GMAC: A Matlab toolbox for spectral Granger causality analysis of fMRI data

Schizophrenia MEG Network Analysis Based on Kernel Granger Causality

Network analysis is an important approach to explore complex brain structures under different pathological and physiological conditions. In this paper, we employ the multivariate inhomogeneous polynomial kernel Granger causality (MKGC) to construct directed weighted networks to characterize schizophrenia magnetoencephalography (MEG). We first generate data based on coupled autoregressive processes to test the effectiveness of MKGC in comparison with the bivariate linear Granger causality and bivariate inhomogeneous polynomial kernel Granger causality. The test results suggest that MKGC outperforms the other two methods. Based on these results, we apply MKGC to construct effective connectivity networks of MEG for patients with schizophrenia (SCZs). We measure three network features, i.e., strength, nonequilibrium, and complexity, to characterize schizophrenia MEG. Our results suggest that MEG of the healthy controls (HCs) has a denser effective connectivity network than that of SCZs. The most significant difference in the in-connectivity strength is observed in the right frontal network (p=0.001). The strongest out-connectivity strength for all subjects occurs in the temporal area, with the most significant between-group difference in the left occipital area (p=0.0018). The total connectivity strength of the frontal, temporal, and occipital areas of HCs exhibits higher values compared with SCZs. The nonequilibrium feature over the whole brain of SCZs is significantly higher than that of the HCs (p=0.012); however, the results of Shannon entropy suggest that healthy MEG networks have higher complexity than schizophrenia networks. Overall, MKGC provides a reliable approach to construct MEG brain networks and characterize the network characteristics.

Granger causality test with nonlinear neural-network-based methods: Python package and simulation study

BACKGROUND AND OBJECTIVE

Causality defined by Granger in 1969 is a widely used concept, particularly in neuroscience and economics. As there is an increasing interest in nonlinear causality research, a Python package with a neural-network-based causality analysis approach was created. It allows performing causality tests using neural networks based on Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), or Multilayer Perceptron (MLP). The aim of this paper is to present the nonlinear method for causality analysis and the created Python package.

METHODS

The created functions with the autoregressive (AR) and Generalized Radial Basis Functions (GRBF) neural network models were tested on simulated signals in two cases: with nonlinear dependency and with absence of causality from Y to X signal. The train-test split (70/30) was used. Errors obtained on the test set were compared using the Wilcoxon signed-rank test to determine the presence of the causality. For the chosen model, the proposed method of study the change of causality over time was presented.

RESULTS

In the case when X was a polynomial of Y, nonlinear methods were able to detect the causality, while the AR model did not manage to indicate it. The best results (in terms of the prediction accuracy) were obtained for the MLP for the lag of 150 (MSE equal to 0.011, compared to 0.041 and 0.036 for AR and GRBF, respectively). When there was no causality between the signals, none of the proposed and AR models did indicate false causality, while it was detected by GRBF models in one case. Only the proposed models gave the expected results in each of the tested scenarios.

CONCLUSIONS

The proposed method appeared to be superior to the compared methods. They were able to detect non-linear causality, make accurate forecasting and not indicate false causality. The created package enables easy usage of neural networks to study the causal relationship between signals. The neural-networks-based approach is a suitable method that allows the detection of a nonlinear causal relationship, which cannot be detected by the classical Granger method. Unlike other similar tools, the package allows for the study of changes in causality over time.

Alterations in the Magnetoencephalography Default Mode Effective Connectivity following Concussion

BACKGROUND AND PURPOSE

Magnetoencephalography is sensitive to functional connectivity changes associated with concussion. However, the directional influences between functionally related regions remain unexplored. In this study, we therefore evaluated concussion-related magnetoencephalography-based effective connectivity changes within resting-state default mode network regions.

MATERIALS AND METHODS

Resting-state magnetoencephalography was acquired for 8 high school football players with concussion at 3 time points (preseason, postconcussion, postseason), as well as 8 high school football players without concussion and 8 age-matched controls at 2 time points (preseason, postseason). Time-series from the default mode network regions were extracted, and effective connectivity between them was computed for 5 different frequency bands. The default mode network regions were grouped into anterior and posterior default mode networks. The combined posterior-to-anterior and anterior-to-posterior effective connectivity values were averaged to generate 2 sets of values for each subject. The effective connectivity values were compared using a repeated measures ANOVA across time points for the concussed, nonconcussed, and control groups, separately.

RESULTS

A significant increase in posterior-to-anterior effective connectivity from preseason to postconcussion (corrected P value = .013) and a significant decrease in posterior-to-anterior effective connectivity from postconcussion to postseason (corrected P value = .028) were observed in the concussed group. Changes in effective connectivity were only significant within the delta band. Anterior-to-posterior connectivity demonstrated no significant change. Effective connectivity in the nonconcussed group and controls did not show significant differences.

CONCLUSIONS

The unidirectional increase in effective connectivity postconcussion may elucidate compensatory processes, invoking use of posterior regions to aid the function of susceptible anterior regions following brain injury. These findings support the potential value of magnetoencephalography in exploring directional changes of the brain network following concussion.

