Visualization of the Spatiotemporal Propagation of Interictal Spikes in Temporal Lobe Epilepsy: A MEG Pilot Study.

Magnetoencephalography (MEG) is clinically used to localize interictal spikes in discrete brain areas of epilepsy patients through the equivalent current dipole (ECD) method, but does not account for the temporal dynamics of spike activity. Recent studies found that interictal spike propagation beyond the temporal lobe may be associated with worse postsurgical outcomes, but studies using whole-brain data such as in MEG remain limited. In this pilot study, we developed a tool that visualizes the spatiotemporal dynamics of interictal MEG spikes normalized to spike-free sleep activity to assess their onset and propagation patterns in patients with temporal lobe epilepsy (TLE). We extracted interictal source data containing focal epileptiform activity in awake and asleep states from seven patients whose MEG ECD clusters localized to the temporal lobe and normalized the data against spike-free sleep recordings. We calculated the normalized activity over time per cortical label, confirmed maximal activity at onset, and mapped the activity over a 10 ms interval onto each patient's brain using a custom-built Multi-Modal Visualization Tool. The onset of activity in all patients appeared near the clinically determined epileptogenic zone. By 10 ms, four of the patients had propagated source activity restricted to within the temporal lobe, and three had propagated source activity spread to extratemporal regions. Using this tool, we show that noninvasively identifying the onset and propagation of interictal spike activity in MEG can be achieved, which may help provide further insight into epileptic networks and guide surgical planning and interventions in patients with TLE.

Modular pipeline for reconstruction and localization of implanted intracranial ECoG and sEEG electrodes.

Implantation of electrodes in the brain has been used as a clinical tool for decades to stimulate and record brain activity. As this method increasingly becomes the standard of care for several disorders and diseases, there is a growing need to quickly and accurately localize the electrodes once they are placed within the brain. We share here a protocol pipeline for localizing electrodes implanted in the brain, which we have applied to more than 260 patients, that is accessible to multiple skill levels and modular in execution. This pipeline uses multiple software packages to prioritize flexibility by permitting multiple different parallel outputs while minimizing the number of steps for each output. These outputs include co-registered imaging, electrode coordinates, 2D and 3D visualizations of the implants, automatic surface and volumetric localizations of the brain regions per electrode, and anonymization and data sharing tools. We demonstrate here some of the pipeline's visualizations and automatic localization algorithms which we have applied to determine appropriate stimulation targets, to conduct seizure dynamics analysis, and to localize neural activity from cognitive tasks in previous studies. Further, the output facilitates the extraction of information such as the probability of grey matter intersection or the nearest anatomic structure per electrode contact across all data sets that go through the pipeline. We expect that this pipeline will be a useful framework for researchers and clinicians alike to localize implanted electrodes in the human brain.

Characterizing brain dynamics during ketamine-induced dissociation and subsequent interactions with propofol using human intracranial neurophysiology.

Ketamine produces antidepressant effects in patients with treatment-resistant depression, but its usefulness is limited by its psychotropic side effects. Ketamine is thought to act via NMDA receptors and HCN1 channels to produce brain oscillations that are related to these effects. Using human intracranial recordings, we found that ketamine produces gamma oscillations in prefrontal cortex and hippocampus, structures previously implicated in ketamine's antidepressant effects, and a 3 Hz oscillation in posteromedial cortex, previously proposed as a mechanism for its dissociative effects. We analyzed oscillatory changes after subsequent propofol administration, whose GABAergic activity antagonizes ketamine's NMDA-mediated disinhibition, alongside a shared HCN1 inhibitory effect, to identify dynamics attributable to NMDA-mediated disinhibition versus HCN1 inhibition. Our results suggest that ketamine engages different neural circuits in distinct frequency-dependent patterns of activity to produce its antidepressant and dissociative sensory effects. These insights may help guide the development of brain dynamic biomarkers and novel therapeutics for depression.

Closed-loop enhancement and neural decoding of cognitive control in humans.

