3D hand motion trajectory prediction from EEG mu and beta bandpower

Korik, Attila; Sosnik, Ronen; Siddique, Nazmul; Coyle, Damien

doi:10.1016/bs.pbr.2016.05.001

Cited by 21 publications

(26 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is commonly considered that time-series of the low delta band-pass filtered EEG potentials provides the best accuracy in the motion trajectory prediction based BCIs as it is hypothesized that the trajectory of the movement is coded in the theta band ( Robinson et al, 2015 ). However, this has been recently questioned when instead of times-series, time-varying spatiotemporal power distribution of theta, mu, and beta EEG oscillations is considered, being in line with the extensive use of mu and beta bands for classification of limbs of the multiclass sensorimotor rhythms based BCIs ( Korik et al, 2016a , b ). We provide evidence that during MI, delta EEG oscillation exhibits a phase-locking while alpha exhibits both power and phase-locking and beta exhibits power time-frequency mechanism ( Cebolla et al, 2015 ).…”

Section: Discussionmentioning

confidence: 99%

“…This is widely used in the multiclass sensorimotor rhythm non-invasive BCIs approach ( Coyle and Sosnik, 2015 ; Marchesotti et al, 2016 ; Zhang et al, 2016 ). It has also been shown that the spatiotemporal power distribution of theta, mu and beta oscillations used as input to a multiple linear regression based kinetic estimator or coupled to a feed-forward neural network gave better accuracy for motion trajectory prediction of hand motion BCIs than the potential time-series of the delta band ( Korik et al, 2016a , b ). Comparing MI to the related real movement highlights the confounding aspects of motor inhibition ( Angelini et al, 2015 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EEG Spectral Generators Involved in Motor Imagery: A swLORETA Study

et al. 2017

View full text Add to dashboard Cite

In order to characterize the neural generators of the brain oscillations related to motor imagery (MI), we investigated the cortical, subcortical, and cerebellar localizations of their respective electroencephalogram (EEG) spectral power and phase locking modulations. The MI task consisted in throwing a ball with the dominant upper limb while in a standing posture, within an ecological virtual reality (VR) environment (tennis court). The MI was triggered by the visual cues common to the control condition, during which the participant remained mentally passive. As previously developed, our paradigm considers the confounding problem that the reference condition allows two complementary analyses: one which uses the baseline before the occurrence of the visual cues in the MI and control resting conditions respectively; and the other which compares the analog periods between the MI and the control resting-state conditions. We demonstrate that MI activates specific, complex brain networks for the power and phase modulations of the EEG oscillations. An early (225 ms) delta phase-locking related to MI was generated in the thalamus and cerebellum and was followed (480 ms) by phase-locking in theta and alpha oscillations, generated in specific cortical areas and the cerebellum. Phase-locking preceded the power modulations (mainly alpha–beta ERD), whose cortical generators were situated in the frontal BA45, BA11, BA10, central BA6, lateral BA13, and posterior cortex BA2. Cerebellar-thalamic involvement through phase-locking is discussed as an underlying mechanism for recruiting at later stages the cortical areas involved in a cognitive role during MI.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

EEG Spectral Generators Involved in Motor Imagery: A swLORETA Study

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Recently, authors in Korik et al (2016, 2018) have advocated the utilization of the time-resolved power feature extracted from other frequency bands (e.g., the mu and beta bands, etc.) EEG signals for decoding the 3D executed or imagined movement trajectories.…”

Section: Methodsmentioning

confidence: 99%

“…In Kim et al (2015), the low-delta band EEG was further utilized for decoding complex 3D movement trajectories with non-linear models, in executed, observed/imagined complicated upper limb motor tasks. Researchers in Korik et al (2016, 2018) have also shown that the low-delta band EEG is informative about the 3D hand joint trajectories in either executed or imagined arm movements, though it is not the best representation for such a 3D imagined movement decoding task according to their experimental results. As for an executed 2D center-out reaching paradigm, the low frequency EEG signal obtained with wavelet analysis has succeeded to be applied for estimating the hand kinematics adaptively (Robinson et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

“…During the task, the EEG signals and the continuous hand movement parameters were recorded simultaneously. We firstly performed the neurophysiological analysis on the phase feature and the amplitude feature of the EEG slow oscillations on representative channels, and then compared the decoding accuracies of the low-delta EEG phase based linear decoder, the other bands EEG amplitude based counterparts, as well as the recently proposed mu/beta band power based ones by Korik et al (2016, 2018). To the best of our knowledge, our work is the first study to extend the application of such a slow-oscillation phase representation for decoding continuous movement parameters, with the non-invasive EEG signals.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Advantage of Low-Delta Electroencephalogram Phase Feature for Reconstructing the Center-Out Reaching Hand Movements

Zeng

Sun

et al. 2019

Front. Neurosci.

View full text Add to dashboard Cite

It is an emerging frontier of research on the use of neural signals for prosthesis control, in order to restore lost function to amputees and patients after spinal cord injury. Compared to the invasive neural signal based brain-machine interface (BMI), a non-invasive alternative, i.e., the electroencephalogram (EEG)-based BMI would be more widely accepted by the patients above. Ideally, a real-time continuous neuroprosthestic control is required for practical applications. However, conventional EEG-based BMIs mainly deal with the discrete brain activity classification. Until recently, the literature has reported several attempts for achieving the real-time continuous control by reconstructing the continuous movement parameters (e.g., speed, position, etc.) from the EEG recordings, and the low-frequency band EEG is consistently reported to encode the continuous motor control information. Previous studies with executed movement tasks have extensively relied on the amplitude representation of such slow oscillations of EEG signals for building models to decode kinematic parameters. Inspired by the recent successes of instantaneous phase of low-frequency invasive brain signals in the motor control and sensory processing domains, this study examines the extension of such a slow-oscillation phase representation to the reconstructing two-dimensional hand movements, with the non-invasive EEG signals for the first time. The data for analysis are collected on five healthy subjects performing 2D hand center-out reaching along four directions in two sessions. On representative channels over the cortices encoding the execution information of reaching movements, we show that the low-delta EEG phase representation is characterized by higher signal-to-noise ratio and stronger modulation by the movement tasks, compared to the low-delta EEG amplitude representation. Furthermore, we have tested the low-delta EEG phase representation with two commonly used linear decoding models. The results demonstrate that the low-delta EEG phase based decoders lead to superior performance for 2D executed movement reconstruction to its amplitude based counterparts, as well as the other-frequency band amplitude and power based features. Thus, our study contributes to improve the movement reconstruction from EEG by introducing a new feature set based on the low-delta EEG phase patterns, and demonstrates its potential for continuous fine motion control of neuroprostheses.

show abstract

DAMTRNN: A Delta Attention-Based Multi-task RNN for Intention Recognition

Chen

Yue

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

3D hand motion trajectory prediction from EEG mu and beta bandpower

Cited by 21 publications

References 50 publications

EEG Spectral Generators Involved in Motor Imagery: A swLORETA Study

EEG Spectral Generators Involved in Motor Imagery: A swLORETA Study

The Advantage of Low-Delta Electroencephalogram Phase Feature for Reconstructing the Center-Out Reaching Hand Movements

DAMTRNN: A Delta Attention-Based Multi-task RNN for Intention Recognition

Contact Info

Product

Resources

About