construction methods of brain networks in GAT2 model. We also used a larger synthetic graph dataset with 4000 samples to validate the utility and power of GAT2 model. Second, in the interpreting stage, we used GNNExplainer to interpret learned GAT2 model with feature importance. We experimentally compared GNNExplainer with two well-known interpretation methods including Saliency Map and DeepLIFT to interpret the learned model, and the results showed GNNExplainer achieved the best interpretation performance. We further used the interpretation method to identify the features that contributed most in classifying ASD versus HC.

CONCLUSION

We propose a two-stage learning and interpreting method GAT-LI to classify functional brain networks and interpret the feature importance in the graph model. The method should also be useful in the classification and interpretation tasks for graph data from other biomedical scenarios.

Learning dynamic graph embeddings for accurate detection of cognitive state changes in functional brain networks

Mounting evidence shows that brain functions and cognitive states are dynamically changing even in the resting state rather than remaining at a single constant state. Due to the relatively small changes in BOLD (blood-oxygen-level-dependent) signals across tasks, it is difficult to detect the change of cognitive status without requiring prior knowledge of the experimental design. To address this challenge, we present a dynamic graph learning approach to generate an ensemble of subject-specific dynamic graph embeddings, which allows us to use brain networks to disentangle cognitive events more accurately than using raw BOLD signals. The backbone of our method is essentially a representation learning process for projecting BOLD signals into a latent vertex-temporal domain with the greater biological underpinning of brain activities. Specifically, the learned representation domain is jointly formed by (1) a set of harmonic waves that govern the topology of whole-brain functional connectivities and (2) a set of Fourier bases that characterize the temporal dynamics of functional changes. In this regard our dynamic graph embeddings provide a new methodology to investigate how these self-organized functional fluctuation patterns oscillate along with the evolving cognitive status. We have evaluated our proposed method on both simulated data and working memory task-based fMRI datasets, where our dynamic graph embeddings achieve higher accuracy in detecting multiple cognitive states than other state-of-the-art methods.

Brain hothubs and dark functional networks: correlation analysis between amplitude and connectivity for Broca's aphasia

Source localization and functional brain network modeling are methods of identifying critical regions during cognitive tasks. The first activity estimates the relative differences of the signal amplitudes in regions of interest (ROI) and the second activity measures the statistical dependence among signal fluctuations. We hypothesized that the source amplitude–functional connectivity relationship decouples or reverses in persons having brain impairments. Five Broca’s aphasics with five matched cognitively healthy controls underwent overt picture-naming magnetoencephalography scans. The gamma-band (30–45 Hz) phase-locking values were calculated as connections among the ROIs. We calculated the partial correlation coefficients between the amplitudes and network measures and detected four node types, including hothubs with high amplitude and high connectivity, coldhubs with high connectivity but lower amplitude, non-hub hotspots, and non-hub coldspots. The results indicate that the high-amplitude regions are not necessarily highly connected hubs. Furthermore, the Broca aphasics utilized different hothub sets for the naming task. Both groups had dark functional networks composed of coldhubs. Thus, source amplitude–functional connectivity relationships could help reveal functional reorganizations in patients. The amplitude–connectivity combination provides a new perspective for pathological studies of the brain’s dark functional networks.

Altered large-scale functional brain organization in posttraumatic stress disorder: A comprehensive review of univariate and network-level neurocircuitry models of PTSD

Classical neural circuitry models of posttraumatic stress disorder (PTSD) are largely derived from univariate activation studies and implicate the fronto-limbic circuit as a main neural correlate of PTSD symptoms. Though well-supported by human neuroimaging literature, these models are limited in their ability to explain the widely distributed neural and behavioral deficits in PTSD. Emerging interest in the application of large-scale network methods to functional neuroimaging provides a new opportunity to overcome such limitations and conceptualize the neural circuitry of PTSD in the context of network patterns. This review aims to evaluate both the classical neural circuitry model and a new, network-based model of PTSD neural circuitry using a breadth of functional brain organization research in subjects with PTSD. Taken together, this literature suggests global patterns of reduced functional connectivity (FC) in PTSD groups as well as altered FC targets that reside disproportionately in canonical functional networks, especially the default mode network. This provides evidence for an integrative model that includes elements of both the classical models and network-based models to characterize the neural circuitry of PTSD.

Antidepressants normalize brain flexibility associated with multi-dimensional symptoms in major depressive patients

BACKGROUND

The fundamental pathophysiology of major depressive disorder (MDD) could be characterized by functional brain networks which tightly and dynamically connect into groups as communities, making the flexible brain possible to external multifarious demands. We aim to scrutinize what brain dynamics go awry in MDD and antidepressants effects on multi-dimensional symptoms.

METHODS

Thirty-five patients and thirty-five controls underwent resting-state functional magnetic resonance imaging (MRI). Patients were scanned before and after 8 or 12 weeks of pharmacotherapy. Group independent component analysis decomposed resting-state images to instinct networks and networks' integrated flexibility was calculated. Network flexibility between patients at baseline and after therapy were compared.