Deficits in cognitive control-that is, in the ability to withhold a default pre-potent response in favour of a more adaptive choice-are common in depression, anxiety, addiction and other mental disorders. Here we report proof-of-concept evidence that, in participants undergoing intracranial epilepsy monitoring, closed-loop direct stimulation of the internal capsule or striatum, especially the dorsal sites, enhances the participants' cognitive control during a conflict task. We also show that closed-loop stimulation upon the detection of lapses in cognitive control produced larger behavioural changes than open-loop stimulation, and that task performance for single trials can be directly decoded from the activity of a small number of electrodes via neural features that are compatible with existing closed-loop brain implants. Closed-loop enhancement of cognitive control might remediate underlying cognitive deficits and aid the treatment of severe mental disorders.

Brain structure and episodic learning rate in cognitively healthy ageing.

Memory normally declines with ageing and these age-related cognitive changes are associated with changes in brain structure. Episodic memory retrieval has been widely studied during ageing, whereas learning has received less attention. Here we examined the neural correlates of episodic learning rate in ageing. Our study sample consisted of 982 cognitively healthy female and male older participants from the Vallecas Project cohort, without a clinical diagnosis of mild cognitive impairment or dementia. The learning rate across the three consecutive recall trials of the verbal memory task (Free and Cued Selective Reminding Test) recall trials was used as a predictor of grey matter (GM) using voxel-based morphometry, and WM microstructure using tract-based spatial statistics on fractional anisotropy (FA) and mean diffusivity (MD) measures. Immediate Recall improved by 1.4 items per trial on average, and this episodic learning rate was faster in women and negatively associated with age. Structurally, hippocampal and anterior thalamic GM volume correlated positively with learning rate. Learning also correlated with the integrity of WM microstructure (high FA and low MD) in an extensive network of tracts including bilateral anterior thalamic radiation, fornix, and long-range tracts. These results suggest that episodic learning rate is associated with key anatomical structures for memory functioning, motivating further exploration of the differential diagnostic properties between episodic learning rate and retrieval in ageing.

Long-term ecological assessment of intracranial electrophysiology synchronized to behavioral markers in obsessive-compulsive disorder.

Detection of neural signatures related to pathological behavioral states could enable adaptive deep brain stimulation (DBS), a potential strategy for improving efficacy of DBS for neurological and psychiatric disorders. This approach requires identifying neural biomarkers of relevant behavioral states, a task best performed in ecologically valid environments. Here, in human participants with obsessive-compulsive disorder (OCD) implanted with recording-capable DBS devices, we synchronized chronic ventral striatum local field potentials with relevant, disease-specific behaviors. We captured over 1,000 h of local field potentials in the clinic and at home during unstructured activity, as well as during DBS and exposure therapy. The wide range of symptom severity over which the data were captured allowed us to identify candidate neural biomarkers of OCD symptom intensity. This work demonstrates the feasibility and utility of capturing chronic intracranial electrophysiology during daily symptom fluctuations to enable neural biomarker identification, a prerequisite for future development of adaptive DBS for OCD and other psychiatric disorders.

Region-Level Functional and Effective Network Analysis of Human Brain During Cognitive Task Engagement.

Mental disorders are a major source of disability, with few effective treatments. It has recently been argued that these diseases might be effectively treated by focusing on decision-making, and specifically remediating decision-making deficits that act as "ingredients" in these disorders. Prior work showed that direct electrical brain stimulation can enhance human cognitive control, and consequently decision-making. This raises a challenge of detecting cognitive control lapses directly from electrical brain activity. Here, we demonstrate approaches to overcome that challenge. We propose a novel method, referred to as maximal variance node merging (MVNM), that merges nodes within a brain region to construct informative inter-region brain networks. We employ this method to estimate functional (correlational) and effective (causal) networks using local field potentials (LFP) during a cognitive behavioral task. The effective networks computed using convergent cross mapping differentiate task engagement from background neural activity with 85% median classification accuracy. We also derive task engagement networks (TENs): networks that constitute the most discriminative inter-region connections. Subsequent graph analysis illustrates the crucial role of the dorsolateral prefrontal cortex (dlPFC) in task engagement, consistent with a widely accepted model for cognition. We also show that task engagement is linked to prefrontal cortex theta (4-8 Hz) oscillations. We, therefore, identify objective biomarkers associated with task engagement. These approaches may generalize to other cognitive functions, forming the basis of a network-based approach to detecting and rectifying decision deficits.

