Literature DB >> 29890509

Detecting intention to execute the next movement while performing current movement from EEG using global optimal constrained ICA.

Elnaz Eilbeigi1, Seyed Kamaledin Setarehdan2.   

Abstract

BACKGROUND: Brain-computer interfaces (BCIs) are a promising tool in neurorehabilitation. The intention to perform a motor action can be detected from brain signals and used to control robotic devices. Most previous studies have focused on the starting of movements from a resting state, while in daily life activities, motions occur continuously and the neural activities correlated to the evolving movements are yet to be investigated.
METHOD: First we investigate the existence of neural correlates of intention to replace an object on the table during a holding phase. Next, we present a new method to extract the movement-related cortical potentials (MRCP) from a single-trial EEG. A novel method called Global optimal constrained ICA (GocICA) is proposed to overcome the limitations of cICA which is implemented based on Particle Swarm Optimization (PSO) and Charged System Search (CSS) techniques. GocICA is then utilized for decoding the intention to grasp and lift and intention to replace movements where the results were compared.
RESULTS: It was found that GocICA significantly improves the intention detection performance. Best results in offline detection were obtained with CSS-cICA for both kinds of intentions. Furthermore, pseudo-online decoding showed that GocICA was able to predict both intentions before the onset of related movements with the highest probability.
CONCLUSIONS: Decoding of the next movement intention during current movement is possible, which can be used to create more natural neuroprostheses. The results demonstrate that GocICA is a promising new algorithm for single-trial MRCP detection which can be used for detecting other types of ERPs such as P300.
Copyright © 2018 Elsevier Ltd. All rights reserved.

Entities:  

Keywords:  Brain-computer interface (BCI); Constrained independent component analysis (cICA); Global optimal constrained ICA (GocICA); High-level commands; Meta-heuristic optimization; Movement related cortical potential (MRCP); Natural neuroprostheses

Mesh:

Year:  2018        PMID: 29890509     DOI: 10.1016/j.compbiomed.2018.05.024

Source DB:  PubMed          Journal:  Comput Biol Med        ISSN: 0010-4825            Impact factor:   4.589


  2 in total

1.  Characteristics of Kinematic Parameters in Decoding Intended Reaching Movements Using Electroencephalography (EEG).

Authors:  Hyeonseok Kim; Natsue Yoshimura; Yasuharu Koike
Journal:  Front Neurosci       Date:  2019-11-01       Impact factor: 4.677

2.  Dimensionality reduction for classification of object weight from electromyography.

Authors:  Elnaz Lashgari; Uri Maoz
Journal:  PLoS One       Date:  2021-08-16       Impact factor: 3.240

  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.