A review of group ICA for fMRI data and ICA for joint inference of imaging, genetic, and ERP data

Regions and Connections: Complementary Approaches to Characterize Brain Organization and Function

Functional magnetic resonance imaging has proved to be a powerful tool to characterize spatiotemporal patterns of human brain activity. Analysis methods broadly fall into two camps: those summarizing properties of a region and those measuring interactions among regions. Here we pose an unappreciated question in the field: What are the strengths and limitations of each approach to study fundamental neural processes? We explore the relative utility of region- and connection-based measures in the context of three topics of interest: neurobiological relevance, brain-behavior relationships, and individual differences in brain organization. In each section, we offer illustrative examples. We hope that this discussion offers a novel and useful framework to support efforts to better understand the macroscale functional organization of the brain and how it relates to behavior.

Canonical maximization of coherence: A novel tool for investigation of neuronal interactions between two datasets

Synchronization between oscillatory signals is considered to be one of the main mechanisms through which neuronal populations interact with each other. It is conventionally studied with mass-bivariate measures utilizing either sensor-to-sensor or voxel-to-voxel signals. However, none of these approaches aims at maximizing synchronization, especially when two multichannel datasets are present. Examples include cortico-muscular coherence (CMC), cortico-subcortical interactions or hyperscanning (where electroencephalographic EEG/magnetoencephalographic MEG activity is recorded simultaneously from two or more subjects). For all of these cases, a method which could find two spatial projections maximizing the strength of synchronization would be desirable. Here we present such method for the maximization of coherence between two sets of EEG/MEG/EMG (electromyographic)/LFP (local field potential) recordings. We refer to it as canonical Coherence (caCOH). caCOH maximizes the absolute value of the coherence between the two multivariate spaces in the frequency domain. This allows very fast optimization for many frequency bins. Apart from presenting details of the caCOH algorithm, we test its efficacy with simulations using realistic head modelling and focus on the application of caCOH to the detection of cortico-muscular coherence. For this, we used diverse multichannel EEG and EMG recordings and demonstrate the ability of caCOH to extract complex patterns of CMC distributed across spatial and frequency domains. Finally, we indicate other scenarios where caCOH can be used for the extraction of neuronal interactions.

Tracking the Main States of Dynamic Functional Connectivity in Resting State

Dynamical changes have recently been tracked in functional connectivity (FC) calculated from resting-state functional magnetic resonance imaging (R-fMRI), when a person is conscious but not carrying out a directed task during scanning. Diverse dynamical FC states (dFC) are believed to represent different internal states of the brain, in terms of brain-regional interactions. In this paper, we propose a novel protocol, the signed community clustering with the optimized modularity by two-step procedures, to track dynamical whole brain functional connectivity (dWFC) states. This protocol is assumption free without a priori threshold for the number of clusters. By applying our method on sliding window based dWFC’s with automated anatomical labeling 2 (AAL2), three main dWFC states were extracted from R-fMRI datasets in Human Connectome Project, that are independent on window size. Through extracting the FC features of these states, we found the functional links in state 1 (WFC-C₁) mainly involved visual, somatomotor, attention and cerebellar (posterior lobe) modules. State 2 (WFC-C₂) was similar to WFC-C₁, but more FC’s linking limbic, default mode, and frontoparietal modules and less linking the cerebellum, sensory and attention modules. State 3 had more FC’s linking default mode, limbic, and cerebellum, compared to WFC-C₁ and WFC-C₂. With tests of robustness and stability, our work provides a solid, hypothesis-free tool to detect dWFC states for the possibility of tracking rapid dynamical change in FCs among large data sets.

Brain networks involved in accented speech processing

We investigated the neural correlates of accented speech processing (ASP) with an fMRI study that overcame prior limitations in this line of research: we preserved intelligibility by using two regional accents that differ in prosody but only mildly in phonetics (Latin American and Castilian Spanish), and we used independent component analysis to identify brain networks as opposed to isolated regions. ASP engaged a speech perception network composed primarily of structures related with the processing of prosody (cerebellum, putamen, and thalamus). This network also included anterior fronto-temporal areas associated with lexical-semantic processing and a portion of the inferior frontal gyrus linked to executive control. ASP also recruited domain-general executive control networks related with cognitive demands (dorsal attentional and default mode networks) and the processing of salient events (salience network). Finally, the reward network showed a preference for the native accent, presumably revealing people's sense of social belonging.

Exploring the supra linear relationship between PetCO2 and fMRI signal change with ICA

The relationships between brain functions and the respiratory system are complex. Disentangling brain activity related to CO₂ changes from nonspecific vasoreactivity is a challenge when studying brain activity involved in the control of breathing with fMRI. In this work, we analyzed a dose dependent relationship between arterial CO₂ levels and brain response. To accomplish this goal, we developed a gas administration protocol, together with multi-subject ICA and specific nonlinear post-processing analysis. Our results highlighted a supra-linear response to CO₂ challenges in brainstem, thalamus and putamen. Results were discussed in the light of current knowledge about the central respiratory network.

Selection of Efficient Clustering Index to Estimate the Number of Dynamic Brain States from Functional Network Connectivity

Clustering analysis is employed in brain dynamic functional connectivity to cluster the data into a set of dynamic states. These states correspond to different patterns of functional connectivity that iterate through time. Although several methods to determine the best clustering partition exists, the appropriateness of methods to apply in the case of dynamic connectivity analysis has not been determined. In this work we examine the use of the Davies-Bouldin clustering validity index via simulation and real data analysis. Currently employed indexes, such as the Silhouette index, do not provide an effective estimation requiring the use of an elbow criterion. All elbow criteria rely on users experience and introduce uncertainty into the estimation. We demonstrate the feasibility of using the Davies-Bouldin index as a method delivering a unique discrete response to provide automated selection of the number of clusters.

Spatial Topography of Individual-Specific Cortical Networks Predicts Human Cognition, Personality, and Emotion

Resting-state functional magnetic resonance imaging (rs-fMRI) offers the opportunity to delineate individual-specific brain networks. A major question is whether individual-specific network topography (i.e., location and spatial arrangement) is behaviorally relevant. Here, we propose a multi-session hierarchical Bayesian model (MS-HBM) for estimating individual-specific cortical networks and investigate whether individual-specific network topography can predict human behavior. The multiple layers of the MS-HBM explicitly differentiate intra-subject (within-subject) from inter-subject (between-subject) network variability. By ignoring intra-subject variability, previous network mappings might confuse intra-subject variability for inter-subject differences. Compared with other approaches, MS-HBM parcellations generalized better to new rs-fMRI and task-fMRI data from the same subjects. More specifically, MS-HBM parcellations estimated from a single rs-fMRI session (10 min) showed comparable generalizability as parcellations estimated by 2 state-of-the-art methods using 5 sessions (50 min). We also showed that behavioral phenotypes across cognition, personality, and emotion could be predicted by individual-specific network topography with modest accuracy, comparable to previous reports predicting phenotypes based on connectivity strength. Network topography estimated by MS-HBM was more effective for behavioral prediction than network size, as well as network topography estimated by other parcellation approaches. Thus, similar to connectivity strength, individual-specific network topography might also serve as a fingerprint of human behavior.