Effective Connectivity During Rest and Music Listening: An EEG Study on Parkinson's Disease

Music-based interventions seem to enhance motor, sensory and cognitive functions in Parkinson’s disease (PD), but the underlying action mechanisms are still largely unknown. This electroencephalography (EEG) study aimed to investigate the effective connectivity patterns characterizing PD in the resting state and during music listening. EEG recordings were obtained from fourteen non-demented PD patients and 12 healthy controls, at rest and while listening to three music tracks. Theta- and alpha-band power spectral density and multivariate partial directed coherence were computed. Power and connectivity measures were compared between patients and controls in the four conditions and in music vs. rest. Compared to controls, patients showed enhanced theta-band power and slightly enhanced alpha-band power, but markedly reduced theta- and alpha-band interactions among EEG channels, especially concerning the information received by the right central channel. EEG power differences were partially reduced by music listening, which induced power increases in controls but not in patients. Connectivity differences were slightly compensated by music, whose effects largely depended on the track. In PD, music enhanced the frontotemporal inter-hemispheric communication. Our findings suggest that PD is characterized by enhanced activity but reduced information flow within the EEG network, being only partially normalized by music. Nevertheless, music capability to facilitate inter-hemispheric communication might underlie its beneficial effects on PD pathophysiology and should be further investigated.

The role of inferior frontal junction in controlling the spatially global effect of feature-based attention in human visual areas

Feature-based attention has a spatially global effect, i.e., responses to stimuli that share features with an attended stimulus are enhanced not only at the attended location but throughout the visual field. However, how feature-based attention modulates cortical neural responses at unattended locations remains unclear. Here we used functional magnetic resonance imaging (fMRI) to examine this issue as human participants performed motion- (Experiment 1) and color- (Experiment 2) based attention tasks. Results indicated that, in both experiments, the respective visual processing areas (middle temporal area [MT+] for motion and V4 for color) as well as early visual, parietal, and prefrontal areas all showed the classic feature-based attention effect, with neural responses to the unattended stimulus significantly elevated when it shared the same feature with the attended stimulus. Effective connectivity analysis using dynamic causal modeling (DCM) showed that this spatially global effect in the respective visual processing areas (MT+ for motion and V4 for color), intraparietal sulcus (IPS), frontal eye field (FEF), medial frontal gyrus (mFG), and primary visual cortex (V1) was derived by feedback from the inferior frontal junction (IFJ). Complementary effective connectivity analysis using Granger causality modeling (GCM) confirmed that, in both experiments, the node with the highest outflow and netflow degree was IFJ, which was thus considered to be the source of the network. These results indicate a source for the spatially global effect of feature-based attention in the human prefrontal cortex.

Attentional selection is the mechanism by which relevant sensory information is processed preferentially. Feature-based attention plays a key role in identifying an attentional target in a complex scene, because we often know the features of the target but not its exact location. The ability to quickly select the target is mainly attributed to enhancement of responses to stimuli that share features with an attended stimulus, not only at the attended location but throughout the whole visual field. However, little is known regarding how feature-based attention modulates brain responses at unattended locations. Here we used fMRI and advanced connectivity analyses to examine human subjects as they performed either motion- or color-based attention tasks. Our results indicated that the visual processing areas for motion and color showed the feature-based attention effect. Effective connectivity analysis showed that this feature-based attention effect was derived by feedback from the inferior frontal junction, an area of the posterior lateral prefrontal cortex involved in many different cognitive processes, including spatial attention and working memory. Further modeling confirmed that the inferior frontal junction showed connectivity features supporting its role as the source of the network. Our results support the hypothesis that the inferior frontal junction plays a key role in the spatially global effect of feature-based attention.

Resting State Effective Connectivity Allows Auditory Hallucination Discrimination

Hallucinations are elusive phenomena that have been associated with psychotic behavior, but that have a high prevalence in healthy population. Some generative mechanisms of Auditory Hallucinations (AH) have been proposed in the literature, but so far empirical evidence is scarce. The most widely accepted generative mechanism hypothesis nowadays consists in the faulty workings of a network of brain areas including the emotional control, the audio and language processing, and the inhibition and self-attribution of the signals in the auditive cortex. In this paper, we consider two methods to analyze resting state fMRI (rs-fMRI) data, in order to measure effective connections between the brain regions involved in the AH generation process. These measures are the Dynamic Causal Modeling (DCM) cross-covariance function (CCF) coefficients, and the partially directed coherence (PDC) coefficients derived from Granger Causality (GC) analysis. Effective connectivity measures are treated as input classifier features to assess their significance by means of cross-validation classification accuracy results in a wrapper feature selection approach. Experimental results using Support Vector Machine (SVM) classifiers on an rs-fMRI dataset of schizophrenia patients with and without a history of AH confirm that the main regions identified in the AH generative mechanism hypothesis have significant effective connection values, under both DCM and PDC evaluation.

