Spectral factorization-based current source density analysis of ongoing neural oscillations

Explainable AI-based Deep-SHAP for mapping the multivariate relationships between regional neuroimaging biomarkers and cognition

BACKGROUND

Mild cognitive impairment (MCI)/Alzheimer's disease (AD) is associated with cognitive decline beyond normal aging and linked to the alterations of brain volume quantified by magnetic resonance imaging (MRI) and amyloid-beta (Aβ) quantified by positron emission tomography (PET). Yet, the complex relationships between these regional imaging measures and cognition in MCI/AD remain unclear. Explainable artificial intelligence (AI) may uncover such relationships.

METHOD

We integrate the AI-based deep learning neural network and Shapley additive explanations (SHAP) approaches and introduce the Deep-SHAP method to investigate the multivariate relationships between regional imaging measures and cognition. After validating this approach on simulated data, we apply it to real experimental data from MCI/AD patients.

RESULTS

Deep-SHAP significantly predicted cognition using simulated regional features and identified the ground-truth simulated regions as the most significant multivariate predictors. When applied to experimental MRI data, Deep-SHAP revealed that the insula, lateral occipital, medial frontal, temporal pole, and occipital fusiform gyrus are the primary contributors to global cognitive decline in MCI/AD. Furthermore, when applied to experimental amyloid Pittsburgh compound B (PiB)-PET data, Deep-SHAP identified the key brain regions for global cognitive decline in MCI/AD as the inferior temporal, parahippocampal, inferior frontal, supratemporal, and lateral frontal gray matter.

CONCLUSION

Deep-SHAP method uncovered the multivariate relationships between regional brain features and cognition, offering insights into the most critical modality-specific brain regions involved in MCI/AD mechanisms.

Dysregulated Salience Network Control over Default-Mode and Central-Executive Networks in Schizophrenia Revealed Using Stochastic Dynamical Causal Modeling

Introduction: Neuroimaging studies suggest that the human brain consists of intrinsically organized, large-scale neural networks. Among these networks, the interplay among the default-mode network (DMN), salience network (SN), and central-executive network (CEN) has been widely used to understand the functional interaction patterns in health and disease. This triple network model suggests that the SN causally controls over the DMN and CEN in healthy individuals. This interaction is often referred to as SN's dynamic regulating mechanism. However, such interactions are not well understood in individuals with schizophrenia. Methods: In this study, we leveraged resting-state functional magnetic resonance imaging data from schizophrenia (n = 67) and healthy controls (n = 81) and evaluated the directional functional interactions among DMN, SN, and CEN using stochastic dynamical causal modeling methodology. Results: In healthy controls, our analyses replicated previous findings that SN regulates DMN and CEN activities (Mann-Whitney U test; p < 10-8). In schizophrenia, however, our analyses revealed a disrupted SN-based controlling mechanism over the DMN and CEN (Mann-Whitney U test; p < 10-16). Conclusions: These results indicate that the disrupted controlling mechanism of SN over the other two neural networks may be a candidate neuroimaging phenotype in schizophrenia.

Spike-phase coupling patterns reveal laminar identity in primate cortex

The cortical column is one of the fundamental computational circuits in the brain. In order to understand the role neurons in different layers of this circuit play in cortical function it is necessary to identify the boundaries that separate the laminar compartments. While histological approaches can reveal ground truth they are not a practical means of identifying cortical layers in vivo. The gold standard for identifying laminar compartments in electrophysiological recordings is current-source density (CSD) analysis. However, laminar CSD analysis requires averaging across reliably evoked responses that target the input layer in cortex, which may be difficult to generate in less well-studied cortical regions. Further, the analysis can be susceptible to noise on individual channels resulting in errors in assigning laminar boundaries. Here, we have analyzed linear array recordings in multiple cortical areas in both the common marmoset and the rhesus macaque. We describe a pattern of laminar spike-field phase relationships that reliably identifies the transition between input and deep layers in cortical recordings from multiple cortical areas in two different non-human primate species. This measure corresponds well to estimates of the location of the input layer using CSDs, but does not require averaging or specific evoked activity. Laminar identity can be estimated rapidly with as little as a minute of ongoing data and is invariant to many experimental parameters. This method may serve to validate CSD measurements that might otherwise be unreliable or to estimate laminar boundaries when other methods are not practical.