Unraveling Diagnostic Biomarkers of Schizophrenia Through Structure-Revealing Fusion of Multi-Modal Neuroimaging Data

Fusing complementary information from different modalities can lead to the discovery of more accurate diagnostic biomarkers for psychiatric disorders. However, biomarker discovery through data fusion is challenging since it requires extracting interpretable and reproducible patterns from data sets, consisting of shared/unshared patterns and of different orders. For example, multi-channel electroencephalography (EEG) signals from multiple subjects can be represented as a third-order tensor with modes: subject, time, and channel, while functional magnetic resonance imaging (fMRI) data may be in the form of subject by voxel matrices. Traditional data fusion methods rearrange higher-order tensors, such as EEG, as matrices to use matrix factorization-based approaches. In contrast, fusion methods based on coupled matrix and tensor factorizations (CMTF) exploit the potential multi-way structure of higher-order tensors. The CMTF approach has been shown to capture underlying patterns more accurately without imposing strong constraints on the latent neural patterns, i.e., biomarkers. In this paper, EEG, fMRI, and structural MRI (sMRI) data collected during an auditory oddball task (AOD) from a group of subjects consisting of patients with schizophrenia and healthy controls, are arranged as matrices and higher-order tensors coupled along the subject mode, and jointly analyzed using structure-revealing CMTF methods [also known as advanced CMTF (ACMTF)] focusing on unique identification of underlying patterns in the presence of shared/unshared patterns. We demonstrate that joint analysis of the EEG tensor and fMRI matrix using ACMTF reveals significant and biologically meaningful components in terms of differentiating between patients with schizophrenia and healthy controls while also providing spatial patterns with high resolution and improving the clustering performance compared to the analysis of only the EEG tensor. We also show that these patterns are reproducible, and study reproducibility for different model parameters. In comparison to the joint independent component analysis (jICA) data fusion approach, ACMTF provides easier interpretation of EEG data by revealing a single summary map of the topography for each component. Furthermore, fusion of sMRI data with EEG and fMRI through an ACMTF model provides structural patterns; however, we also show that when fusing data sets from multiple modalities, hence of very different nature, preprocessing plays a crucial role.

The Translational Potential of Neuroimaging Genomic Analyses To Diagnosis And Treatment In The Mental Disorders

Imaging genomics focuses on characterizing genomic influence on the variation of neurobiological traits, holding promise for illuminating the pathogenesis, reforming the diagnostic system, and precision medicine of mental disorders. This paper aims to provide an overall picture of the current status of neuroimaging-genomic analyses in mental disorders, and how we can increase their translational potential into clinical practice. The review is organized around three perspectives. (a) Towards reliability, generalizability and interpretability, where we summarize the multivariate models and discuss the considerations and trade-offs of using these methods and how reliable findings may be reached, to serve as ground for further delineation. (b) Towards improved diagnosis, where we outline the advantages and challenges of constructing a dimensional transdiagnostic model and how imaging genomic analyses map into this framework to aid in deconstructing heterogeneity and achieving an optimal stratification of patients that better inform treatment planning. (c) Towards improved treatment. Here we highlight recent efforts and progress in elucidating the functional annotations that bridge between genomic risk and neurobiological abnormalities, in detecting genomic predisposition and prodromal neurodevelopmental changes, as well as in identifying imaging genomic biomarkers for predicting treatment response. Providing an overview of the challenges and promises, this review hopefully motivates imaging genomic studies with multivariate, dimensional and transdiagnostic designs for generalizable and interpretable findings that facilitate development of personalized treatment.

Discovering hierarchical common brain networks via multimodal deep belief network

Studying a common architecture reflecting both brain's structural and functional organizations across individuals and populations in a hierarchical way has been of significant interest in the brain mapping field. Recently, deep learning models exhibited ability in extracting meaningful hierarchical structures from brain imaging data, e.g., fMRI and DTI. However, deep learning models have been rarely used to explore the relation between brain structure and function yet. In this paper, we proposed a novel multimodal deep believe network (DBN) model to discover and quantitatively represent the hierarchical organizations of common and consistent brain networks from both fMRI and DTI data. A prominent characteristic of DBN is that it is capable of extracting meaningful features from complex neuroimaging data with a hierarchical manner. With our proposed DBN model, three hierarchical layers with hundreds of common and consistent brain networks across individual brains are successfully constructed through learning a large dimension of representative features from fMRI/DTI data.

Mindfulness-based therapy modulates default-mode network connectivity in patients with opioid dependence

Recently, mindfulness-based programs have shown promising clinical effects in the treatment of substance-use disorders (SUD). While several studies linked mindfulness to decreased default mode network (DMN) connectivity in meditators, only a few studies investigated its effects in patients with SUD. This study aimed to detect changes in DMN connectivity in opiate dependent patients receiving mindfulness based therapy (MBT) during their first month of treatment. Data from 32 patients that were assigned to MBT or treatment as usual (TAU) groups was investigated using resting-state functional MRI at 1.5 T before and after four weeks of treatment. Independent Component Analysis was used to investigate distinct (anterior vs. posterior) DMN subsystems. Connectivity changes after treatment were related to measures of impulsivity, distress tolerance and mindfulness. Increased mindfulness scores after treatment were found in patients receiving MBT compared to TAU. Within the anterior DMN, decreased right inferior frontal cortical connectivity was detected in patients who received MBT compared to TAU. In addition, within the MBT-group decreased right superior frontal cortex connectivity was detected after treatment. Inferior frontal cortex function was significantly associated with mindfulness measures. The data suggest that MBT can be useful during abstinence from opiates. In opiate-dependent patients distinct functional connectivity changes within the DMN are associated with MBT.

Bottom-up sensory processing can induce negative BOLD responses and reduce functional connectivity in nodes of the default mode-like network in rats

The default mode network is a large-scale brain network that is active during rest and internally focused states and deactivates as well as desynchronizes during externally oriented (top-down) attention demanding cognitive tasks. However, it is not sufficiently understood if salient stimuli, able to trigger bottom-up attentional processes, could also result in similar reduction of activity and functional connectivity in the DMN. In this study, we investigated whether bottom-up sensory processing could influence the default mode-like network (DMLN) in rats. DMLN activity was examined using block-design visual functional magnetic resonance imaging (fMRI) while its synchronization was investigated by comparing functional connectivity during a resting versus a continuously stimulated brain state by unpredicted light flashes. We demonstrated that the BOLD response in DMLN regions was decreased during visual stimulus blocks and increased during blanks. Furthermore, decreased inter-network functional connectivity between the DMLN and visual networks as well as decreased intra-network functional connectivity within the DMLN was observed during the continuous visual stimulation. These results suggest that triggering of bottom-up attention mechanisms in sedated rats can lead to a cascade similar to top-down orienting of attention in humans and is able to deactivate and desynchronize the DMLN.

