Deakin University
Browse

File(s) under permanent embargo

Hybrid deep shallow network for assessment of depression using electroencephalogram signals

conference contribution
posted on 2020-01-01, 00:00 authored by A Qayyum, Imran RazzakImran Razzak, W Mumtaz
© 2020, Springer Nature Switzerland AG. Depression is a mental health disorder characterised by persistently depressed mood or loss of interest in activities resulting impairment in daily life significantly. Electroencephalography (EEG) can assist with the accurate diagnosis of depression. In this paper, we present two different hybrid deep learning models for classification and assessment of patient suffering with depression. We have combined convolutional neural network with Gated recurrent units (RGUs), thus the proposed network is shallow and much smaller in size in comparison to its counter LSTM network. In addition to this, proposed approach is less sensitive to parameter settings. Extensive experiments on EEG dataset shows that the proposed hybrid model achieve highest accuracy, f1 score 99.66%, 99.93% and 98.87%, 99.12% for eye open and eye close dataset respectively in comparison to state of the art methods. Based on high performance, the proposed hybrid approach can be used for assessment of depression for clinical applications and can deployed remotely in hospital or private clinics for clinical evaluation.

History

Volume

12534

Pagination

245-257

Location

Online from Bangkok, Thailand

Start date

2020-11-18

End date

2020-11-22

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783030638351

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

Yang H, Pasupa K, Leung AC-S, Kwok JT, Chan JH, King I

Title of proceedings

ICONIP 2020 : Proceedings of the 27th International Conference on Neural Information Processing

Event

Neural Information Processing. International Conference (27th : 2020 : Online from Bangkok, Thailand)

Publisher

Springer International Publishing

Place of publication

Cham, Switzerland

Series

Neural Information Processing International Conference

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC