A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia

Cosimo Ieracitano, Nadia Mammone, Amir Hussain, Francesco C. Morabito

Research output: Contribution to journalArticle

Abstract

Electroencephalographic (EEG) recordings generate an electrical map of the human brain that are useful for clinical inspection of patients and in biomedical smart Internet-of-Things (IoT) and Brain-Computer Interface (BCI) applications. From a signal processing perspective, EEGs yield a nonlinear and nonstationary, multivariate representation of the underlying neural circuitry interactions. In this paper, a novel multi-modal Machine Learning (ML) based approach is proposed to integrate EEG engineered features for automatic classification of brain states. EEGs are acquired from neurological patients with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) and the aim is to discriminate Healthy Control (HC) subjects from patients. Specifically, in order to effectively cope with nonstationarities, 19-channels EEG signals are projected into the time–frequency (TF) domain by means of the Continuous Wavelet Transform (CWT) and a set of appropriate features (denoted as CWT features) are extracted from δ, θ, α1, α2, β EEG sub-bands. Furthermore, to exploit nonlinear phase-coupling information of EEG signals, higher order statistics (HOS) are extracted from the bispectrum (BiS) representation. BiS generates a second set of features (denoted as BiS features) which are also evaluated in the five EEG sub-bands. The CWT and BiS features are fed into a number of ML classifiers to perform both 2-way (AD vs. HC, AD vs. MCI, MCI vs. HC) and 3-way (AD vs. MCI vs. HC) classifications. As an experimental benchmark, a balanced EEG dataset that includes 63 AD, 63 MCI and 63 HC is analyzed. Comparative results show that when the concatenation of CWT and BiS features (denoted as multi-modal (CWT+BiS) features) is used as input, the Multi-Layer Perceptron (MLP) classifier outperforms all other models, specifically, the Autoencoder (AE), Logistic Regression (LR) and Support Vector Machine (SVM). Consequently, our proposed multi-modal ML scheme can be considered a viable alternative to state-of-the-art computationally intensive deep learning approaches.

Original languageEnglish
Pages (from-to)176-190
Number of pages15
JournalNeural Networks
Volume123
DOIs
Publication statusPublished - Mar 2020

Fingerprint

Wavelet Analysis
Wavelet transforms
Dementia
Learning systems
Alzheimer Disease
Electroencephalography
Brain
Classifiers
Disease control
Brain-Computer Interfaces
Higher order statistics
Benchmarking
Brain computer interface
Neural Networks (Computer)
Multilayer neural networks
Internet
Support vector machines
Logistics
Healthy Volunteers
Signal processing

Keywords

  • Alzheimer's disease
  • Bispectrum
  • Continuous wavelet transform
  • Data fusion
  • Machine learning
  • Mild cognitive impairment

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia. / Ieracitano, Cosimo; Mammone, Nadia; Hussain, Amir; Morabito, Francesco C.

In: Neural Networks, Vol. 123, 03.2020, p. 176-190.

Research output: Contribution to journalArticle

Ieracitano, Cosimo ; Mammone, Nadia ; Hussain, Amir ; Morabito, Francesco C. / A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia. In: Neural Networks. 2020 ; Vol. 123. pp. 176-190.
@article{aa2e28be4850484f9b2e70549435bd2d,
title = "A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia",
abstract = "Electroencephalographic (EEG) recordings generate an electrical map of the human brain that are useful for clinical inspection of patients and in biomedical smart Internet-of-Things (IoT) and Brain-Computer Interface (BCI) applications. From a signal processing perspective, EEGs yield a nonlinear and nonstationary, multivariate representation of the underlying neural circuitry interactions. In this paper, a novel multi-modal Machine Learning (ML) based approach is proposed to integrate EEG engineered features for automatic classification of brain states. EEGs are acquired from neurological patients with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) and the aim is to discriminate Healthy Control (HC) subjects from patients. Specifically, in order to effectively cope with nonstationarities, 19-channels EEG signals are projected into the time–frequency (TF) domain by means of the Continuous Wavelet Transform (CWT) and a set of appropriate features (denoted as CWT features) are extracted from δ, θ, α1, α2, β EEG sub-bands. Furthermore, to exploit nonlinear phase-coupling information of EEG signals, higher order statistics (HOS) are extracted from the bispectrum (BiS) representation. BiS generates a second set of features (denoted as BiS features) which are also evaluated in the five EEG sub-bands. The CWT and BiS features are fed into a number of ML classifiers to perform both 2-way (AD vs. HC, AD vs. MCI, MCI vs. HC) and 3-way (AD vs. MCI vs. HC) classifications. As an experimental benchmark, a balanced EEG dataset that includes 63 AD, 63 MCI and 63 HC is analyzed. Comparative results show that when the concatenation of CWT and BiS features (denoted as multi-modal (CWT+BiS) features) is used as input, the Multi-Layer Perceptron (MLP) classifier outperforms all other models, specifically, the Autoencoder (AE), Logistic Regression (LR) and Support Vector Machine (SVM). Consequently, our proposed multi-modal ML scheme can be considered a viable alternative to state-of-the-art computationally intensive deep learning approaches.",
keywords = "Alzheimer's disease, Bispectrum, Continuous wavelet transform, Data fusion, Machine learning, Mild cognitive impairment",
author = "Cosimo Ieracitano and Nadia Mammone and Amir Hussain and Morabito, {Francesco C.}",
year = "2020",
month = "3",
doi = "10.1016/j.neunet.2019.12.006",
language = "English",
volume = "123",
pages = "176--190",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia

AU - Ieracitano, Cosimo

AU - Mammone, Nadia

AU - Hussain, Amir

AU - Morabito, Francesco C.