RESULTS

All patients completed the clinical trial and MRI scans. Following antidepressants treatment, we found significant normalization of reduced network flexibility in default mode network (DMN) and cognitive control network (CCN) of MDD patients. Selectively significant correlations between network flexibility and multi-dimensional symptoms such as anxiety/somatization and hysteresis factor were also found.

CONCLUSIONS

"Hypoflexible" CCN may involve in anxiety syndrome. Low flexibility in DMN may be indicative of hysteresis. These suggest an important pathophysiology of depressive manifestation of MDD. The antidepressant-induced normalization of the "hypoflexibility" suggests a selective pathway through which antidepressants may alleviate symptoms in depression.

Bilateral nucleus basalis of Meynert deep brain stimulation for dementia with Lewy bodies: A randomised clinical trial

BACKGROUND

Dementia with Lewy bodies (DLB) is the second most common form of dementia. Current symptomatic treatment with medications remains inadequate. Deep brain stimulation of the nucleus basalis of Meynert (NBM DBS) has been proposed as a potential new treatment option in dementias.

OBJECTIVE

To assess the safety and tolerability of low frequency (20 Hz) NBM DBS in DLB patients and explore its potential effects on both clinical symptoms and functional connectivity in underlying cognitive networks.

METHODS

We conducted an exploratory randomised, double-blind, crossover trial of NBM DBS in six DLB patients recruited from two UK neuroscience centres. Patients were aged between 50 and 80 years, had mild-moderate dementia symptoms and were living with a carer-informant. Patients underwent image guided stereotactic implantation of bilateral DBS electrodes with the deepest contacts positioned in the Ch4i subsector of NBM. Patients were subsequently assigned to receive either active or sham stimulation for six weeks, followed by a two week washout period, then the opposite condition for six weeks. Safety and tolerability of both the surgery and stimulation were systematically evaluated throughout. Exploratory outcomes included the difference in scores on standardised measurements of cognitive, psychiatric and motor symptoms between the active and sham stimulation conditions, as well as differences in functional connectivity in discrete cognitive networks on resting state fMRI.

RESULTS

Surgery and stimulation were well tolerated by all six patients (five male, mean age 71.33 years). One serious adverse event occurred: one patient developed antibiotic-associated colitis, prolonging his hospital stay by two weeks. No consistent improvements were observed in exploratory clinical outcome measures, but the severity of neuropsychiatric symptoms reduced with NBM DBS in 3/5 patients. Active stimulation was associated with functional connectivity changes in both the default mode network and the frontoparietal network.

CONCLUSION

Low frequency NBM DBS can be safely conducted in DLB patients. This should encourage further exploration of the possible effects of stimulation on neuropsychiatric symptoms and corresponding changes in functional connectivity in cognitive networks.

TRIAL REGISTRATION NUMBER

NCT02263937.

Functional brain network architecture supporting the learning of social networks in humans

Most humans have the good fortune to live their lives embedded in richly structured social groups. Yet, it remains unclear how humans acquire knowledge about these social structures to successfully navigate social relationships. Here we address this knowledge gap with an interdisciplinary neuroimaging study drawing on recent advances in network science and statistical learning. Specifically, we collected BOLD MRI data while participants learned the community structure of both social and non-social networks, in order to examine whether the learning of these two types of networks was differentially associated with functional brain network topology. We found that participants learned the community structure of the networks, as evidenced by a slower reaction time when a trial moved between communities than when a trial moved within a community. Learning the community structure of social networks was also characterized by significantly greater functional connectivity of the hippocampus and temporoparietal junction when transitioning between communities than when transitioning within a community. Furthermore, temporoparietal regions of the default mode were more strongly connected to hippocampus, somatomotor, and visual regions for social networks than for non-social networks. Collectively, our results identify neurophysiological underpinnings of social versus non-social network learning, extending our knowledge about the impact of social context on learning processes. More broadly, this work offers an empirical approach to study the learning of social network structures, which could be fruitfully extended to other participant populations, various graph architectures, and a diversity of social contexts in future studies.

Functional brain networks underlying evidence integration and delusional ideation

Cognitive biases affecting evidence integration contribute to delusions and delusional ideation in the psychosis continuum. In previously published work we observed hyperactivity in a visual attention network (VsAN) during confirmatory evidence integration, and hypoactivity in a cognitive evaluation network (CEN) during disconfirmatory evidence integration in schizophrenia patients with delusions, suggesting that a task-specific imbalance of these networks may contribute to delusion maintenance. In the current study, we investigated whether patterns of aberrant functional connectivity observed in past work were associated with delusional ideation in 41 healthy individuals by examining associations between cognitive biases, subclinical schizotypal traits, and functional brain activity during evidence integration. Behaviourally, we replicated positive associations between schizotypal traits and cognitive biases and further showed that this association was driven by delusional ideation specifically. Constrained principal component analysis for fMRI (fMRI-CPCA) revealed recruitment of the brain networks observed in our previous clinical and non-clinical evidence integration studies: default-mode network (DMN); cognitive evaluation network (CEN); and visual attention (VsAN) network. Moreover, as with clinically-significant delusions, delusional ideation was associated with decreased CEN activity during the processing of disconfirmatory evidence and increased VsAN activity during the processing of confirmatory evidence. These findings suggest that this altered pattern of activation across networks during evidence integration may underlie delusional ideation and delusions in the psychosis continuum.