Anti-seizure medication correlated changes of cortical morphology in childhood epilepsy with centrotemporal spikes.

To investigate the morphological changes of cerebral cortex correlating with anti-seizure medication in Childhood Epilepsy with Centrotemporal Spikes (CECTS), and their relationships with seizure control. This study included a total of 188 children, including 62 patients with CECTS taking anti-seizure drugs, 56 patients with drug-naive, and 70 healthy controls. A portion of cases were also followed-up for longitudinal analysis. Cortical morphological parameters were quantitatively measured by applying surface-based morphometry analysis to high-resolution three-dimension T1 weighted images. Among the three groups, the morphological indices were compared to quantify any cortical changes affected by seizures and medication. The relationships among anti-seizure medication, seizure controls and cortical morphometry were investigated using causal mediator analysis. The Rolandic cortex of the drug-naive patients showed abnormal cortical thickness by comparing with that of healthy controls, and thinning by comparing with that of patients with medication. The cortical thickness in the Rolandic regions was negatively correlated with duration of medication and duration of seizure-free. Longitudinal analysis further demonstrated that the thickness of Rolandic cortex thinned in post-medication state relative to the pre-medication state. Mediation analysis revealed that morphological alteration of the Rolandic cortex might act as a mediator in the path of anti-seizure medication on seizure control. Our findings highlighted that anti-seizure medication was associated with regression of abnormal increment of cortical thickness in the Rolandic regions in CECTS. The neuroanatomical alteration might be a mediating factor in the process of seizure control by anti-seizure medication.

Spatial fidelity of MEG/EEG source estimates: A general evaluation approach.

Low spatial resolution is often cited as the most critical limitation of magneto- and electroencephalography (MEG and EEG), but a unifying framework for quantifying the spatial fidelity of M/EEG source estimates has yet to be established; previous studies have focused on linear estimation methods under ideal scenarios without noise. Here we present an approach that quantifies the spatial fidelity of M/EEG estimates from simulated patch activations over the entire neocortex superposed on measured resting-state data. This approach grants more generalizability in the evaluation process that allows for, e.g., comparing linear and non-linear estimates in the whole brain for different signal-to-noise ratios (SNR), number of active sources and activation waveforms. Using this framework, we evaluated the MNE, dSPM, sLORETA, eLORETA, and MxNE methods and found that the spatial fidelity varies significantly with SNR, following a largely sigmoidal curve whose shape varies depending on which aspect of spatial fidelity that is being quantified and the source estimation method. We believe that these methods and results will be useful when interpreting M/EEG source estimates as well as in methods development.

Case Report of Dual-Site Neurostimulation and Chronic Recording of Cortico-Striatal Circuitry in a Patient With Treatment Refractory Obsessive Compulsive Disorder.

Psychiatric disorders are increasingly understood as dysfunctions of hyper- or hypoconnectivity in distributed brain circuits. A prototypical example is obsessive compulsive disorder (OCD), which has been repeatedly linked to hyper-connectivity of cortico-striatal-thalamo-cortical (CSTC) loops. Deep brain stimulation (DBS) and lesions of CSTC structures have shown promise for treating both OCD and related disorders involving over-expression of automatic/habitual behaviors. Physiologically, we propose that this CSTC hyper-connectivity may be reflected in high synchrony of neural firing between loop structures, which could be measured as coherent oscillations in the local field potential (LFP). Here we report the results from the pilot patient in an Early Feasibility study (https://clinicaltrials.gov/ct2/show/NCT03184454) in which we use the Medtronic Activa PC+ S device to simultaneously record and stimulate in the supplementary motor area (SMA) and ventral capsule/ventral striatum (VC/VS). We hypothesized that frequency-mismatched stimulation should disrupt coherence and reduce compulsive symptoms. The patient reported subjective improvement in OCD symptoms and showed evidence of improved cognitive control with the addition of cortical stimulation, but these changes were not reflected in primary rating scales specific to OCD and depression, or during blinded cortical stimulation. This subjective improvement was correlated with increased SMA and VC/VS coherence in the alpha, beta, and gamma bands, signals which persisted after correcting for stimulation artifacts. We discuss the implications of this research, and propose future directions for research in network modulation in OCD and more broadly across psychiatric disorders.