Classification of cocaine-dependent participants with dynamic functional connectivity from functional magnetic resonance imaging data

Static functional connectivity (FC) analyses based on functional magnetic resonance imaging (fMRI) data have been extensively explored for studying various psychiatric conditions in the brain, including cocaine addiction. A recently emerging, more powerful technique, dynamic functional connectivity (DFC), studies how the FC dynamics change during the course of the fMRI experiments. The aim in this paper was to develop a computational approach, using a machine learning framework, to determine if DFC features were more successful than FC features in the classification of cocaine-dependent patients and healthy controls. fMRI data were obtained from of 25 healthy and 58 cocaine-dependent participants while performing a motor response inhibition task, stop signal task. Group independent component analysis was carried out on all participant data to compute spatially independent components (ICs). Eight ICs were selected manually as relevant brain networks, which were used to classify healthy versus cocaine-dependent participants. FC and DFC measures of the chosen IC pairs were used as features for the classification algorithm. Support Vector Machines were used for both feature selection/reduction and participant classification. Based on DFC with only seven IC pairs, participants were successfully classified with 95% accuracy (and with 90% accuracy with three IC pairs), whereas static FC yielded only 81% accuracy. Visual, sensorimotor, default mode, and executive control networks, amygdala, and insula played the most significant role in the DFC-based classification. These findings support the use of DFC-based classification of fMRI data as a potential biomarker for the identification of cocaine dependence.

Network-level dysconnectivity in patients with nasopharyngeal carcinoma (NPC) early post-radiotherapy: longitudinal resting state fMRI study

In this study, we seek to longitudinally investigate the network-level functional connectivity (FC) alternations and its association with irradiation dose and cognition changes in the early stage post radiotherapy (RT) in nasopharyngeal carcinoma (NPC) patients. We performed independent component analysis (ICA) of resting state blood oxygen level-dependent functional magnetic resonance imaging (BOLD-fMRI) from 39 newly diagnosed NPC patients before receiving treatment (baseline), and 3 months post-RT. the default mode network (DMN), salience network (SN), and executive control network (ECN) were extracted with well-validated software (GIFT). Inter-network connectivity was assessed using the functional network connectivity (FNC) toolbox. The inter- and intra-network FC was compared between time points, and the z value of FC alternation was correlated with the RT dose value and cognitive changes. Compared with baseline, the FC of the left anterior cingulate cortex (ACC) within the DMN, and the right insular within the SN, significantly reduced 3 months post-RT, with greater effects at higher doses in the right insular. Bilateral ECN FC was also significantly lower 3 months post-RT compared to the baseline. Chemotherapy was not associated with inter- and intra- network FC change. We found intra- and inter-network FC disruption in NPC patients 3 months post-RT, with the right insular showing a dose-dependent effect. Thus, this network-level FC may serve as a potential biomarker of the RT-induced brain functional impairments, and provide valuable targets for further functional recovery treatment.

Phase fMRI informs whole-brain function connectivity balance across lifespan with connection-specific aging effects during the resting state

A functional magnetic resonance imaging (fMRI) experiment produces complex-valued images consisting of pairwise magnitude and phase images. As different perspective on the same magnetic source, fMRI magnitude and phase data are complementary for brain function analysis. We collected 600-subject fMRI data during rest, decomposed via group-level independent component analysis (ICA) (mICA and pICA for magnitude and phase respectively), and calculated brain functional network connectivity matrices (mFC and pFC). The pFC matrix shows a fewer of significant connections balanced across positive and negative relationships. In comparison, the mFC matrix contains a positively-biased pattern with more significant connections. Our experiment data analyses also show that human brain maintains a whole-brain connection balance in resting state across an age span from 10 to 76 years, however, phase and magnitude data analyses reveal different connection-specific age effects on significant positive and negative subnetwork couplings.

Add-on atypical anti-psychotic treatment alleviates auditory verbal hallucinations in patients with chronic post-traumatic stress disorder

Auditory verbal hallucinations are common symptoms of post traumatic distress disorder. Previous studies have demonstrated alterations in the salience network (SN) in patients with post traumatic distress disorder and that hyperactivity of the SN is associated with AVHs in patients with psychosis. Patients with post traumatic distress disorder may benefit from aripiprazole; however, studies investigating the effect of aripiprazole on AVHs and activity in the SN in patients with post traumatic distress disorder are scarce. Therefore, we conducted an outcomes analysis using functional magnetic resonance imaging to explore the effects of add-on aripiprazole treatment on AVHs and brain functional connectivity in patients with post traumatic distress disorder. AVHs were alleviated by add-on aripiprazole treatment (Auditory Hallucination Rating Scale [AHRS] score reduced by ≥ 50%) in 22.7% of patients. Functional activity in the SN was obviously decreased in patients in whom AHRS scores were reduced ≥ 50% following add-on aripiprazole treatment compared to patients in whom AHRS scores were reduced by <50%. The decrease in functional connectivity within the SN was significantly correlated with the reduction in total AHRS scores. Although this study was associated with several limitations, the findings suggest that add-on aripiprazole treatment can alleviate AVHs in patients with post traumatic distress disorder by reducing activity in the SN.

Using resting-state DMN effective connectivity to characterize the neurofunctional architecture of empathy

Neuroimaging studies in social neuroscience have largely relied on functional connectivity (FC) methods to characterize the functional integration between different brain regions. However, these methods have limited utility in social-cognitive studies that aim to understand the directed information flow among brain areas that underlies complex psychological processes. In this study we combined functional and effective connectivity approaches to characterize the functional integration within the Default Mode Network (DMN) and its role in self-perceived empathy. Forty-two participants underwent a resting state fMRI scan and completed a questionnaire of dyadic empathy. Independent Component Analysis (ICA) showed that higher empathy scores were associated with an increased contribution of the medial prefrontal cortex (mPFC) to the DMN spatial mode. Dynamic causal modelling (DCM) combined with Canonical Variance Analysis (CVA) revealed that this association was mediated indirectly by the posterior cingulate cortex (PCC) via the right inferior parietal lobule (IPL). More specifically, in participants with higher scores in empathy, the PCC had a greater effect on bilateral IPL and the right IPL had a greater influence on mPFC. These results highlight the importance of using analytic approaches that address directed and hierarchical connectivity within networks, when studying complex psychological phenomena, such as empathy.