Exact topological inference of the resting-state brain networks in twins

A cycle in a brain network is a subset of a connected component with redundant additional connections. If there are many cycles in a connected component, the connected component is more densely connected. Whereas the number of connected components represents the integration of the brain network, the number of cycles represents how strong the integration is. However, it is unclear how to perform statistical inference on the number of cycles in the brain network. In this study, we present a new statistical inference framework for determining the significance of the number of cycles through the Kolmogorov-Smirnov (KS) distance, which was recently introduced to measure the similarity between networks across different filtration values by using the zeroth Betti number. In this paper, we show how to extend the method to the first Betti number, which measures the number of cycles. The performance analysis was conducted using the random network simulations with ground truths. By using a twin imaging study, which provides biological ground truth, the methods are applied in determining if the number of cycles is a statistically significant heritable network feature in the resting-state functional connectivity in 217 twins obtained from the Human Connectome Project. The MATLAB codes as well as the connectivity matrices used in generating results are provided at http://www.stat.wisc.edu/∼mchung/TDA.

In this paper, we propose a new topological distance based on the Kolmogorov-Smirnov (KS) distance that is adapted for brain networks, and compare them against other topological network distances including the Gromov-Hausdorff (GH) distances. KS-distance is recently introduced to measure the similarity between networks across different filtration values by using the zeroth Betti number, which measures the number of connected components. In this paper, we show how to extend the method to the first Betti number, which measures the number of cycles. The performance analysis was conducted using random network simulations with ground truths. Using a twin imaging study, which provides biological ground truth (of network differences), we demonstrate that the KS distances on the zeroth and first Betti numbers have the ability to determine heritability.

Automatic recognition of holistic functional brain networks using iteratively optimized convolutional neural networks (IO-CNN) with weak label initialization

fMRI data decomposition techniques have advanced significantly from shallow models such as Independent Component Analysis (ICA) and Sparse Coding and Dictionary Learning (SCDL) to deep learning models such Deep Belief Networks (DBN) and Convolutional Autoencoder (DCAE). However, interpretations of those decomposed networks are still open questions due to the lack of functional brain atlases, no correspondence across decomposed or reconstructed networks across different subjects, and significant individual variabilities. Recent studies showed that deep learning, especially deep convolutional neural networks (CNN), has extraordinary ability of accommodating spatial object patterns, e.g., our recent works using 3D CNN for fMRI-derived network classifications achieved high accuracy with a remarkable tolerance for mistakenly labelled training brain networks. However, the training data preparation is one of the biggest obstacles in these supervised deep learning models for functional brain network map recognitions, since manual labelling requires tedious and time-consuming labours which will sometimes even introduce label mistakes. Especially for mapping functional networks in large scale datasets such as hundreds of thousands of brain networks used in this paper, the manual labelling method will become almost infeasible. In response, in this work, we tackled both the network recognition and training data labelling tasks by proposing a new iteratively optimized deep learning CNN (IO-CNN) framework with an automatic weak label initialization, which enables the functional brain networks recognition task to a fully automatic large-scale classification procedure. Our extensive experiments based on ABIDE-II 1099 brains' fMRI data showed the great promise of our IO-CNN framework.

Decreased integration of EEG source-space networks in disorders of consciousness

Increasing evidence links disorders of consciousness (DOC) with disruptions in functional connectivity between distant brain areas. However, to which extent the balance of brain network segregation and integration is modified in DOC patients remains unclear. Using high-density electroencephalography (EEG), the objective of our study was to characterize the local and global topological changes of DOC patients' functional brain networks.

Resting state high-density-EEG data were collected and analyzed from 82 participants: 61 DOC patients recovering from coma with various levels of consciousness (EMCS (n = 6), MCS+ (n = 29), MCS- (n = 17) and UWS (n = 9)), and 21 healthy subjects (i.e., controls). Functional brain networks in five different EEG frequency bands and the broadband signal were estimated using an EEG connectivity approach at the source level. Graph theory-based analyses were used to evaluate their relationship with decreasing levels of consciousness as well as group differences between healthy volunteers and DOC patient groups.

Results showed that networks in DOC patients are characterized by impaired global information processing (network integration) and increased local information processing (network segregation) as compared to controls. The large-scale functional brain networks had integration decreasing with lower level of consciousness.

•

Long-distance communication between brain regions is altered in patients suffering from disorders of consciousness.

•

Impaired consciousness is associated with disruptions in brain network integration.

•

The left orbitofrontal and left precuneus were identified in all patients groups.

Changes in functional brain networks and neurocognitive function in Chinese gynecological cancer patients after chemotherapy: a prospective longitudinal study

Background

Previous neurocognitive assessments in non-central nervous system cancers highlight the high incidence of neurocognitive dysfunction in this study population. However, there have been few studies exploring neurocognitive dysfunction induced by chemotherapy in gynecological cancer patients. This prospective longitudinal study was conducted to assess neurocognitive functioning and functional brain networks in Chinese gynecological cancer patients pre- and post-chemotherapy, while additionally including age-matched healthy subjects as the control group.

