Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia

Francesco Carlo Morabito, Maurizio Campolo, Nadia Mammone, Mario Versaci, Silvana Franceschetti, Fabrizio Tagliavini, Vito Sofia, Daniela Fatuzzo, Antonio Gambardella, Angelo Labate, Laura Mumoli, Giovanbattista Gaspare Tripodi, Sara Gasparini, Vittoria Cianci, Chiara Sueri, Edoardo Ferlazzo, Umberto Aguglia

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89%, an average sensitivity of 92%, and an average specificity of 89% in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.

Original languageEnglish
Article number1650039
JournalInternational Journal of Neural Systems
Volume27
Issue number2
DOIs
Publication statusPublished - Mar 1 2017

Fingerprint

Electroencephalography
Supervised learning
Processing
Unsupervised learning
Multilayer neural networks
Deep learning
Wavelet transforms
Support vector machines
Labels
Multilayers
Classifiers
Entropy
Tuning

Keywords

  • Alzheimer's disease
  • CJD
  • classification
  • continuous wavelet transform
  • deep learning
  • dementia
  • EEG
  • subacute encephalopathies
  • SVM

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia. / Morabito, Francesco Carlo; Campolo, Maurizio; Mammone, Nadia; Versaci, Mario; Franceschetti, Silvana; Tagliavini, Fabrizio; Sofia, Vito; Fatuzzo, Daniela; Gambardella, Antonio; Labate, Angelo; Mumoli, Laura; Tripodi, Giovanbattista Gaspare; Gasparini, Sara; Cianci, Vittoria; Sueri, Chiara; Ferlazzo, Edoardo; Aguglia, Umberto.

In: International Journal of Neural Systems, Vol. 27, No. 2, 1650039, 01.03.2017.

Research output: Contribution to journalArticle

Morabito, FC, Campolo, M, Mammone, N, Versaci, M, Franceschetti, S, Tagliavini, F, Sofia, V, Fatuzzo, D, Gambardella, A, Labate, A, Mumoli, L, Tripodi, GG, Gasparini, S, Cianci, V, Sueri, C, Ferlazzo, E & Aguglia, U 2017, 'Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia', International Journal of Neural Systems, vol. 27, no. 2, 1650039. https://doi.org/10.1142/S0129065716500398
Morabito, Francesco Carlo ; Campolo, Maurizio ; Mammone, Nadia ; Versaci, Mario ; Franceschetti, Silvana ; Tagliavini, Fabrizio ; Sofia, Vito ; Fatuzzo, Daniela ; Gambardella, Antonio ; Labate, Angelo ; Mumoli, Laura ; Tripodi, Giovanbattista Gaspare ; Gasparini, Sara ; Cianci, Vittoria ; Sueri, Chiara ; Ferlazzo, Edoardo ; Aguglia, Umberto. / Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia. In: International Journal of Neural Systems. 2017 ; Vol. 27, No. 2.
@article{be0d7a78bc5b47c5842979acf5bdd01b,
title = "Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia",
abstract = "A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89{\%}, an average sensitivity of 92{\%}, and an average specificity of 89{\%} in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.",
keywords = "Alzheimer's disease, CJD, classification, continuous wavelet transform, deep learning, dementia, EEG, subacute encephalopathies, SVM",
author = "Morabito, {Francesco Carlo} and Maurizio Campolo and Nadia Mammone and Mario Versaci and Silvana Franceschetti and Fabrizio Tagliavini and Vito Sofia and Daniela Fatuzzo and Antonio Gambardella and Angelo Labate and Laura Mumoli and Tripodi, {Giovanbattista Gaspare} and Sara Gasparini and Vittoria Cianci and Chiara Sueri and Edoardo Ferlazzo and Umberto Aguglia",
year = "2017",
month = "3",
day = "1",
doi = "10.1142/S0129065716500398",
language = "English",
volume = "27",
journal = "International Journal of Neural Systems",
issn = "0129-0657",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "2",

}

TY - JOUR

T1 - Deep learning representation from electroencephalography of early-stage creutzfeldt-jakob disease and features for differentiation from rapidly progressive dementia

AU - Morabito, Francesco Carlo

AU - Campolo, Maurizio

AU - Mammone, Nadia

AU - Versaci, Mario

AU - Franceschetti, Silvana

AU - Tagliavini, Fabrizio

AU - Sofia, Vito

AU - Fatuzzo, Daniela

AU - Gambardella, Antonio

AU - Labate, Angelo

AU - Mumoli, Laura

AU - Tripodi, Giovanbattista Gaspare

AU - Gasparini, Sara

AU - Cianci, Vittoria

AU - Sueri, Chiara

AU - Ferlazzo, Edoardo

AU - Aguglia, Umberto

PY - 2017/3/1

Y1 - 2017/3/1

N2 - A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89%, an average sensitivity of 92%, and an average specificity of 89% in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.

AB - A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89%, an average sensitivity of 92%, and an average specificity of 89% in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.

KW - Alzheimer's disease

KW - CJD

KW - classification

KW - continuous wavelet transform

KW - deep learning

KW - dementia

KW - EEG

KW - subacute encephalopathies

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=85007400475&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007400475&partnerID=8YFLogxK

U2 - 10.1142/S0129065716500398

DO - 10.1142/S0129065716500398

M3 - Article

AN - SCOPUS:85007400475

VL - 27

JO - International Journal of Neural Systems

JF - International Journal of Neural Systems

SN - 0129-0657

IS - 2

M1 - 1650039

ER -