Modeling brain dynamic state changes with adaptive mixture independent component analysis

Abstract

There is a growing interest in neuroscience in assessing the continuous, endogenous, and nonstationary dynamics of brain network activity supporting the fluidity of human cognition and behavior. This non-stationarity may involve ever-changing formation and dissolution of active cortical sources and brain networks. However, unsupervised approaches to identify and model these changes in brain dynamics as continuous transitions between quasi-stable brain states using unlabeled, noninvasive recordings of brain activity have been limited. This study explores the use of adaptive mixture independent component analysis (AMICA) to model multichannel electroencephalographic (EEG) data with a set of ICA models, each of which decomposes an adaptively learned portion of the data into statistically independent sources. We first show that AMICA can segment simulated quasi-stationary EEG data and accurately identify ground-truth sources and source model transitions. Next, we demonstrate that AMICA decomposition, applied to 6-13 channel scalp recordings from the CAP Sleep Database, can characterize sleep stage dynamics, allowing 75% accuracy in identifying transitions between six sleep stages without use of EEG power spectra. Finally, applied to 30-channel data from subjects in a driving simulator, AMICA identifies models that account for EEG during faster and slower response to driving challenges, respectively. We show changes in relative probabilities of these models allow effective prediction of subject response speed and moment-by-moment characterization of state changes within single trials. AMICA thus provides a generic unsupervised approach to identifying and modeling changes in EEG dynamics. Applied to continuous, unlabeled multichannel data, AMICA may likely be used to detect and study any changes in cognitive states.

Prediction of gait intention from pre-movement EEG signals: a feasibility study

“…AMICA achieves better ICA decomposition than other ICA approaches as reported in [50]. Moreover, multi-model AMICA can be used as a data-driven approach to address the non-stationarity and dynamic changes of continuous EEG data [51]. The resulting independent components thus obtained were then inspected, and artefactual IC's were rejected by visual inspection.…”

Alteration of brain dynamics during natural dual-task walking

“…muscle activity, noise) by visual inspection. Then, for the identification and removal of eye blink artifacts we utilized the procedure of adaptive mixture independent component analysis (AMICA, Palmer et al, 2012;Hsu et al, 2018). To this end, the raw data were bandpass filtered to 1 Hz -90 Hz to improve the decomposition to independent components (ICs) (Winkler et al, 2015).…”

Functional near-infrared spectroscopy in conjunction with electroencephalography of cerebellar transcranial direct current stimulation responses in the latent neurovascular coupling space – a chronic stroke study

“…EEG data was first down-sampled to 250Hz and then was de-trended by high-pass filtering with 1 Hz cutoff frequency, line noise, and its harmonics were removed using "CleanLine," and then EEG data were re-referenced to the common average. Then, the continuous EEG data were processed with Artifact Subspace Reconstruction (ASR) [60], and then Adaptive Mixture Independent Component Analysis (AMICA) was performed in EEGlab [61]. The artifactual components were removed by IClabel [62] and visual inspection [63], which made the fNIRS-EEG data discontinuous with the discontinuity marked as events in the EEGlab.…”