Large-Scale Brain Networks Underlying Successful and Unsuccessful Encoding, Maintenance, and Retrieval of Everyday Scenes in Visuospatial Working Memory

Recent research on working memory (WM) identified the contribution of several large-scale brain networks operating during WM tasks, such as the frontoparietal attention network (AN), the default mode network (DMN), and the salience network (SN). To date, however, the dynamical interplay among these networks is largely unexplored during successful or unsuccessful WM performance, especially with complex and ecological stimuli. Here we systematically characterized the selective contribution of these networks during a visuospatial WM task requiring the encoding, maintenance and retrieval of real-life scenes. While undergoing fMRI scans, participants were presented with everyday life visual scenes for 4 s (encoding phase). After a delay of 8 s (maintenance phase), participants were presented with a target-object extracted from the previous scene. Participants had to judge whether the target-object was presented at the same or in a different location compared to the original scene (retrieval phase) and then provide a confidence judgment. Using the independent component analysis (ICA), we found that subsequent remembering was associated with the activity of the AN at encoding, the attention and SN at maintenance, plus the visual network at retrieval. Conversely, subsequent forgetting was associated with the activity of the DMN at maintenance, and the SN at retrieval. Overall, these findings reveal a dynamical interplay between large-scale brain networks during visuospatial WM performance related to complex, real-life stimuli.

FuNP (Fusion of Neuroimaging Preprocessing) Pipelines: A Fully Automated Preprocessing Software for Functional Magnetic Resonance Imaging

The preprocessing of functional magnetic resonance imaging (fMRI) data is necessary to remove unwanted artifacts and transform the data into a standard format. There are several neuroimaging data processing tools that are widely used, such as SPM, AFNI, FSL, FreeSurfer, Workbench, and fMRIPrep. Different data preprocessing pipelines yield differing results, which might reduce the reproducibility of neuroimaging studies. Here, we developed a preprocessing pipeline for T1-weighted structural MRI and fMRI data by combining components of well-known software packages to fully incorporate recent developments in MRI preprocessing into a single coherent software package. The developed software, called FuNP (Fusion of Neuroimaging Preprocessing) pipelines, is fully automatic and provides both volume- and surface-based preprocessing pipelines with a user-friendly graphical interface. The reliability of the software was assessed by comparing resting-state networks (RSNs) obtained using FuNP with pre-defined RSNs using open research data (n = 90). The obtained RSNs were well-matched with the pre-defined RSNs, suggesting that the pipelines in FuNP are reliable. In addition, image quality metrics (IQMs) were calculated from the results of three different software packages (i.e., FuNP, FSL, and fMRIPrep) to compare the quality of the preprocessed data. We found that our FuNP outperformed other software in terms of temporal characteristics and artifacts removal. We validated our pipeline with independent local data (n = 28) in terms of IQMs. The IQMs of our local data were similar to those obtained from the open research data. The codes for FuNP are available online to help researchers.

Dysconnectivity of Multiple Resting-State Networks Associated With Higher-Order Functions in Sensorineural Hearing Loss

Objects: Sensorineural hearing loss (SNHL) involves wide-ranging functional reorganization, and is associated with accumulating risk of cognitive and emotional dysfunction. The coordination of multiple functional networks supports normal brain functions. Here, we aimed to evaluate the functional connectivity (FC) patterns involving multiple resting-state networks (RSNs), and the correlations between the functional remodeling of RSNs and the potential cognitive or emotional impairments in SNHL.

Methods: Thirty long-term bilateral SNHL patients and 39 well-matched healthy controls were recruited for assessment of resting-state functional magnetic resonance imaging and neuropsychological tests.

Results: Using independent component analysis, 11 RSNs were identified. Relative to the healthy controls, patients with SNHL presented apparent abnormalities of intra-network FC involving right frontoparietal network, posterior temporal network, and sensory motor network. Disrupted between-network FC was also revealed in the SNHL patients across both higher-order cognitive control networks and multiple sensory networks. Eight of the eleven RSNs showed altered functional synchronization using a seed network to whole brain FC method, particularly in the ventromedial prefrontal cortex. In addition, these functional abnormalities were correlated with cognition- and emotion-related performances.

Interpretations: These findings supported our hypotheses that long-term SNHL involves notable dysconnectivity of multiple RSNs. Our study provides important insights into the pathophysiological mechanisms of SNHL, and sheds lights on the neural substrates underlying the possible cognitive and emotional dysfunctions following SNHL.

Positron emission tomography of tau in Iraq and Afghanistan Veterans with blast neurotrauma

Military personnel are often exposed to multiple instances of various types of head trauma. As a result, there has been increasing concern recently over identifying when head trauma has resulted in a brain injury and what, if any, long-term consequences those brain injuries may have. Efforts to develop equipment to protect soldiers from these long-term consequences will first require understanding the types of head trauma that are likely responsible. In this study, we sought to identify the types of head trauma most likely to lead to the deposition of tau, a protein identified as a likely indicator of long-term negative consequences of brain injury. To define the types of head trauma in a military population, we applied a factor analysis to interviews from a larger cohort of 428 Veterans enrolled in the Translational Research Center for Traumatic Brain Injury and Stress Disorders. Three factors were identified: Blast Exposure, Symptom Duration, and Blunt Concussion. Sixteen male Veterans from this study and one additional male civilian (aged 25-69, mean 35.2 years) underwent simultaneous positron emission tomography/magnetic resonance imaging using a tracer that binds to tau protein, the ligand T807/AV-1451 (Flortaucipir). Standard uptake value ratios to the isthmus of the cingulate were calculated from a 20-minute time frame 70 min post-injection. We found that tracer uptake throughout the brain was associated with Blast Exposure factor beta weights, but not with either Symptom Duration or Blunt Concussion. Associations with uptake were located primarily in the cerebellar, occipital, inferior temporal and frontal regions. The data suggest that in this small, relatively young cohort of Veterans, elevated T807/AV-1451 uptake is associated with exposure to blast neurotrauma. These findings are unanticipated, as they do not match histopathological descriptions of tau pathology associated with head trauma. Continued work will be necessary to understand the nature of the regional T807/AV-1451 uptake and any associations with clinical symptoms.

State-Dependent Effects of Ventromedial Prefrontal Cortex Continuous Thetaburst Stimulation on Cocaine Cue Reactivity in Chronic Cocaine Users

Cue-induced craving is a significant barrier to obtaining abstinence from cocaine. Neuroimaging research has shown that cocaine cue exposure evokes elevated activity in a network of frontal-striatal brain regions involved in drug craving and drug seeking. Prior research from our laboratory has demonstrated that when targeted at the medial prefrontal cortex (mPFC), continuous theta burst stimulation (cTBS), an inhibitory form of non-invasive brain stimulation, can decrease drug cue-related activity in the striatum in cocaine users and alcohol users. However, it is known that there are individual differences in response to repetitive transcranial magnetic stimulation (rTMS), with some individuals being responders and others non-responders. There is some evidence that state-dependent effects influence response to rTMS, with baseline neural state predicting rTMS treatment outcomes. In this single-blind, active sham-controlled crossover study, we assess the striatum as a biomarker of treatment response by determining if baseline drug cue reactivity in the striatum influences striatal response to mPFC cTBS. The brain response to cocaine cues was measured in 19 cocaine-dependent individuals immediately before and after real and sham cTBS (110% resting motor threshold, 3600 total pulses). Group independent component analysis (ICA) revealed a prominent striatum network comprised of bilateral caudate, putamen, and nucleus accumbens, which was modulated by the cocaine cue reactivity task. Baseline drug cue reactivity in this striatal network was inversely related to change in striatum reactivity after real (vs. sham) cTBS treatment (ρ = -.79; p < .001; R ² _Adj = .58). Specifically, individuals with a high striatal response to cocaine cues at baseline had significantly attenuated striatal activity after real but not sham cTBS (t ₉ = -3.76; p ≤ .005). These data demonstrate that the effects of mPFC cTBS on the neural circuitry of craving are not uniform and may depend on an individual's baseline frontal-striatal reactivity to cues. This underscores the importance of assessing individual variability as we develop brain stimulation treatments for addiction.