Seizure onset location shapes dynamics of initiation.

OBJECTIVE: Ictal electrographic patterns are widely thought to reflect underlying neural mechanisms of seizures. Here we studied the degree to which seizure patterns are consistent in a given patient, relate to particular brain regions and if two candidate biomarkers (high-frequency oscillations, HFOs; infraslow activity, ISA) and network activity, as assessed with cross-frequency interactions, can discriminate between seizure types. METHODS: We analyzed temporal changes in low and high frequency oscillations recorded during seizures, as well as phase-amplitude coupling (PAC) to monitor the interactions between delta/theta and ripple/fast ripple frequency bands at seizure onset. RESULTS: Seizures of multiple electrographic patterns were observed in a given patient and brain region. While there was an increase in HFO rate across different electrographic patterns, there are specific relationships between types of HFO activity and onset region. Similarly, changes in PAC dynamics were more closely related to seizure onset region than they were to electrographic patterns while ISA was a poor indicator for seizure onset. CONCLUSIONS: Our findings suggest that the onset region sculpts neurodynamics at seizure initiation and that unique features of the cytoarchitecture and/or connectivity of that region play a significant role in determining seizure mechanism. SIGNIFICANCE: To learn how seizures are initiated, researchers would do well to consider other aspects of their manifestation, in addition to their electrographic patterns. Examination of onset pattern in conjunction with the interactions between different oscillatory frequencies in the context of different brain regions might be more informative and lead to more reliable clinical inference as well as novel therapeutic approaches.

Reactivation of Motor-Related Gamma Activity in Human NREM Sleep.

Models of memory consolidation posit a central role for reactivation of brain activity patterns during sleep, especially in non-Rapid Eye Movement (NREM) sleep. While such "replay" of recent waking experiences has been well-demonstrated in rodents, electrophysiological evidence of reactivation in human sleep is still largely lacking. In this intracranial study in patients with epilepsy (N = 9) we explored the spontaneous electroencephalographic reactivation during sleep of spatial patterns of brain activity evoked by motor learning. We first extracted the gamma-band (60-140 Hz) patterns underlying finger movements during a tapping task and underlying no-movement during a short rest period just prior to the task, and trained a binary classifier to discriminate between motor movements vs. rest. We then used the trained model on NREM sleep data immediately after the task and on NREM sleep during a control sleep period preceding the task. Compared with the control sleep period, we found, at the subject level, an increase in the detection rate of motor-related patterns during sleep following the task, but without association with performance changes. These data provide electrophysiological support for the reoccurrence in NREM sleep of the neural activity related to previous waking experience, i.e. that a basic tenet of the reactivation theory does occur in human sleep.

Caudate stimulation enhances learning.

Neuromodulation is a promising treatment modality for disorders of learning and memory, offering the possibility of precise alteration of disordered neural circuits. Studies to date have failed to identify an optimal target and stimulation paradigm. Six epilepsy patients with depth electrodes implanted for seizure localization participated in our study. We recorded local field potentials from implanted electrodes while subjects participated in an associative learning task requiring them to learn an association between presented images and a button press. Three subjects participated in stimulation sessions during which caudate or putamen stimulation was delivered for some images during feedback after correct responses. Caudate stimulation enhanced learning. Both caudate and dorsolateral prefrontal cortex demonstrated a beta power increase during the feedback period of the learning task that was greater following correct than incorrect trials. In dorsolateral prefrontal cortex, this difference increased with learning and persisted beyond the end of the feedback period. Caudate stimulation was associated with increased dorsolateral prefrontal cortex beta power following feedback. These findings suggest that temporally specific caudate stimulation is a promising neuromodulation strategy to improve learning in disorders of learning and memory.

Decoding task engagement from distributed network electrophysiology in humans.

