EEG-Based Neonatal Sleep-Wake Classification Using Multilayer Perceptron Neural Network

Saadullah Farooq Abbasi; Jawad Ahmad; Ahsen Tahir; Muhammad Awais; Chen Chen; Muhammad Irfan; Hafiza Ayesha Siddiqa; Abu Bakar Waqas; Xi Long; Bin Yin; Saeed Akbarzadeh; Chunmei Lu; Wei Chen; Laishuan S Wang

doi:10.1109/ACCESS.2020.3028182

EEG-Based Neonatal Sleep-Wake Classification Using Multilayer Perceptron Neural Network

Saadullah Farooq Abbasi
, Jawad Ahmad
, Ahsen Tahir
, Muhammad Awais
, Chen Chen
, Muhammad Irfan
, Hafiza Ayesha Siddiqa
, Abu Bakar Waqas
, Xi Long
, Bin Yin
, Saeed Akbarzadeh
, Chunmei Lu
, Wei Chen
, Laishuan S Wang

Electronic, Electrical and Systems Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Objective: Classification of sleep-wake states using multichannel electroencephalography (EEG) data that reliably work for neonates.

Methods: A deep multilayer perceptron (MLP) neural network is developed to classify sleep-wake states using multichannel bipolar EEG signals, which takes an input vector of size 108 containing the joint features of 9 channels. The network avoids any post-processing step in order to work as a full-fledged real-time application. For training and testing the model, EEG recordings of 3525 30-second segments from 19 neonates (postmenstrual age of 37 ± 05 weeks) are used.

Results: For sleep-wake classification, mean Cohen’s kappa between the network estimate and the ground truth annotation by human experts is 0.62. The maximum mean accuracy can reach up to 83% which, to date, is the highest accuracy for sleep-wake classification.

Original language	English
Pages (from-to)	183025-183034
Journal	IEEE Access
Volume	8
DOIs	https://doi.org/10.1109/ACCESS.2020.3028182
Publication status	Published - 1 Oct 2020

Access to Document

10.1109/ACCESS.2020.3028182Licence: Creative Commons: Attribution (CC BY)

Cite this

@article{39fb5735fa944f9e987913845efc43f7,

title = "EEG-Based Neonatal Sleep-Wake Classification Using Multilayer Perceptron Neural Network",

abstract = "Objective: Classification of sleep-wake states using multichannel electroencephalography (EEG) data that reliably work for neonates. Methods: A deep multilayer perceptron (MLP) neural network is developed to classify sleep-wake states using multichannel bipolar EEG signals, which takes an input vector of size 108 containing the joint features of 9 channels. The network avoids any post-processing step in order to work as a full-fledged real-time application. For training and testing the model, EEG recordings of 3525 30-second segments from 19 neonates (postmenstrual age of 37 ± 05 weeks) are used. Results: For sleep-wake classification, mean Cohen{\textquoteright}s kappa between the network estimate and the ground truth annotation by human experts is 0.62. The maximum mean accuracy can reach up to 83\% which, to date, is the highest accuracy for sleep-wake classification.",

author = "Abbasi, \{Saadullah Farooq\} and Jawad Ahmad and Ahsen Tahir and Muhammad Awais and Chen Chen and Muhammad Irfan and Siddiqa, \{Hafiza Ayesha\} and Waqas, \{Abu Bakar\} and Xi Long and Bin Yin and Saeed Akbarzadeh and Chunmei Lu and Wei Chen and Wang, \{Laishuan S\}",

year = "2020",

month = oct,

day = "1",

doi = "10.1109/ACCESS.2020.3028182",

language = "English",

volume = "8",

pages = "183025--183034",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "IEEE Xplore",

}

TY - JOUR

T1 - EEG-Based Neonatal Sleep-Wake Classification Using Multilayer Perceptron Neural Network

AU - Abbasi, Saadullah Farooq

AU - Ahmad, Jawad

AU - Tahir, Ahsen

AU - Awais, Muhammad

AU - Chen, Chen

AU - Irfan, Muhammad

AU - Siddiqa, Hafiza Ayesha

AU - Waqas, Abu Bakar

AU - Long, Xi

AU - Yin, Bin

AU - Akbarzadeh, Saeed

AU - Lu, Chunmei

AU - Chen, Wei

AU - Wang, Laishuan S

PY - 2020/10/1

Y1 - 2020/10/1

N2 - Objective: Classification of sleep-wake states using multichannel electroencephalography (EEG) data that reliably work for neonates. Methods: A deep multilayer perceptron (MLP) neural network is developed to classify sleep-wake states using multichannel bipolar EEG signals, which takes an input vector of size 108 containing the joint features of 9 channels. The network avoids any post-processing step in order to work as a full-fledged real-time application. For training and testing the model, EEG recordings of 3525 30-second segments from 19 neonates (postmenstrual age of 37 ± 05 weeks) are used. Results: For sleep-wake classification, mean Cohen’s kappa between the network estimate and the ground truth annotation by human experts is 0.62. The maximum mean accuracy can reach up to 83% which, to date, is the highest accuracy for sleep-wake classification.

AB - Objective: Classification of sleep-wake states using multichannel electroencephalography (EEG) data that reliably work for neonates. Methods: A deep multilayer perceptron (MLP) neural network is developed to classify sleep-wake states using multichannel bipolar EEG signals, which takes an input vector of size 108 containing the joint features of 9 channels. The network avoids any post-processing step in order to work as a full-fledged real-time application. For training and testing the model, EEG recordings of 3525 30-second segments from 19 neonates (postmenstrual age of 37 ± 05 weeks) are used. Results: For sleep-wake classification, mean Cohen’s kappa between the network estimate and the ground truth annotation by human experts is 0.62. The maximum mean accuracy can reach up to 83% which, to date, is the highest accuracy for sleep-wake classification.

U2 - 10.1109/ACCESS.2020.3028182

DO - 10.1109/ACCESS.2020.3028182

M3 - Article

SN - 2169-3536

VL - 8

SP - 183025

EP - 183034

JO - IEEE Access

JF - IEEE Access

ER -

EEG-Based Neonatal Sleep-Wake Classification Using Multilayer Perceptron Neural Network

Abstract

Access to Document

Fingerprint

Cite this