Methods

All research participants were evaluated using a resting-state functional magnetic resonance imaging and neurocognition assessment. Behavioral data were conducted using SPSS for descriptive statistics, correlation and comparison analyses. Preprocessing of MRI (Magnetic Resonance Imaging) data and network analyses were performed using GRETNA (Graph Theoretical Network Analysis).

Results

A total of 40 subjects joined this study, with 20 subjects in each group. With the exception of the mean of psychomotor speed, there was no significant difference pre-chemotherapy between patients and healthy controls in neurocognitive test mean scores (Ps > 0.05). During the post-chemotherapy assessment, there were significant differences in the mean scores of neurocognitive tests (including Digit Span tests, verbal memory, immediate recall, delayed recall, and information processing speed tests) (all Ps < 0 .05). Longitudinal graph analysis revealed statistically significant differences in the patient group, with significant decreases in both local efficiency (P < 0.01) and global efficiency (P = 0.04). Lower raw TMT-A scores were significantly associated with lower local efficiency (r = 0.37, P = 0.03). Lower verbal memory scores were statistically significant and associated with lower global efficiency (r = 0.54, P = 0.02) in the patient group, but not in the healthy control group.

Conclusions

This study found that the risk of brain function and neurocognitive changes following chemotherapy could potentially guide patients in making appropriate treatment decisions, and this study may identify a cohort that could be suited for study of an intervention.

Anxious brain networks: A coordinate-based activation likelihood estimation meta-analysis of resting-state functional connectivity studies in anxiety

Anxiety and anxiety disorders are associated with specific alterations to functional brain networks, including intra-networks and inter-networks. Given the heterogeneity within anxiety disorders and inconsistencies in functional network differences across studies, identifying common patterns of altered brain networks in anxiety is imperative. Here, we conducted an activation likelihood estimation meta-analysis of resting-state functional connectivity studies in anxiety and anxiety disorders (including 835 individuals with different levels of anxiety or anxiety disorders and 508 controls). Results show that anxiety can be characterized by hypo-connectivity of the affective network with executive control network (ECN) and default mode network (DMN), as well as decoupling of the ECN with the DMN. The connectivity within the salience network and its connectivity with sensorimotor network are also attenuated. These results reveal consistent dysregulations of affective and cognitive control related networks over networks related to emotion processing in anxiety and anxiety disorders. The current findings provide an empirical foundation for an integrated model of brain network alterations that are common across anxiety and anxiety disorders.

Frequency-based brain networks: From a multiplex framework to a full multilayer description

We explore how to study dynamical interactions between brain regions by using functional multilayer networks whose layers represent different frequency bands at which a brain operates. Specifically, we investigate the consequences of considering the brain as (i) a multilayer network, in which all brain regions can interact with each other at different frequency bands; and as (ii) a multiplex network, in which interactions between different frequency bands are allowed only within each brain region and not between them. We study the second-smallest eigenvalue λ 2 of the combinatorial supra-Laplacian matrix of both the multiplex and multilayer networks, as λ 2 has been used previously as an indicator of network synchronizability and as a biomarker for several brain diseases. We show that the heterogeneity of interlayer edge weights and, especially, the fraction of missing edges crucially modify the value of λ 2, and we illustrate our results with both synthetic network models and real data obtained from resting-state magnetoencephalography. Our work highlights the differences between using a multiplex approach and a full multilayer approach when studying frequency-based multilayer brain networks.

Resting-state brain network features associated with short-term skill learning ability in humans and the influence of N-methyl-d-aspartate receptor antagonism

Graph theoretical functional magnetic resonance imaging (fMRI) studies have demonstrated that brain networks reorganize significantly during motor skill acquisition, yet the associations between motor learning ability, brain network features, and the underlying biological mechanisms remain unclear. In the current study, we applied a visually guided sequential pinch force learning task and graph theoretical analyses to investigate the associations between short-term motor learning ability and resting-state brain network metrics in 60 healthy subjects. We further probed the test-retest reliability (n = 26) and potential effects of the N-methyl-d-aspartate (NMDA) antagonist ketamine (n = 19) in independent healthy volunteers. Our results show that the improvement of motor performance after short-term training was positively correlated with small-worldness (p = 0.032) and global efficiency (p = 0.025), whereas negatively correlated with characteristic path length (p = 0.014) and transitivity (p = 0.025). In addition, using network-based statistics (NBS), we identified a learning ability–associated (p = 0.037) and ketamine-susceptible (p = 0.027) cerebellar-cortical network with fair to good reliability (intraclass correlation coefficient [ICC] > 0.7) and higher functional connectivity in better learners. Our results provide new evidence for the association of intrinsic brain network features with motor learning and suggest a role of NMDA-related glutamatergic processes in learning-associated subnetworks.

Learning a new motor skill prompts immediate reconfigurations of distributed brain networks followed by adaptive changes in intrinsic brain circuits related to synaptic plasticity. Here, we identify global brain network properties and a cerebellar-cortical functional subnetwork that are both significantly associated with motor learning ability in a previously trained visuomotor task in humans. We further show that the associated functional subnetwork connectivity but not the global brain network properties are susceptible to ketamine. Our findings suggest a distinct functional role for learning-related global versus local network metrics and support the idea of a preferential susceptibility of learning-associated subnetworks to N-methyl-d-aspartate antagonist and plasticity-related consolidation effects.