OBJECTIVE: Here, our objective was to develop a binary decoder to detect task engagement in humans during two distinct, conflict-based behavioral tasks. Effortful, goal-directed decision-making requires the coordinated action of multiple cognitive processes, including attention, working memory and action selection. That type of mental effort is often dysfunctional in mental disorders, e.g. when a patient attempts to overcome a depression or anxiety-driven habit but feels unable. If the onset of engagement in this type of focused mental activity could be reliably detected, decisional function might be augmented, e.g. through neurostimulation. However, there are no known algorithms for detecting task engagement with rapid time resolution. APPROACH: We defined a new network measure, fixed canonical correlation (FCCA), specifically suited for neural decoding applications. We extracted FCCA features from local field potential recordings in human volunteers to give a temporally continuous estimate of mental effort, defined by engagement in experimental conflict tasks. MAIN RESULTS: Using a small number of features per participant, we accurately decoded and distinguished task engagement from other mental activities. Further, the decoder distinguished between engagement in two different conflict-based tasks within seconds of their onset. SIGNIFICANCE: These results demonstrate that network-level brain activity can detect specific types of mental efforts. This could form the basis of a responsive intervention strategy for decision-making deficits.

Decoding Hidden Cognitive States From Behavior and Physiology Using a Bayesian Approach.

Cognitive processes, such as learning and cognitive flexibility, are both difficult to measure and to sample continuously using objective tools because cognitive processes arise from distributed, high-dimensional neural activity. For both research and clinical applications, that dimensionality must be reduced. To reduce dimensionality and measure underlying cognitive processes, we propose a modeling framework in which a cognitive process is defined as a low-dimensional dynamical latent variable-called a cognitive state, which links high-dimensional neural recordings and multidimensional behavioral readouts. This framework allows us to decompose the hard problem of modeling the relationship between neural and behavioral data into separable encoding-decoding approaches. We first use a state-space modeling framework, the behavioral decoder, to articulate the relationship between an objective behavioral readout (e.g., response times) and cognitive state. The second step, the neural encoder, involves using a generalized linear model (GLM) to identify the relationship between the cognitive state and neural signals, such as local field potential (LFP). We then use the neural encoder model and a Bayesian filter to estimate cognitive state using neural data (LFP power) to generate the neural decoder. We provide goodness-of-fit analysis and model selection criteria in support of the encoding-decoding result. We apply this framework to estimate an underlying cognitive state from neural data in human participants (N=8) performing a cognitive conflict task. We successfully estimated the cognitive state within the 95% confidence intervals of that estimated using behavior readout for an average of 90% of task trials across participants. In contrast to previous encoder-decoder models, our proposed modeling framework incorporates LFP spectral power to encode and decode a cognitive state. The framework allowed us to capture the temporal evolution of the underlying cognitive processes, which could be key to the development of closed-loop experiments and treatments.

Consistent linear and non-linear responses to invasive electrical brain stimulation across individuals and primate species with implanted electrodes.

BACKGROUND: Electrical neuromodulation via implanted electrodes is used in treating numerous neurological disorders, yet our knowledge of how different brain regions respond to varying stimulation parameters is sparse. OBJECTIVE/HYPOTHESIS: We hypothesized that the neural response to electrical stimulation is both region-specific and non-linearly related to amplitude and frequency. METHODS: We examined evoked neural responses following 400 ms trains of 10-400 Hz electrical stimulation ranging from 0.1 to 10 mA. We stimulated electrodes implanted in cingulate cortex (dorsal anterior cingulate and rostral anterior cingulate) and subcortical regions (nucleus accumbens, amygdala) of non-human primates (NHP, N = 4) and patients with intractable epilepsy (N = 15) being monitored via intracranial electrodes. Recordings were performed in prefrontal, subcortical, and temporal lobe locations. RESULTS: In subcortical regions as well as dorsal and rostral anterior cingulate cortex, response waveforms depended non-linearly on frequency (Pearson's linear correlation r < 0.39), but linearly on current (r > 0.58). These relationships between location, and input-output characteristics were similar in homologous brain regions with average Pearson's linear correlation values r > 0.75 between species and linear correlation values between participants r > 0.75 across frequency and current values per brain region. Evoked waveforms could be described by three main principal components (PCs) which allowed us to successfully predict response waveforms across individuals and across frequencies using PC strengths as functions of current and frequency using brain region specific regression models. CONCLUSIONS: These results provide a framework for creation of an atlas of input-output relationships which could be used in the principled selection of stimulation parameters per brain region.

