Abstract:Brain-Computer Interfaces (BCI) aim to bypass the peripheral nervous system to link the brain to external devices via successful modeling of decoding mechanisms. BCI based on electrocorticogram or ECoG represent a viable compromise between clinical practicality, spatial resolution, and signal quality when it comes to extracellular electrical potentials from local neuronal assemblies. Classic analysis of ECoG traces usually falls under the umbrella of Time-Frequency decompositions with adaptations from Fourier analysis and wavelets as its most prominent variants. However, analyzing such high-dimensional, multivariate time series demands for specialized signal processing and neurophysiological principles. We propose a generative model for single-channel ECoGs that is able to fully characterize reoccurring rhythm-specific neuromodulations as weighted activations of prototypical templates over time. The set of timings, weights and indexes comprise a temporal marked point process (TMPP) that accesses a set of bases from vector spaces of different dimensions-a dictionary. The shallow nature of the model admits the equivalence between latent variables and representations. In this way, learning the model parameters is a case of unsupervised representation learning. We exploit principles of Minimum Description Length (MDL) encoding to effectively yield a data-driven framework where prototypical neuromodulations (not restricted to a particular duration) can be estimated alongside the timings and features of the TMPP. We validate the proposed methodology on discrimination of movement-related tasks utilizing 32-electrode grids implanted in the frontal cortex of six epileptic subjects. We show that the learned representations from the high-gamma band (85-145 Hz) are not only interpretable, but also discriminant in a lower dimensional space. The results also underscore the practicality of our algorithm, i.e., 2 main hyperparameters that can be readily set via neurophysiology, and emphasize the need of principled and interpretable representation learning in order to model encoding mechanisms in the brain.

A machine learning approach for real-time cortical state estimation

Discrimination of Movement-Related Cortical Potentials Exploiting Unsupervised Learned Representations From ECoGs

Real-Time Machine Learning Strategies for a New Kind of Neuroscience Experiments

Hidden-Markov Factor analysis as a spatiotemporal model for electrocorticography

Brain-state modeling using electroencephalography: Application to adaptive closed-loop neuromodulation for epilepsy

Modeling brain dynamic state changes with adaptive mixture independent component analysis

Implementation of Structural Synchrony and Linear Measures of Brain Network Connectivity for Real-Time State Estimation.

State-aware detection of sensory stimuli in the cortex of the awake mouse

Explicit Modeling of Brain State Duration Using Hidden Semi Markov Models in EEG Data

Estimation of neural dynamics from MEG/EEG cortical current density maps: application to the reconstruction of large-scale cortical synchrony

Evolutionary State-Space Model and Its Application to Time-Frequency Analysis of Local Field Potentials

Brain Waves Analysis Via a Non-Parametric Bayesian Mixture of Autoregressive Kernels

A hierarchical Bayesian approach for learning sparse spatio-temporal decompositions of multichannel EEG.

Spatiotemporal Dynamics of Human High Gamma Discriminate Naturalistic Behavioral States

Towards a dynamical understanding of microstate analysis of M/EEG data

Real-Time Non-Invasive Imaging and Detection of Spreading Depolarizations through EEG: An Ultra-Light Explainable Deep Learning Approach

Solving large-scale MEG/EEG source localization and functional connectivity problems simultaneously using state-space models

Latent dynamics of primary sensory cortical population activity is structured by fluctuations in the local field potential

Dynamic Tracking of Non-Stationarity in Human ECoG Activity

Online Estimating Pairwise Neuronal Functional Connectivity in Brain-Machine Interface

Precision data-driven modeling of cortical dynamics reveals idiosyncratic mechanisms underlying canonical oscillations