Assessment of driver drowsiness using electroencephalogram signals based on multiple functional brain networks

This paper proposes a comprehensive approach to explore whether functional brain network (FBN) changes from the alert state to the drowsy state and to find out ideal neurophysiology indicators able to detect driver drowsiness in terms of FBN. A driving simulation experiment consisting of two driving tasks is designed and conducted using fifteen participant drivers. Collected EEG signals are then decomposed into multiple frequency bands by wavelet packet transform (WPT). Based on this, two novel FBN approaches, synchronization likelihood (SL) and minimum spanning tree (MST) are combined and applied to feature recognition and classification system. Unlike other methods, our approaches focus on the interaction and correlation between different brain regions. Statistical analysis of network features indicates that the difference between alert state and drowsy state are significant and further confirmed that brain network configuration should be related to drowsiness. For classification, these brain network features are selected and then fed into four classifiers considered namely Support Vector Machines (SVM), K Nearest Neighbors classifier (KNN), Logistic Regression (LR) and Decision Trees (DT). It is found that combining MST method and SL method is actually increasing the classification accuracy with all classifiers considered in this work especially the KNN classifier from 95.4% to 98.6%. Moreover, KNN classifier also gives the highest precision of 98.3%, sensitivity of 98.8% and specificity of 98.9%. Thus this kind of methodology might be a useful tool for further understanding the neurophysiology mechanisms of driver drowsiness, and as a reference work for future studies or future 'systems'.

δ-MAPS: from spatio-temporal data to a weighted and lagged network between functional domains

In real physical systems the underlying spatial components might not have crisp boundaries and their interactions might not be instantaneous. To this end, we propose δ-MAPS; a method that identifies spatially contiguous and possibly overlapping components referred to as domains, and identifies the lagged functional relationships between them. Informally, a domain is a spatially contiguous region that somehow participates in the same dynamic effect or function. The latter will result in highly correlated temporal activity between grid cells of the same domain. δ-MAPS first identifies the epicenters of activity of a domain. Next, it identifies a domain as the maximum possible set of spatially contiguous grid cells that include the detected epicenters and satisfy a homogeneity constraint. After identifying the domains, δ-MAPS infers a functional network between them. The proposed network inference method examines the statistical significance of each lagged correlation between two domains, applies a multiple-testing process to control the rate of false positives, infers a range of potential lag values for each edge, and assigns a weight to each edge reflecting the magnitude of interaction between two domains. δ-MAPS is related to clustering, multivariate statistical techniques and network community detection. However, as we discuss and also show with synthetic data, it is also significantly different, avoiding many of the known limitations of these methods. We illustrate the application of δ-MAPS on data from two domains: climate science and neuroscience. First, the sea-surface temperature climate network identifies some well-known teleconnections (such as the lagged connection between the El Nin õ Southern Oscillation and the Indian Ocean). Second, the analysis of resting state fMRI cortical data confirms the presence of known functional resting state networks (default mode, occipital, motor/somatosensory and auditory), and shows that the cortical network includes a backbone of relatively few regions that are densely interconnected.

Constructing fine-granularity functional brain network atlases via deep convolutional autoencoder

State-of-the-art functional brain network reconstruction methods such as independent component analysis (ICA) or sparse coding of whole-brain fMRI data can effectively infer many thousands of volumetric brain network maps from a large number of human brains. However, due to the variability of individual brain networks and the large scale of such networks needed for statistically meaningful group-level analysis, it is still a challenging and open problem to derive group-wise common networks as network atlases. Inspired by the superior spatial pattern description ability of the deep convolutional neural networks (CNNs), a novel deep 3D convolutional autoencoder (CAE) network is designed here to extract spatial brain network features effectively, based on which an Apache Spark enabled computational framework is developed for fast clustering of larger number of network maps into fine-granularity atlases. To evaluate this framework, 10 resting state networks (RSNs) were manually labeled from the sparsely decomposed networks of Human Connectome Project (HCP) fMRI data and 5275 network training samples were obtained, in total. Then the deep CAE models are trained by these functional networks' spatial maps, and the learned features are used to refine the original 10 RSNs into 17 network atlases that possess fine-granularity functional network patterns. Interestingly, it turned out that some manually mislabeled outliers in training networks can be corrected by the deep CAE derived features. More importantly, fine granularities of networks can be identified and they reveal unique network patterns specific to different brain task states. By further applying this method to a dataset of mild traumatic brain injury study, it shows that the technique can effectively identify abnormal small networks in brain injury patients in comparison with controls. In general, our work presents a promising deep learning and big data analysis solution for modeling functional connectomes, with fine granularities, based on fMRI data.