Interactions between the anterior cingulate-insula network and the fronto-parietal network during perceptual decision-making

Information processing in the human brain during cognitively demanding goal-directed tasks is thought to involve several large-scale brain networks, including the anterior cingulate-insula network (aCIN) and the fronto-parietal network (FPN). Recent functional MRI (fMRI) studies have provided clues that the aCIN initiates activity changes in the FPN. However, when and how often these networks interact remains largely unknown to date. Here, we systematically examined the oscillatory interactions between the aCIN and the FPN by using the spectral Granger causality analysis of reconstructed brain source signals from the scalp electroencephalography (EEG) recorded from human participants performing a face-house perceptual categorization task. We investigated how the aCIN and the FPN interact, what the temporal sequence of events in these nodes is, and what frequency bands of information flow bind these nodes in networks. We found that beta band (13-30Hz) and gamma (30-100Hz) bands of interactions are involved between the aCIN and the FPN during decision-making tasks. In gamma band, the aCIN initiated the Granger causal control over the FPN in 25-225 ms timeframe. In beta band, the FPN achieved a control over the aCIN in 225-425 ms timeframe. These band-specific time-dependent Granger causal controls of the aCIN and the FPN were retained for behaviorally harder decision-making tasks. These findings of times and frequencies of oscillatory interactions in the aCIN and FPN provide us new insights into the general neural mechanisms for sensory information-guided, goal-directed behaviors, including perceptual decision-making processes.

Face or House Image Perception: Beta and Gamma Bands of Oscillations in Brain Networks Carry Out Decision-Making

Previous functional magnetic resonance imaging studies have consistently shown that perception of visual objects, such as faces and houses, involves distributed brain networks that include the fusiform face area (FFA), parahippocampal place area (PPA), and dorsolateral prefrontal cortex (DLPFC). These regions are commonly observed to be coactivated in BOLD measurements during perception of visual objects. In this study, we aimed to disentangle node-level and network-level activities in millisecond timescale of perception and decision-making in attempts to answer questions about timing and frequency of brain oscillatory activities. We used clear and noisy face-house image categorization tasks and human scalp electroencephalography recordings combined with source reconstruction techniques to study when and how oscillatory activity organizes within the FFA, PPA, and DLPFC. We uncovered the dynamics of two oscillatory networks-beta (13-30 Hz) and gamma (30-100 Hz). In beta band, the node and network activities were enhanced in time frame of 125-250 msec after stimulus onset, the FFA and PPA acted as main outflow hubs and the DLPFC as a main inflow hub, and network activities negatively correlated with behavior measures of noise levels (response times). In gamma band, node and network activities were elevated in time frame of 0-125 msec after stimulus onset, the DLPFC acted as a main outflow hub, and finally network activities were positively correlated with the noise level. These findings broaden our understanding of temporal evolution of node and network features associated with visual perceptual decision-making.

The salience network dynamics in perceptual decision-making

Recent neuroimaging studies have demonstrated that the network consisting of the right anterior insula (rAI), left anterior insula (lAI) and dorsal anterior cingulate cortex (dACC) is activated in sensory stimulus-guided goal-directed behaviors. This network is often known as the salience network (SN). When and how a sensory signal enters and organizes within SN before reaching the central executive network including the prefrontal cortices is still a mystery. Previous electrophysiological studies focused on individual nodes of SN, either on dACC or rAI, have reports of conflicting findings of the earliest cortical activity within the network. Functional magnetic resonance imaging (fMRI) studies are not able to answer these questions in the time-scales of human sensory perception and decision-making. Here, using clear and noisy face-house image categorization tasks and human scalp electroencephalography (EEG) recordings combined with source reconstruction techniques, we study when and how oscillatory activity organizes SN during a perceptual decision. We uncovered that the beta-band (13-30Hz) oscillations bound SN, became most active around 100ms after the stimulus onset and the rAI acted as a main outflow hub within SN for easier decision making task. The SN activities (Granger causality measures) were negatively correlated with the decision response time (decision difficulty). These findings suggest that the SN activity precedes the executive control in mediating sensory and cognitive processing to arrive at visual perceptual decisions.

Interactions Among the Brain Default-Mode, Salience, and Central-Executive Networks During Perceptual Decision-Making of Moving Dots

Cognitively demanding goal-directed tasks in the human brain are thought to involve the dynamic interplay of several large-scale neural networks, including the default-mode network (DMN), salience network (SN), and central-executive network (CEN). Resting-state functional magnetic resonance imaging (rsfMRI) studies have consistently shown that the CEN and SN negatively regulate activity in the DMN, and this switching is argued to be controlled by the right anterior insula (rAI) of the SN. However, what remains to be investigated is the pattern of directed network interactions during difficult perceptual decision-making tasks. We recorded fMRI data while participants categorized the left-right motion of moving dots. We defined regions of interest, extracted fMRI time series, and performed directed connectivity analysis using Granger causality techniques. Our analyses demonstrated that the slow oscillation (0.07-0.19 Hz) mediated the interactions within and between the DMN, SN, and CEN nodes both for easier and harder decision-making tasks. We found that the rAI, a key node of the SN, played a causal control over the DMN and CEN for easier decision-making tasks. The combined effort of the rAI and dorsal anterior cingulate cortex of the SN had the causal control over the DMN and CEN for a harder task. These findings provide important insights into how a sensory signal organizes among the DMN, SN, and CEN during sensory information-guided, goal-directed tasks.