Classifying early and late mild cognitive impairment stages of Alzheimer's disease by fusing default mode networks extracted with multiple seeds

BACKGROUND

The default mode network (DMN) in resting state has been increasingly used in disease diagnosis since it was found in 2001. Prior work has mainly focused on extracting a single DMN with various techniques. However, by using seeding-based analysis with more than one desirable seed, we can obtain multiple DMNs, which are likely to have complementary information, and thus are more promising for disease diagnosis. In the study, we used 18 early mild cognitive impairment (EMCI) participants and 18 late mild cognitive impairment (LMCI) participants of Alzheimer's disease (AD). First, we used seeding-based analysis with four seeds to extract four DMNs for each subject. Then, we conducted fusion analysis for all different combinations of the four DMNs. Finally, we carried out nonlinear support vector machine classification based on the mixing coefficients from the fusion analysis.

RESULTS

We found that (1) the four DMNs corresponding to the four different seeds indeed capture different functional regions of each subject; (2) Maps of the four DMNs in the most different joint source from fusion analysis are centered at the regions of the corresponding seeds; (3) Classification results reveal the effectiveness of using multiple seeds to extract DMNs. When using a single seed, the regions of posterior cingulate cortex (PCC) extractions of EMCI and LMCI show the largest difference. For multiple-seed cases, the regions of PCC extraction and right lateral parietal cortex (RLP) extraction provide complementary information for each other in fusion, which improves the classification accuracy. Furthermore, the regions of left lateral parietal cortex (LLP) extraction and RLP extraction also have complementary effect in fusion. In summary, AD diagnosis can be improved by exploiting complementary information of DMNs extracted with multiple seeds.

CONCLUSIONS

In this study, we applied fusion analysis to the DMNs extracted by using different seeds for exploiting the complementary information hidden among the separately extracted DMNs, and the results supported our expectation that using the complementary information can improve classification accuracy.

Dynamic Functional Magnetic Resonance Imaging Connectivity Tensor Decomposition: A New Approach to Analyze and Interpret Dynamic Brain Connectivity

There is a growing interest in using so-called dynamic functional connectivity, as the conventional static brain connectivity models are being questioned. Brain network analyses yield complex network data that are difficult to analyze and interpret. To deal with the complex structures, decomposition/factorization techniques that simplify the data are often used. For dynamic network analyses, data simplification is of even greater importance, as dynamic connectivity analyses result in a time series of complex networks. A new challenge that must be faced when using these decomposition/factorization techniques is how to interpret the resulting connectivity patterns. Connectivity patterns resulting from decomposition analyses are often visualized as networks in brain space, in the same way that pairwise correlation networks are visualized. This elevates the risk of conflating connections between nodes that represent correlations between nodes' time series with connections between nodes that result from decomposition analyses. Moreover, dynamic connectivity data may be represented with three-dimensional or four-dimensional (4D) tensors and decomposition results require unique interpretations. Thus, the primary goal of this article is to (1) address the issues that must be considered when interpreting the connectivity patterns from decomposition techniques and (2) show how the data structure and decomposition method interact to affect this interpretation. The outcome of our analyses is summarized as follows. (1) The edge strength in decomposition connectivity patterns represents complex relationships not pairwise interactions between the nodes. (2) The structure of the data significantly alters the connectivity patterns, for example, 4D data result in connectivity patterns with higher regional connections. (3) Orthogonal decomposition methods outperform in feature reduction applications, whereas nonorthogonal decomposition methods are better for mechanistic interpretation.

Spatial dynamics within and between brain functional domains: A hierarchical approach to study time-varying brain function

The analysis of time-varying activity and connectivity patterns (i.e., the chronnectome) using resting-state magnetic resonance imaging has become an important part of ongoing neuroscience discussions. The majority of previous work has focused on variations of temporal coupling among fixed spatial nodes or transition of the dominant activity/connectivity pattern over time. Here, we introduce an approach to capture spatial dynamics within functional domains (FDs), as well as temporal dynamics within and between FDs. The approach models the brain as a hierarchical functional architecture with different levels of granularity, where lower levels have higher functional homogeneity and less dynamic behavior and higher levels have less homogeneity and more dynamic behavior. First, a high-order spatial independent component analysis is used to approximate functional units. A functional unit is a pattern of regions with very similar functional activity over time. Next, functional units are used to construct FDs. Finally, functional modules (FMs) are calculated from FDs, providing an overall view of brain dynamics. Results highlight the spatial fluidity within FDs, including a broad spectrum of changes in regional associations, from strong coupling to complete decoupling. Moreover, FMs capture the dynamic interplay between FDs. Patients with schizophrenia show transient reductions in functional activity and state connectivity across several FDs, particularly the subcortical domain. Activity and connectivity differences convey unique information in many cases (e.g., the default mode) highlighting their complementarity information. The proposed hierarchical model to capture FD spatiotemporal variations provides new insight into the macroscale chronnectome and identifies changes hidden from existing approaches.

Multivariate Deep Learning Classification of Alzheimer's Disease Based on Hierarchical Partner Matching Independent Component Analysis

Machine learning and pattern recognition have been widely investigated in order to look for the biomarkers of Alzheimer’s disease (AD). However, most existing methods extract features by seed-based correlation, which not only requires prior information but also ignores the relationship between resting state functional magnetic resonance imaging (rs-fMRI) voxels. In this study, we proposed a deep learning classification framework with multivariate data-driven based feature extraction for automatic diagnosis of AD. Specifically, a three-level hierarchical partner matching independent components analysis (3LHPM-ICA) approach was proposed first in order to address the issues in spatial individual ICA, including the uncertainty of the numbers of components, the randomness of initial values, and the correspondence of ICs of multiple subjects, resulting in stable and reliable ICs which were applied as the intrinsic brain functional connectivity (FC) features. Second, Granger causality (GC) was utilized to infer directional interaction between the ICs that were identified by the 3LHPM-ICA method and extract the effective connectivity features. Finally, a deep learning classification framework was developed to distinguish AD from controls by fusing the functional and effective connectivities. A resting state fMRI dataset containing 34 AD patients and 34 normal controls (NCs) was applied to the multivariate deep learning platform, leading to a classification accuracy of 95.59%, with a sensitivity of 97.06% and a specificity of 94.12% with leave-one-out cross validation (LOOCV). The experimental results demonstrated that the measures of neural connectivities of ICA and GC followed by deep learning classification represented the most powerful methods of distinguishing AD clinical data from NCs, and these aberrant brain connectivities might serve as robust brain biomarkers for AD. This approach also allows for expansion of the methodology to classify other psychiatric disorders.