Spatio-temporal modeling of connectome-scale brain network interactions via time-evolving graphs

Many recent literature studies have revealed interesting dynamics patterns of functional brain networks derived from fMRI data. However, it has been rarely explored how functional networks spatially overlap (or interact) and how such connectome-scale network interactions temporally evolve. To explore these unanswered questions, this paper presents a novel framework for spatio-temporal modeling of connectome-scale functional brain network interactions via two main effective computational methodologies. First, to integrate, pool and compare brain networks across individuals and their cognitive states under task performances, we designed a novel group-wise dictionary learning scheme to derive connectome-scale consistent brain network templates that can be used to define the common reference space of brain network interactions. Second, the temporal dynamics of spatial network interactions is modeled by a weighted time-evolving graph, and then a data-driven unsupervised learning algorithm based on the dynamic behavioral mixed-membership model (DBMM) is adopted to identify behavioral patterns of brain networks during the temporal evolution process of spatial overlaps/interactions. Experimental results on the Human Connectome Project (HCP) task fMRI data showed that our methods can reveal meaningful, diverse behavior patterns of connectome-scale network interactions. In particular, those networks' behavior patterns are distinct across HCP tasks such as motor, working memory, language and social tasks, and their dynamics well correspond to the temporal changes of specific task designs. In general, our framework offers a new approach to characterizing human brain function by quantitative description for the temporal evolution of spatial overlaps/interactions of connectome-scale brain networks in a standard reference space.

EEG sensorimotor rhythms' variation and functional connectivity measures during motor imagery: linear relations and classification approaches

Hands motor imagery (MI) has been reported to alter synchronization patterns amongst neurons, yielding variations in the mu and beta bands' power spectral density (PSD) of the electroencephalography (EEG) signal. These alterations have been used in the field of brain-computer interfaces (BCI), in an attempt to assign distinct MI tasks to commands of such a system. Recent studies have highlighted that information may be missing if knowledge about brain functional connectivity is not considered. In this work, we modeled the brain as a graph in which each EEG electrode represents a node. Our goal was to understand if there exists any linear correlation between variations in the synchronization patterns-that is, variations in the PSD of mu and beta bands-induced by MI and alterations in the corresponding functional networks. Moreover, we (1) explored the feasibility of using functional connectivity parameters as features for a classifier in the context of an MI-BCI; (2) investigated three different types of feature selection (FS) techniques; and (3) compared our approach to a more traditional method using the signal PSD as classifier inputs. Ten healthy subjects participated in this study. We observed significant correlations (p < 0.05) with values ranging from 0.4 to 0.9 between PSD variations and functional network alterations for some electrodes, prominently in the beta band. The PSD method performed better for data classification, with mean accuracies of (90 ± 8)% and (87 ± 7)% for the mu and beta band, respectively, versus (83 ± 8)% and (83 ± 7)% for the same bands for the graph method. Moreover, the number of features for the graph method was considerably larger. However, results for both methods were relatively close, and even overlapped when the uncertainties of the accuracy rates were considered. Further investigation regarding a careful exploration of other graph metrics may provide better alternatives.

Consistency of Regions of Interest as nodes of fMRI functional brain networks

The functional network approach, where fMRI BOLD time series are mapped to networks depicting functional relationships between brain areas, has opened new insights into the function of the human brain. In this approach, the choice of network nodes is of crucial importance. One option is to consider fMRI voxels as nodes. This results in a large number of nodes, making network analysis and interpretation of results challenging. A common alternative is to use predefined clusters of anatomically close voxels, Regions of Interest (ROIs). This approach assumes that voxels within ROIs are functionally similar. Because these two approaches result in different network structures, it is crucial to understand what happens to network connectivity when moving from the voxel level to the ROI level. We show that the consistency of ROIs, defined as the mean Pearson correlation coefficient between the time series of their voxels, varies widely in resting-state experimental data. Therefore the assumption of similar voxel dynamics within each ROI does not generally hold. Further, the time series of low-consistency ROIs may be highly correlated, resulting in spurious links in ROI-level networks. Based on these results, we recommend that averaging BOLD signals over anatomically defined ROIs should be carefully considered.

Network methods have opened new insights on structure and functional dynamics of the human brain. However, constructing functional brain networks is far from trivial—the neuroscientific community still lacks a standard definition of the nodes of brain networks. In the present article, we consider the two most commonly used approaches: using either imaging voxels or predefined Regions of Interest (ROIs) as nodes of the network. We investigate what happens when voxel-level signals are averaged for obtaining ROI-level networks. We introduce the concept of ROI consistency to characterize the similarity of the dynamics of voxels in an ROI. With the help of consistency, we show that although voxels in an ROI are assumed to behave similarly, this assumption does not hold for all ROIs.

Expertise-related functional brain network efficiency in healthy older adults

Background

In view of age-related brain changes, identifying factors that are associated with healthy aging are of great interest. In the present study, we compared the functional brain network characteristics of three groups of healthy older participants aged 61–75 years who had a different cognitive and motor training history (multi-domain group: participants who had participated in a multi-domain training; visuomotor group: participants who had participated in a visuomotor training; control group: participants with no specific training history). The study’s basic idea was to examine whether these different training histories are associated with differences in behavioral performance as well as with task-related functional brain network characteristics. Based on a high-density electroencephalographic measurement one year after training, we calculated graph-theoretical measures representing the efficiency of functional brain networks.