A Hitchhiker's Guide to Functional Magnetic Resonance Imaging

Functional Magnetic Resonance Imaging (fMRI) studies have become increasingly popular both with clinicians and researchers as they are capable of providing unique insights into brain functions. However, multiple technical considerations (ranging from specifics of paradigm design to imaging artifacts, complex protocol definition, and multitude of processing and methods of analysis, as well as intrinsic methodological limitations) must be considered and addressed in order to optimize fMRI analysis and to arrive at the most accurate and grounded interpretation of the data. In practice, the researcher/clinician must choose, from many available options, the most suitable software tool for each stage of the fMRI analysis pipeline. Herein we provide a straightforward guide designed to address, for each of the major stages, the techniques, and tools involved in the process. We have developed this guide both to help those new to the technique to overcome the most critical difficulties in its use, as well as to serve as a resource for the neuroimaging community.

Investigation of the electrophysiological correlates of negative BOLD response during intermittent photic stimulation: An EEG-fMRI study

Although the occurrence of concomitant positive BOLD responses (PBRs) and negative BOLD responses (NBRs) to visual stimuli is increasingly investigated in neuroscience, it still lacks a definite explanation. Multimodal imaging represents a powerful tool to study the determinants of negative BOLD responses: the integration of functional Magnetic Resonance Imaging (fMRI) and electroencephalographic (EEG) recordings is especially useful, since it can give information on the neurovascular coupling underlying this complex phenomenon. In the present study, the brain response to intermittent photic stimulation (IPS) was investigated in a group of healthy subjects using simultaneous EEG-fMRI, with the main objective to study the electrophysiological mechanisms associated with the intense NBRs elicited by IPS in extra-striate visual cortex. The EEG analysis showed that IPS induced a desynchronization of the basal rhythm, followed by the instauration of a novel rhythm driven by the visual stimulation. The most interesting results emerged from the EEG-informed fMRI analysis, which suggested a relationship between the neuronal rhythms at 10 and 12 Hz and the BOLD dynamics in extra-striate visual cortex. These findings support the hypothesis that NBRs to visual stimuli may be neuronal in origin rather than reflecting pure vascular phenomena. Hum Brain Mapp 37:2247-2262, 2016. © 2016 Wiley Periodicals, Inc.

Monitoring Neuro-Motor Recovery From Stroke With High-Resolution EEG, Robotics and Virtual Reality: A Proof of Concept

A novel system for the neuro-motor rehabilitation of upper limbs was validated in three sub-acute post-stroke patients. The system permits synchronized cortical and kinematic measures by integrating high-resolution EEG, passive robotic device and Virtual Reality. The brain functional re-organization was monitored in association with motor patterns replicating activities of daily living (ADL). Patients underwent 13 rehabilitation sessions. At sessions 1, 7 and 13, clinical tests were administered to assess the level of motor impairment, and EEG was recorded during rehabilitation task execution. For each session and rehabilitation task, four kinematic indices of motor performance were calculated and compared with the outcome of clinical tests. Functional source maps were obtained from EEG data and projected on the real patients' anatomy (MRI data). Laterality indices were calculated for hemispheric dominance assessment. All patients showed increased participation in the rehabilitation process. Cortical activation changes during recovery were detected in relation to different motor patterns, hence verifying the system's suitability to add quantitative measures of motor performance and neural recovery to classical tests. We conclude that this system seems a promising tool for novel robot-based rehabilitation paradigms tailored to individual needs and neuro-motor responses of the patients.

Causal interaction following the alteration of target region activation during motor imagery training using real-time fMRI

Motor imagery training is an effective approach for motor skill learning and motor function rehabilitation. As a novel method of motor imagery training, real-time fMRI (rtfMRI) enables individuals to acquire self-control of localized brain activation, achieving desired changes in behavior. The regulation of target region activation by rtfMRI often alters the activation of related brain regions. However, the interaction between the target region and these related regions is unclear. The Granger causality model (GCM) is a data-driven method that can explore the causal interaction between brain regions. In this study, we employed rtfMRI to train subjects to regulate the activation of the ipsilateral dorsal premotor area (dPMA) during motor imagery training, and we calculated the causal interaction of the dPMA with other motor-related regions based on the GCM. The results demonstrated that as the activity of the dPMA changed during rtfMRI training, the interaction of the target region with other related regions became significantly altered, and behavioral performance was improved after training. The altered interaction primarily exhibited as an increased unidirectional interaction from the dPMA to the other regions. These findings support the dominant role of the dPMA in motor skill learning via rtfMRI training and may indicate how activation of the target region interacts with the activation of other related regions.

Coupling of fMRI and NIRS measurements in the study of negative BOLD response to intermittent photic stimulation

Functional Magnetic Resonance Imaging (fMRI) in combination with Near Infrared Spectroscopy (NIRS) is finding widespread use in the analysis of brain function. While most of the studies deal with the detection of positive responses, here we focus on negative responses to visual stimulation. In a group fMRI study on Intermittent Photic Stimulation (IPS) we detected a sustained Negative BOLD Response (NBR) in the extrastriate visual cortex. To confirm and better characterize NBR, we repeated the same protocol during NIRS recordings. In this paper we show fMRI results and demonstrate the NBR on the basis of NIRS findings.