Blind fMRI source unmixing via higher-order tensor decompositions

BACKGROUND

The growing interest in neuroimaging technologies generates a massive amount of biomedical data of high dimensionality. Tensor-based analysis of brain imaging data has been recognized as an effective analysis that exploits its inherent multi-way nature. In particular, the advantages of tensorial over matrix-based methods have previously been demonstrated in the context of functional magnetic resonance imaging (fMRI) source localization. However, such methods can also become ineffective in realistic challenging scenarios, involving, e.g., strong noise and/or significant overlap among the activated regions. Moreover, they commonly rely on the assumption of an underlying multilinear model generating the data.

NEW METHOD

This paper aims at investigating the possible gains from exploiting the 4-dimensional nature of the brain images, through a higher-order tensorization of the fMRI signal, and the use of less restrictive generative models. In this context, the higher-order block term decomposition (BTD) and the PARAFAC2 tensor models are considered for the first time in fMRI blind source separation. A novel PARAFAC2-like extension of BTD (BTD2) is also proposed, aiming at combining the effectiveness of BTD in handling strong instances of noise and the potential of PARAFAC2 to cope with datasets that do not follow the strict multilinear assumption.

COMPARISON WITH EXISTING METHODS

The methods were tested using both synthetic and real data and compared with state of the art methods.

CONCLUSIONS

The simulation results demonstrate the effectiveness of BTD and BTD2 for challenging scenarios (presence of noise, spatial overlap among activation regions and inter-subject variability in the haemodynamic response function (HRF)).

Machine learning of brain gray matter differentiates sex in a large forensic sample

Differences between males and females have been extensively documented in biological, psychological, and behavioral domains. Among these, sex differences in the rate and typology of antisocial behavior remains one of the most conspicuous and enduring patterns among humans. However, the nature and extent of sexual dimorphism in the brain among antisocial populations remains mostly unexplored. Here, we seek to understand sex differences in brain structure between incarcerated males and females in a large sample (n = 1,300) using machine learning. We apply source-based morphometry, a contemporary multivariate approach for quantifying gray matter measured with magnetic resonance imaging, and carry these parcellations forward using machine learning to classify sex. Models using components of brain gray matter volume and concentration were able to differentiate between males and females with greater than 93% generalizable accuracy. Highly differentiated components include orbitofrontal and frontopolar regions, proportionally larger in females, and anterior medial temporal regions proportionally larger in males. We also provide a complimentary analysis of a nonforensic healthy control sample and replicate our 93% sex discrimination. These findings demonstrate that the brains of males and females are highly distinguishable. Understanding sex differences in the brain has implications for elucidating variability in the incidence and progression of disease, psychopathology, and differences in psychological traits and behavior. The reliability of these differences confirms the importance of sex as a moderator of individual differences in brain structure and suggests future research should consider sex specific models.

Capturing Dynamic Connectivity from Resting State fMRI using Time-Varying Graphical Lasso

Functional connectivity (FC) within the human brain evaluated through functional magnetic resonance imaging (fMRI) data has attracted increasing attention and has been employed to study the development of the brain or health conditions of the brain. Many different approaches have been proposed to estimate FC from fMRI data, whereas many of them rely on an implicit assumption that functional connectivity should be static throughout the fMRI scan session. Recently, the fMRI community has realized the limitation of assuming static connectivity and dynamic approaches are more prominent in the resting state fMRI (rs-fMRI) analysis. The sliding window technique has been widely used in many studies to capture network dynamics, but has a number of limitations. In this study, we apply a time-varying graphical lasso (TVGL) model, an extension from the traditional graphical lasso, to address the challenge, which can greatly improve the estimation of FC. The performance of estimating dynamic FC is evaluated with the TVGL through both simulated experiments and real rs-fMRI data from the Philadelphia Neurodevelopmental Cohort (PNC) project. Improved performance is achieved over the sliding window technique. In particular, group differences and transition behaviours between young adults and children are investigated using the estimated dynamic connectivity networks, which help us to better unveil the mechanisms underlying the evolution of the brain over time.

Constrained temporal parallel decomposition for EEG-fMRI fusion

OBJECTIVE

Multimodal neuroimaging has become a common practice in neuroscience research. Simultaneous EEG-fMRI is a popular multimodal recording approach due to the complementary spatiotemporal relationship between the two modalities. Several data fusion techniques have been proposed in the literature for EEG-fMRI fusion, including joint-ICA and parallel-ICA frameworks. Previous EEG-fMRI fusion approaches have used sensor-level EEG features. Recently, we introduced source-space ICA for EEG-MEG source reconstruction and component identification, which was shown to be a superior alternative to sensor-space ICA.

APPROACH

Here, we extend source-space ICA to the fusion of EEG-fMRI data. Additionally, we incorporate the use of a paradigm signal (constrained) and a lag-based signal decomposition approach to accommodate recent findings demonstrating the potentially variable lag structure between electrophysiological and BOLD signals. We evaluated this method on simulated concurrent EEG-fMRI during a boxcar task design, as well as real concurrent EEG-fMRI data from three participants performing an N-Back working memory task. The block diagram of the algorithm and corresponding source codes are provided.

MAIN RESULTS

Based on the results of the real working memory task, for all three subjects, one frontal theta component, and one right posterior alpha component had the highest contribution coefficients (~0.5) to the paradigm-related fused component. There were also two more alpha band components with contribution coefficients of 0.3. The highest contributing fMRI component (~0.8) was one known in the literature to be related to the attention network. The second fMRI component was related to the well-known default mode network, with a contribution coefficient of 0.3.

SIGNIFICANCE

The proposed EEG-fMRI fusion approach, is capable of estimating the brain maps of the EEG and fMRI for the fused components and account for the variable lag structure between electrophysiological and BOLD signals.