Results

Behaviorally, the multi-domain group performed significantly better than the visuomotor and the control groups on a multi-domain task including an inhibition domain, a visuomotor domain, and a spatial navigation domain. In terms of the functional brain network features, the multi-domain group showed significantly higher functional connectivity in a network encompassing visual, motor, executive, and memory-associated brain areas in the theta frequency band compared to the visuomotor group. These brain areas corresponded to the multi-domain task demands. Furthermore, mean connectivity of this network correlated positively with performance across both the multi-domain and the visuomotor group. In addition, the multi-domain group showed significantly enhanced processing efficiency reflected by a higher mean weighted node degree (strength) of the network as compared to the visuomotor group.

Conclusions

Taken together, our study shows expertise-dependent differences in task-related functional brain networks. These network differences were evident even a year after the acquisition of the different expertise levels. Hence, the current findings can foster understanding of how expertise is positively associated with brain functioning during aging.

Assessing the effects of cocaine dependence and pathological gambling using group-wise sparse representation of natural stimulus FMRI data

Assessing functional brain activation patterns in neuropsychiatric disorders such as cocaine dependence (CD) or pathological gambling (PG) under naturalistic stimuli has received rising interest in recent years. In this paper, we propose and apply a novel group-wise sparse representation framework to assess differences in neural responses to naturalistic stimuli across multiple groups of participants (healthy control, cocaine dependence, pathological gambling). Specifically, natural stimulus fMRI (N-fMRI) signals from all three groups of subjects are aggregated into a big data matrix, which is then decomposed into a common signal basis dictionary and associated weight coefficient matrices via an effective online dictionary learning and sparse coding method. The coefficient matrices associated with each common dictionary atom are statistically assessed for each group separately. With the inter-group comparisons based on the group-wise correspondence established by the common dictionary, our experimental results demonstrated that the group-wise sparse coding and representation strategy can effectively and specifically detect brain networks/regions affected by different pathological conditions of the brain under naturalistic stimuli.

A novel pattern mining approach for identifying cognitive activity in EEG based functional brain networks

The complex nature of neuronal interactions of the human brain has posed many challenges to the research community. To explore the underlying mechanisms of neuronal activity of cohesive brain regions during different cognitive activities, many innovative mathematical and computational models are required. This paper presents a novel Common Functional Pattern Mining approach to demonstrate the similar patterns of interactions due to common behavior of certain brain regions. The electrode sites of EEG-based functional brain network are modeled as a set of transactions and node-based complex network measures as itemsets. These itemsets are transformed into a graph data structure called Functional Pattern Graph. By mining this Functional Pattern Graph, the common functional patterns due to specific brain functioning can be identified. The empirical analyses show the efficiency of the proposed approach in identifying the extent to which the electrode sites (transactions) are similar during various cognitive load states.

Cognition and resting-state functional connectivity in schizophrenia

Individuals with schizophrenia consistently display deficits in a multitude of cognitive domains, but the neurobiological source of these cognitive impairments remains unclear. By analyzing the functional connectivity of resting-state functional magnetic resonance imaging (rs-fcMRI) data in clinical populations like schizophrenia, research groups have begun elucidating abnormalities in the intrinsic communication between specific brain regions, and assessing relationships between these abnormalities and cognitive performance in schizophrenia. Here we review studies that have reported analysis of these brain-behavior relationships. Through this systematic review we found that patients with schizophrenia display abnormalities within and between regions comprising (1) the cortico-cerebellar-striatal-thalamic loop and (2) task-positive and task-negative cortical networks. Importantly, we did not observe unique relationships between specific functional connectivity abnormalities and distinct cognitive domains, suggesting that the observed functional systems may underlie mechanisms that are shared across cognitive abilities, the disturbance of which could contribute to the "generalized" cognitive deficit found in schizophrenia. We also note several areas of methodological change that we believe will strengthen this literature.

Test-retest reliability of graph metrics of resting state MRI functional brain networks: A review

The employment of graph theory to analyze spontaneous fluctuations in resting state BOLD fMRI data has become a dominant theme in brain imaging studies and neuroscience. Analysis of resting state functional brain networks based on graph theory has proven to be a powerful tool to quantitatively characterize functional architecture of the brain and it has provided a new platform to explore the overall structure of local and global functional connectivity in the brain. Due to its increased use and possible expansion to clinical use, it is essential that the reliability of such a technique is very strongly assessed. In this review, we explore the outcome of recent studies in network reliability which apply graph theory to analyze connectome resting state networks. Therefore, we investigate which preprocessing steps may affect reproducibility the most. In order to investigate network reliability, we compared the test-retest (TRT) reliability of functional data of published neuroimaging studies with different preprocessing steps. In particular we tested influence of global signal regression, correlation metric choice, binary versus weighted link definition, frequency band selection and length of time-series. Statistical analysis shows that only frequency band selection and length of time-series seem to affect TRT reliability. Our results highlight the importance of the choice of the preprocessing steps to achieve more reproducible measurements.

