In recent years, the research community around the globe has contributed significantly to improve the brain-computer interface based assistive technologies. Electroencephalographic brain-computer interface enables the person to communicate with the outside world by creating an advanced communication protocol between the brain and the computer. Motor imagery-based BCIs aim to predict the specific patterns elicited by imagining some planned movements. Standard BCI systems incorporate the use of spatial features from the motor cortex. However, several researchers claim to have the intercommunication of different brain regions during the motor task. Thus, a unique approach like brain connectivity is essential to extract the intercommunication of brain regions through several electrode channels during a MI task. In this work, brain effective connectivity has been estimated using partial directed coherence, and it has been used as the feature extraction method. An extensive 2-class motor imagery dataset from Physionet database incorporating 91 subjects has been used for the validation purposes. Our proposed work reached the average classification accuracy of 97.45% using an SVM classifier. The findings of this study revealed the significance of brain connectivity features over the conventional features extracted from a single brain region.


    Access

    Download


    Export, share and cite



    Title :

    Partial Directed Coherence for the Classification of Motor Imagery-Based Brain-Computer Interface


    Additional title:

    Advances in Engineering res



    Conference:

    Proceedings of the Multimedia University Engineering ; 2022 ; Cyberjaya and Melaka, Malaysia July 25, 2022 - July 27, 2022



    Publication date :

    2022-12-23


    Size :

    11 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English





    Kinesthetic Motor Imagery Based Brain-Computer Interface for Power Wheelchair Manoeuvring

    T., Jackie / M.P., Paulraj / Adom, A.H. et al. | BASE | 2018

    Free access

    Improving Motor Imagery Brain-Computer Interface Performance Through Data Screening

    Zheng, Shiwei / Jiang, Lizi / Mai, Xun et al. | Springer Verlag | 2023


    Improving Motor Imagery Brain-Computer Interface Performance Through Data Screening

    Zheng, Shiwei / Jiang, Lizi / Mai, Xun et al. | TIBKAT | 2023


    COHERENCE-BASED TARGET DETECTION AND CLASSIFICATION FOR SIDE SCAN SONAR IMAGERY

    Tucker, J.D. / Azimi-Sadjadi, M.R. | British Library Online Contents | 2008


    Classification of Motor Imagery Based on Sample Entropy

    Zhou, P. / Ge, J.-y. / Cao, H.-b. et al. | British Library Online Contents | 2008