Decentralized temporal independent component analysis: Leveraging fMRI data in collaborative settings

The field of neuroimaging has recently witnessed a strong shift towards data sharing; however, current collaborative research projects may be unable to leverage institutional architectures that collect and store data in local, centralized data centers. Additionally, though research groups are willing to grant access for collaborations, they often wish to maintain control of their data locally. These concerns may stem from research culture as well as privacy and accountability concerns. In order to leverage the potential of these aggregated larger data sets, we require tools that perform joint analyses without transmitting the data. Ideally, these tools would have similar performance and ease of use as their current centralized counterparts. In this paper, we propose and evaluate a new Algorithm, decentralized joint independent component analysis (djICA), which meets these technical requirements. djICA shares only intermediate statistics about the data, plausibly retaining privacy of the raw information to local sites, thus making it amenable to further privacy protections, for example via differential privacy. We validate our method on real functional magnetic resonance imaging (fMRI) data and show that it enables collaborative large-scale temporal ICA of fMRI, a rich vein of analysis as of yet largely unexplored, and which can benefit from the larger-N studies enabled by a decentralized approach. We show that djICA is robust to different distributions of data over sites, and that the temporal components estimated with djICA show activations similar to the temporal functional modes analyzed in previous work, thus solidifying djICA as a new, decentralized method oriented toward the frontiers of temporal independent component analysis.

A structural equation model for imaging genetics using spatial transcriptomics

Imaging genetics deals with relationships between genetic variation and imaging variables, often in a disease context. The complex relationships between brain volumes and genetic variants have been explored with both dimension reduction methods and model-based approaches. However, these models usually do not make use of the extensive knowledge of the spatio-anatomical patterns of gene activity. We present a method for integrating genetic markers (single nucleotide polymorphisms) and imaging features, which is based on a causal model and, at the same time, uses the power of dimension reduction. We use structural equation models to find latent variables that explain brain volume changes in a disease context, and which are in turn affected by genetic variants. We make use of publicly available spatial transcriptome data from the Allen Human Brain Atlas to specify the model structure, which reduces noise and improves interpretability. The model is tested in a simulation setting and applied on a case study of the Alzheimer's Disease Neuroimaging Initiative.

Integrating Space, Time, and Orientation in Spiking Neural Networks: A Case Study on Multimodal Brain Data Modeling

Recent progress in a noninvasive brain data sampling technology has facilitated simultaneous sampling of multiple modalities of brain data, such as functional magnetic resonance imaging, electroencephalography, diffusion tensor imaging, and so on. In spite of the potential benefits from integrating predictive modeling of multiple modality brain data, this area of research remains mostly unexplored due to a lack of methodological advancements. The difficulty in fusing multiple modalities of brain data within a single model lies in the heterogeneous temporal and spatial characteristics of the data sources. Recent advances in spiking neural network systems, however, provide the flexibility to incorporate multidimensional information within the model. This paper proposes a novel, unsupervised learning algorithm for fusing temporal, spatial, and orientation information in a spiking neural network architecture that could potentially be used to understand and perform predictive modeling using multimodal data. The proposed algorithm is evaluated both qualitatively and quantitatively using synthetically generated data to characterize its behavior and its ability to utilize spatial, temporal, and orientation information within the model. This leads to improved pattern recognition capabilities and performance along with robust interpretability of the brain data. Furthermore, a case study is presented, which aims to build a computational model that discriminates between people with schizophrenia who respond or do not respond to monotherapy with the antipsychotic clozapine.

Functional and Structural Plasticity of Brain in Elite Karate Athletes

The structural and functional neural differences between the elite karate athletes and control group have been investigated in the concept of this study. 13 elite karate athletes and age-gender matched 13 volunteers who have not performed regular exercises participated in the study. Magnetic resonance imaging was used to acquire the anatomical and functional maps. T1-weighted anatomical images were segmented to form gray and white matter images. Voxel-based morphometry is used to elucidate the differences between the groups. Moreover, resting state functional measurements had been done, and group independent component analysis was implemented in order to exhibit the resting state networks. Then, second-level general linear models were used to compute the statistical maps. It has been revealed that increased GM volume values of inferior/superior temporal, occipital, premotor cortex, and temporal pole superior were present for the elite athletes. Additionally, WM values were found to be increased in caudate nucleus, hypothalamus, and mammilary region for the elite karate players. Similarly, for the elite karate players, the brain regions involved in the movement planning and visual perception are found to have higher connectivity values. The differences in these findings can be thought to be originated from the advances gained through the several years of training which is required to be an elite karate athlete.

An approach to directly link ICA and seed-based functional connectivity: Application to schizophrenia

Independent component analysis (ICA) and seed-based analyses are widely used techniques for studying intrinsic neuronal activity in task-based or resting scans. In this work, we show there is a direct link between the two, and show that there are some important differences between the two approaches in terms of what information they capture. We developed an enhanced connectivity-matrix independent component analysis (cmICA) for calculating whole brain voxel maps of functional connectivity, which reduces the computational complexity of voxel-based connectivity analysis on performing many temporal correlations. We also show there is a mathematical equivalency between parcellations on voxel-to-voxel functional connectivity and simplified cmICA. Next, we used this cost-efficient data-driven method to examine the resting state fMRI connectivity in schizophrenia patients (SZ) and healthy controls (HC) on a whole brain scale and further quantified the relationship between brain functional connectivity and cognitive performances measured by the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) battery. Current results suggest that SZ exhibit a wide-range abnormality, primarily a decrease, in functional connectivity both between networks and within different network hubs. Specific functional connectivity decreases were associated with MATRICS performance deficits. In addition, we found that resting state functional connectivity decreases was extensively associated with aging regardless of groups. In contrast, there was no relationship between positive and negative symptoms in the patients and functional connectivity. In sum, we have developed a novel mathematical relationship between ICA and seed-based connectivity that reduces computational complexity, which has broad applicability, and showed a specific application of this approach to characterize connectivity changes associated with cognitive scores in SZ.

Group independent component analysis reveals alternation of right executive control network in Internet gaming disorder

OBJECTIVE

Previous studies have demonstrated that individuals with Internet gaming disorder (IGD) showed attentional bias toward gaming-related cues and exhibited impaired executive functions. The purpose of this study was to explore the alternations in related functional brain networks underlying attentional bias in IGD subjects.

METHODS

Eighteen IGD subjects and 19 healthy controls (HC) were scanned with functional magnetic resonance imaging while they were performing an addiction Stroop task. Networks of functional connectivity were identified using group independent component analysis (ICA).

RESULTS

ICA identified 4 functional networks that showed differences between the 2 groups, which were related to the right executive control network and visual related networks in our study. Within the right executive control network, in contrast to controls, IGD subjects showed increased functional connectivity in the temporal gyrus and frontal gyrus, and reduced functional connectivity in the posterior cingulate cortex, temporal gyrus, and frontal gyrus.

CONCLUSION

These findings suggest that IGD is related to abnormal functional connectivity of the right executive control network, and may be described as addiction-related abnormally increased cognitive control processing and diminished response inhibition during an addiction Stroop task. The results suggest that IGD subjects show increased susceptibility towards gaming-related cues but weakened strength of inhibitory control.