PY - 2020/3

Y1 - 2020/3

N2 - Electroencephalographic (EEG) recordings generate an electrical map of the human brain that are useful for clinical inspection of patients and in biomedical smart Internet-of-Things (IoT) and Brain-Computer Interface (BCI) applications. From a signal processing perspective, EEGs yield a nonlinear and nonstationary, multivariate representation of the underlying neural circuitry interactions. In this paper, a novel multi-modal Machine Learning (ML) based approach is proposed to integrate EEG engineered features for automatic classification of brain states. EEGs are acquired from neurological patients with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) and the aim is to discriminate Healthy Control (HC) subjects from patients. Specifically, in order to effectively cope with nonstationarities, 19-channels EEG signals are projected into the time–frequency (TF) domain by means of the Continuous Wavelet Transform (CWT) and a set of appropriate features (denoted as CWT features) are extracted from δ, θ, α1, α2, β EEG sub-bands. Furthermore, to exploit nonlinear phase-coupling information of EEG signals, higher order statistics (HOS) are extracted from the bispectrum (BiS) representation. BiS generates a second set of features (denoted as BiS features) which are also evaluated in the five EEG sub-bands. The CWT and BiS features are fed into a number of ML classifiers to perform both 2-way (AD vs. HC, AD vs. MCI, MCI vs. HC) and 3-way (AD vs. MCI vs. HC) classifications. As an experimental benchmark, a balanced EEG dataset that includes 63 AD, 63 MCI and 63 HC is analyzed. Comparative results show that when the concatenation of CWT and BiS features (denoted as multi-modal (CWT+BiS) features) is used as input, the Multi-Layer Perceptron (MLP) classifier outperforms all other models, specifically, the Autoencoder (AE), Logistic Regression (LR) and Support Vector Machine (SVM). Consequently, our proposed multi-modal ML scheme can be considered a viable alternative to state-of-the-art computationally intensive deep learning approaches.

AB - Electroencephalographic (EEG) recordings generate an electrical map of the human brain that are useful for clinical inspection of patients and in biomedical smart Internet-of-Things (IoT) and Brain-Computer Interface (BCI) applications. From a signal processing perspective, EEGs yield a nonlinear and nonstationary, multivariate representation of the underlying neural circuitry interactions. In this paper, a novel multi-modal Machine Learning (ML) based approach is proposed to integrate EEG engineered features for automatic classification of brain states. EEGs are acquired from neurological patients with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) and the aim is to discriminate Healthy Control (HC) subjects from patients. Specifically, in order to effectively cope with nonstationarities, 19-channels EEG signals are projected into the time–frequency (TF) domain by means of the Continuous Wavelet Transform (CWT) and a set of appropriate features (denoted as CWT features) are extracted from δ, θ, α1, α2, β EEG sub-bands. Furthermore, to exploit nonlinear phase-coupling information of EEG signals, higher order statistics (HOS) are extracted from the bispectrum (BiS) representation. BiS generates a second set of features (denoted as BiS features) which are also evaluated in the five EEG sub-bands. The CWT and BiS features are fed into a number of ML classifiers to perform both 2-way (AD vs. HC, AD vs. MCI, MCI vs. HC) and 3-way (AD vs. MCI vs. HC) classifications. As an experimental benchmark, a balanced EEG dataset that includes 63 AD, 63 MCI and 63 HC is analyzed. Comparative results show that when the concatenation of CWT and BiS features (denoted as multi-modal (CWT+BiS) features) is used as input, the Multi-Layer Perceptron (MLP) classifier outperforms all other models, specifically, the Autoencoder (AE), Logistic Regression (LR) and Support Vector Machine (SVM). Consequently, our proposed multi-modal ML scheme can be considered a viable alternative to state-of-the-art computationally intensive deep learning approaches.

KW - Alzheimer's disease

KW - Bispectrum

KW - Continuous wavelet transform

KW - Data fusion

KW - Machine learning

KW - Mild cognitive impairment

UR - http://www.scopus.com/inward/record.url?scp=85076854500&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076854500&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2019.12.006

DO - 10.1016/j.neunet.2019.12.006

M3 - Article

C2 - 31884180

AN - SCOPUS:85076854500

VL - 123

SP - 176

EP - 190

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

ER -