Fronto-parietal and cingulo-opercular network integrity and cognition in health and schizophrenia

Growing evidence suggests that coordinated activity within specific functional brain networks supports cognitive ability, and that abnormalities in brain connectivity may underlie cognitive deficits observed in neuropsychiatric diseases, such as schizophrenia. Two functional networks, the fronto-parietal network (FPN) and cingulo-opercular network (CON), are hypothesized to support top-down control of executive functioning, and have therefore emerged as potential drivers of cognitive impairment in disease-states. Graph theoretic analyses of functional connectivity data can characterize network topology, allowing the relationships between cognitive ability and network integrity to be examined. In the current study we applied graph analysis to pseudo-resting state data in 54 healthy subjects and 46 schizophrenia patients, and measured overall cognitive ability as the shared variance in performance from tasks of episodic memory, verbal memory, processing speed, goal maintenance, and visual integration. We found that, across all participants, cognitive ability was significantly positively associated with the local and global efficiency of the whole brain, FPN, and CON, but not with the efficiency of a comparison network, the auditory network. Additionally, the participation coefficient of the right anterior insula, a major hub within the CON, significantly predicted cognition, and this relationship was independent of CON global efficiency. Surprisingly, we did not observe strong evidence for group differences in any of our network metrics. These data suggest that functionally efficient task control networks support better cognitive ability in both health and schizophrenia, and that the right anterior insula may be a particularly important hub for successful cognitive performance across both health and disease.

Accurate age classification of 6 and 12 month-old infants based on resting-state functional connectivity magnetic resonance imaging data

•

SVMs classified 6 versus 12 month-old infants above chance based on fcMRI data alone.

•

We carefully accounted for the effects of fcMRI motion artifact.

•

These results coincide with a period of dramatic change in infant development.

•

Two interpretations about connections supporting this age categorization are given.

Human large-scale functional brain networks are hypothesized to undergo significant changes over development. Little is known about these functional architectural changes, particularly during the second half of the first year of life. We used multivariate pattern classification of resting-state functional connectivity magnetic resonance imaging (fcMRI) data obtained in an on-going, multi-site, longitudinal study of brain and behavioral development to explore whether fcMRI data contained information sufficient to classify infant age. Analyses carefully account for the effects of fcMRI motion artifact. Support vector machines (SVMs) classified 6 versus 12 month-old infants (128 datasets) above chance based on fcMRI data alone. Results demonstrate significant changes in measures of brain functional organization that coincide with a special period of dramatic change in infant motor, cognitive, and social development. Explorations of the most different correlations used for SVM lead to two different interpretations about functional connections that support 6 versus 12-month age categorization.

Identifying functional co-activation patterns in neuroimaging studies via poisson graphical models

Studying the interactions between different brain regions is essential to achieve a more complete understanding of brain function. In this article, we focus on identifying functional co-activation patterns and undirected functional networks in neuroimaging studies. We build a functional brain network, using a sparse covariance matrix, with elements representing associations between region-level peak activations. We adopt a penalized likelihood approach to impose sparsity on the covariance matrix based on an extended multivariate Poisson model. We obtain penalized maximum likelihood estimates via the expectation-maximization (EM) algorithm and optimize an associated tuning parameter by maximizing the predictive log-likelihood. Permutation tests on the brain co-activation patterns provide region pair and network-level inference. Simulations suggest that the proposed approach has minimal biases and provides a coverage rate close to 95% of covariance estimations. Conducting a meta-analysis of 162 functional neuroimaging studies on emotions, our model identifies a functional network that consists of connected regions within the basal ganglia, limbic system, and other emotion-related brain regions. We characterize this network through statistical inference on region-pair connections as well as by graph measures.

Dynamic functional brain networks involved in simple visual discrimination learning

Visual discrimination tasks have been widely used to evaluate many types of learning and memory processes. However, little is known about the brain regions involved at different stages of visual discrimination learning. We used cytochrome c oxidase histochemistry to evaluate changes in regional brain oxidative metabolism during visual discrimination learning in a water-T maze at different time points during training. As compared with control groups, the results of the present study reveal the gradual activation of cortical (prefrontal and temporal cortices) and subcortical brain regions (including the striatum and the hippocampus) associated to the mastery of a simple visual discrimination task. On the other hand, the brain regions involved and their functional interactions changed progressively over days of training. Regions associated with novelty, emotion, visuo-spatial orientation and motor aspects of the behavioral task seem to be relevant during the earlier phase of training, whereas a brain network comprising the prefrontal cortex was found along the whole learning process. This study highlights the relevance of functional interactions among brain regions to investigate learning and memory processes.

Networks of task co-activations

Recent progress in neuroimaging informatics and meta-analytic techniques has enabled a novel domain of human brain connectomics research that focuses on task-dependent co-activation patterns across behavioral tasks and cognitive domains. Here, we review studies utilizing the BrainMap database to investigate data trends in the activation literature using methods such as meta-analytic connectivity modeling (MACM), connectivity-based parcellation (CPB), and independent component analysis (ICA). We give examples of how these methods are being applied to learn more about the functional connectivity of areas such as the amygdala, the default mode network, and visual area V5. Methods for analyzing the behavioral metadata corresponding to regions of interest and to their intrinsically connected networks are described as a tool for local functional decoding. We finally discuss the relation of observed co-activation connectivity results to resting state connectivity patterns, and provide implications for future work in this domain.