Fused Estimation of Sparse Connectivity Patterns From Rest fMRI-Application to Comparison of Children and Adult Brains

In this paper, we consider the problem of estimating multiple sparse, co-activated brain regions from functional magnetic resonance imaging (fMRI) observations belonging to different classes. More precisely, we propose a method to analyze similarities and differences in functional connectivity between children and young adults. Often, analysis is conducted on each class separately, and differences across classes are identified with an additional postprocessing step using adequate statistical tools. Here, we propose to rely on a generalized fused Lasso penalty, which allows us to make use of the entire data set in order to estimate connectivity patterns that are either shared across classes, or specific to a given group. By using the entire population during the estimation, we hope to increase the power of our analysis. The proposed model falls in the category of population-wise matrix decomposition, and a simple and efficient alternating direction method of multipliers algorithm is introduced to solve the associated optimization problem. After validating our approach on simulated data, experiments are performed on resting-state fMRI imaging from the Philadelphia neurodevelopmental cohort data set, comprised of normally developing children from ages 8 to 21. Developmental differences were observed in various brain regions, as a total of three class-specific resting-state components were identified. Statistical analysis of the estimated subject-specific features, as well as classification results (based on age groups, up to 81% accuracy, samples) related to these components demonstrate that the proposed method is able to properly extract meaningful shared and class-specific sub-networks.

Functional brain networks for learning predictive statistics

Making predictions about future events relies on interpreting streams of information that may initially appear incomprehensible. This skill relies on extracting regular patterns in space and time by mere exposure to the environment (i.e., without explicit feedback). Yet, we know little about the functional brain networks that mediate this type of statistical learning. Here, we test whether changes in the processing and connectivity of functional brain networks due to training relate to our ability to learn temporal regularities. By combining behavioral training and functional brain connectivity analysis, we demonstrate that individuals adapt to the environment's statistics as they change over time from simple repetition to probabilistic combinations. Further, we show that individual learning of temporal structures relates to decision strategy. Our fMRI results demonstrate that learning-dependent changes in fMRI activation within and functional connectivity between brain networks relate to individual variability in strategy. In particular, extracting the exact sequence statistics (i.e., matching) relates to changes in brain networks known to be involved in memory and stimulus-response associations, while selecting the most probable outcomes in a given context (i.e., maximizing) relates to changes in frontal and striatal networks. Thus, our findings provide evidence that dissociable brain networks mediate individual ability in learning behaviorally-relevant statistics.

The role of diversity in data-driven analysis of multi-subject fMRI data: Comparison of approaches based on independence and sparsity using global performance metrics

Data-driven methods have been widely used in functional magnetic resonance imaging (fMRI) data analysis. They extract latent factors, generally, through the use of a simple generative model. Independent component analysis (ICA) and dictionary learning (DL) are two popular data-driven methods that are based on two different forms of diversity-statistical properties of the data-statistical independence for ICA and sparsity for DL. Despite their popularity, the comparative advantage of emphasizing one property over another in the decomposition of fMRI data is not well understood. Such a comparison is made harder due to the differences in the modeling assumptions between ICA and DL, as well as within different ICA algorithms where each algorithm exploits a different form of diversity. In this paper, we propose the use of objective global measures, such as time course frequency power ratio, network connection summary, and graph theoretical metrics, to gain insight into the role that different types of diversity have on the analysis of fMRI data. Four ICA algorithms that account for different types of diversity and one DL algorithm are studied. We apply these algorithms to real fMRI data collected from patients with schizophrenia and healthy controls. Our results suggest that no one particular method has the best performance using all metrics, implying that the optimal method will change depending on the goal of the analysis. However, we note that in none of the scenarios we test the highly popular Infomax provides the best performance, demonstrating the cost of exploiting limited form of diversity.

Spatio-Temporal Dynamics of Intrinsic Networks in Functional Magnetic Imaging Data Using Recurrent Neural Networks

We introduce a novel recurrent neural network (RNN) approach to account for temporal dynamics and dependencies in brain networks observed via functional magnetic resonance imaging (fMRI). Our approach directly parameterizes temporal dynamics through recurrent connections, which can be used to formulate blind source separation with a conditional (rather than marginal) independence assumption, which we call RNN-ICA. This formulation enables us to visualize the temporal dynamics of both first order (activity) and second order (directed connectivity) information in brain networks that are widely studied in a static sense, but not well-characterized dynamically. RNN-ICA predicts dynamics directly from the recurrent states of the RNN in both task and resting state fMRI. Our results show both task-related and group-differentiating directed connectivity.

Mapping hybrid functional-structural connectivity traits in the human connectome

One of the crucial questions in neuroscience is how a rich functional repertoire of brain states relates to its underlying structural organization. How to study the associations between these structural and functional layers is an open problem that involves novel conceptual ways of tackling this question. We here propose an extension of the Connectivity Independent Component Analysis (connICA) framework, to identify joint structural-functional connectivity traits. Here, we extend connICA to integrate structural and functional connectomes by merging them into common "hybrid" connectivity patterns that represent the connectivity fingerprint of a subject. We test this extended approach on the 100 unrelated subjects from the Human Connectome Project. The method is able to extract main independent structural-functional connectivity patterns from the entire cohort that are sensitive to the realization of different tasks. The hybrid connICA extracted two main task-sensitive hybrid traits. The first, encompassing the within and between connections of dorsal attentional and visual areas, as well as fronto-parietal circuits. The second, mainly encompassing the connectivity between visual, attentional, DMN and subcortical networks. Overall, these findings confirms the potential of the hybrid connICA for the compression of structural/functional connectomes into integrated patterns from a set of individual brain networks.

Classification and Prediction of Brain Disorders Using Functional Connectivity: Promising but Challenging

Brain functional imaging data, especially functional magnetic resonance imaging (fMRI) data, have been employed to reflect functional integration of the brain. Alteration in brain functional connectivity (FC) is expected to provide potential biomarkers for classifying or predicting brain disorders. In this paper, we present a comprehensive review in order to provide guidance about the available brain FC measures and typical classification strategies. We survey the state-of-the-art FC analysis methods including widely used static functional connectivity (SFC) and more recently proposed dynamic functional connectivity (DFC). Temporal correlations among regions of interest (ROIs), data-driven spatial network and functional network connectivity (FNC) are often computed to reflect SFC from different angles. SFC can be extended to DFC using a sliding-window framework, and intrinsic connectivity states along the time-varying connectivity patterns are typically extracted using clustering or decomposition approaches. We also briefly summarize window-less DFC approaches. Subsequently, we highlight various strategies for feature selection including the filter, wrapper and embedded methods. In terms of model building, we include traditional classifiers as well as more recently applied deep learning methods. Moreover, we review representative applications with remarkable classification accuracy for psychosis and mood disorders, neurodevelopmental disorder, and neurological disorders using fMRI data. Schizophrenia, bipolar disorder, autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), Alzheimer's disease and mild cognitive impairment (MCI) are discussed. Finally, challenges in the field are pointed out with respect to the inaccurate diagnosis labeling, the abundant number of possible features and the difficulty in validation. Some suggestions for future work are also provided.