Cortical Signal Suppression (CSS) for Detection of Subcortical Activity Using MEG and EEG.

Magnetoencephalography (MEG) and electroencephalography (EEG) use non-invasive sensors to detect neural currents. Since the contribution of superficial neural sources to the measured M/EEG signals are orders-of-magnitude stronger than the contribution of subcortical sources, most MEG and EEG studies have focused on cortical activity. Subcortical structures, however, are centrally involved in both healthy brain function as well as in many neurological disorders such as Alzheimer's disease and Parkinson's disease. In this paper, we present a method that can separate and suppress the cortical signals while preserving the subcortical contributions to the M/EEG data. The resulting signal subspace of the data mainly originates from subcortical structures. Our method works by utilizing short-baseline planar gradiometers with short-sighted sensitivity distributions as reference sensors for cortical activity. Since the method is completely data-driven, forward and inverse modeling are not required. In this study, we use simulations and auditory steady state response experiments in a human subject to demonstrate that the method can remove the cortical signals while sparing the subcortical signals. We also test our method on MEG data recorded in an essential tremor patient with a deep brain stimulation implant and show how it can be used to reduce the DBS artifact in the MEG data by ~ 99.9% without affecting low frequency brain rhythms.

The interactive electrode localization utility: software for automatic sorting and labeling of intracranial subdural electrodes.

PURPOSE: Existing methods for sorting, labeling, registering, and across-subject localization of electrodes in intracranial encephalography (iEEG) may involve laborious work requiring manual inspection of radiological images. METHODS: We describe a new open-source software package, the interactive electrode localization utility which presents a full pipeline for the registration, localization, and labeling of iEEG electrodes from CT and MR images. In addition, we describe a method to automatically sort and label electrodes from subdural grids of known geometry. RESULTS: We validated our software against manual inspection methods in twelve subjects undergoing iEEG for medically intractable epilepsy. Our algorithm for sorting and labeling performed correct identification on 96% of the electrodes. CONCLUSIONS: The sorting and labeling methods we describe offer nearly perfect performance and the software package we have distributed may simplify the process of registering, sorting, labeling, and localizing subdural iEEG grid electrodes by manual inspection.

Biologically inspired load balancing mechanism in neocortical competitive learning.

A unique delayed self-inhibitory pathway mediated by layer 5 Martinotti Cells was studied in a biologically inspired neural network simulation. Inclusion of this pathway along with layer 5 basket cell lateral inhibition caused balanced competitive learning, which led to the formation of neuronal clusters as were indeed reported in the same region. Martinotti pathway proves to act as a learning "conscience," causing overly successful regions in the network to restrict themselves and let others fire. It thus spreads connectivity more evenly throughout the net and solves the "dead unit" problem of clustering algorithms in a local and biologically plausible manner.

Constraining compartmental models using multiple voltage recordings and genetic algorithms.

Compartmental models with many nonlinearly and nonhomogeneous distributions of voltage-gated conductances are routinely used to investigate the physiology of complex neurons. However, the number of loosely constrained parameters makes manually constructing the desired model a daunting if not impossible task. Recently, progress has been made using automated parameter search methods, such as genetic algorithms (GAs). However, these methods have been applied to somatically recorded action potentials using relatively simple target functions. Using a genetic minimization algorithm and a reduced compartmental model based on a previously published model of layer 5 neocortical pyramidal neurons we compared the efficacy of five cost functions (based on the waveform of the membrane potential, the interspike interval, trajectory density, and their combinations) to constrain the model. When the model was constrained using somatic recordings only, a combined cost function was found to be the most effective. This combined cost function was then applied to investigate the contribution of dendritic and axonal recordings to the ability of the GA to constrain the model. The more recording locations from the dendrite and the axon that were added to the data set the better was the genetic minimization algorithm able to constrain the compartmental model. Based on these simulations we propose an experimental scheme that, in combination with a genetic minimization algorithm, may be used to constrain compartmental models of neurons.