M/EEG-Based Bio-Markers to Predict the MCI and Alzheimer's Disease: A Review From the ML Perspective

Research output: Contribution to journalArticle

Abstract

This work reviews the state-of-the-art neuromarkers development for the prognosis of Alzheimer’s diseases (AD) and mild cognitive impairment (MCI). The first part of this study is devoted to reviewing the recently emerged machine learning (ML) algorithms based on electroencephalography (EEG) and magnetoencephalography (MEG) modalities. In particular, the methods are categorized by different types of neuromarkers. The second part of the review is dedicated to a series of investigations that further highlight the differences between these two modalities. Firstly, several source reconstruction methods are reviewed and their source-level performances explored, followed by an objective comparison between EEG and MEG from multiple perspectives. Finally, a number of the most recent reports on classification of MCI/AD during resting state using EEG/MEG are documented to show the up-to-date performance for this well-recognized data collecting scenario. It is noticed that the MEG modality may be particularly effective in distinguishing between subjects with MCI and healthy controls, a high classification accuracy of more than 98% was reported recently; whereas the EEG seems to be performing well in classifying AD and healthy subjects, which also reached around 98% of the accuracy. A number of influential factors have also been raised and suggested for careful considerations while evaluating the ML-based diagnosis systems in the real-world scenarios.
LanguageEnglish
Pages2924-2935
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume66
Issue number10
Early online date12 Feb 2019
DOIs
Publication statusPublished - 31 Oct 2019

Fingerprint

Magnetoencephalography
Electroencephalography
Learning systems
Learning algorithms

Keywords

  • Alzheimer's Disease
  • Mild cognitive impairment
  • MEG
  • EEG
  • Biomarkers
  • neuromarkers
  • biomarkers
  • mild cognitive impairment
  • Alzheimer's disease

Cite this

@article{435528c6b19c4055a281cd5ee67ed642,
title = "M/EEG-Based Bio-Markers to Predict the MCI and Alzheimer's Disease: A Review From the ML Perspective",
abstract = "This work reviews the state-of-the-art neuromarkers development for the prognosis of Alzheimer’s diseases (AD) and mild cognitive impairment (MCI). The first part of this study is devoted to reviewing the recently emerged machine learning (ML) algorithms based on electroencephalography (EEG) and magnetoencephalography (MEG) modalities. In particular, the methods are categorized by different types of neuromarkers. The second part of the review is dedicated to a series of investigations that further highlight the differences between these two modalities. Firstly, several source reconstruction methods are reviewed and their source-level performances explored, followed by an objective comparison between EEG and MEG from multiple perspectives. Finally, a number of the most recent reports on classification of MCI/AD during resting state using EEG/MEG are documented to show the up-to-date performance for this well-recognized data collecting scenario. It is noticed that the MEG modality may be particularly effective in distinguishing between subjects with MCI and healthy controls, a high classification accuracy of more than 98{\%} was reported recently; whereas the EEG seems to be performing well in classifying AD and healthy subjects, which also reached around 98{\%} of the accuracy. A number of influential factors have also been raised and suggested for careful considerations while evaluating the ML-based diagnosis systems in the real-world scenarios.",
keywords = "Alzheimer's Disease, Mild cognitive impairment, MEG, EEG, Biomarkers, neuromarkers, biomarkers, mild cognitive impairment, Alzheimer's disease",
author = "Su Yang and {Sanchez Bornot}, Jose and KongFatt Wong-Lin and Girijesh Prasad",
year = "2019",
month = "10",
day = "31",
doi = "10.1109/TBME.2019.2898871",
language = "English",
volume = "66",
pages = "2924--2935",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
number = "10",

}

TY - JOUR

T1 - M/EEG-Based Bio-Markers to Predict the MCI and Alzheimer's Disease: A Review From the ML Perspective

AU - Yang, Su

AU - Sanchez Bornot, Jose

AU - Wong-Lin, KongFatt

AU - Prasad, Girijesh

PY - 2019/10/31

Y1 - 2019/10/31

N2 - This work reviews the state-of-the-art neuromarkers development for the prognosis of Alzheimer’s diseases (AD) and mild cognitive impairment (MCI). The first part of this study is devoted to reviewing the recently emerged machine learning (ML) algorithms based on electroencephalography (EEG) and magnetoencephalography (MEG) modalities. In particular, the methods are categorized by different types of neuromarkers. The second part of the review is dedicated to a series of investigations that further highlight the differences between these two modalities. Firstly, several source reconstruction methods are reviewed and their source-level performances explored, followed by an objective comparison between EEG and MEG from multiple perspectives. Finally, a number of the most recent reports on classification of MCI/AD during resting state using EEG/MEG are documented to show the up-to-date performance for this well-recognized data collecting scenario. It is noticed that the MEG modality may be particularly effective in distinguishing between subjects with MCI and healthy controls, a high classification accuracy of more than 98% was reported recently; whereas the EEG seems to be performing well in classifying AD and healthy subjects, which also reached around 98% of the accuracy. A number of influential factors have also been raised and suggested for careful considerations while evaluating the ML-based diagnosis systems in the real-world scenarios.

AB - This work reviews the state-of-the-art neuromarkers development for the prognosis of Alzheimer’s diseases (AD) and mild cognitive impairment (MCI). The first part of this study is devoted to reviewing the recently emerged machine learning (ML) algorithms based on electroencephalography (EEG) and magnetoencephalography (MEG) modalities. In particular, the methods are categorized by different types of neuromarkers. The second part of the review is dedicated to a series of investigations that further highlight the differences between these two modalities. Firstly, several source reconstruction methods are reviewed and their source-level performances explored, followed by an objective comparison between EEG and MEG from multiple perspectives. Finally, a number of the most recent reports on classification of MCI/AD during resting state using EEG/MEG are documented to show the up-to-date performance for this well-recognized data collecting scenario. It is noticed that the MEG modality may be particularly effective in distinguishing between subjects with MCI and healthy controls, a high classification accuracy of more than 98% was reported recently; whereas the EEG seems to be performing well in classifying AD and healthy subjects, which also reached around 98% of the accuracy. A number of influential factors have also been raised and suggested for careful considerations while evaluating the ML-based diagnosis systems in the real-world scenarios.

KW - Alzheimer's Disease

KW - Mild cognitive impairment

KW - MEG

KW - EEG

KW - Biomarkers

KW - neuromarkers

KW - biomarkers

KW - mild cognitive impairment

KW - Alzheimer's disease

UR - https://ieeexplore.ieee.org/document/8640065

U2 - 10.1109/TBME.2019.2898871

DO - 10.1109/TBME.2019.2898871

M3 - Article

VL - 66

SP - 2924

EP - 2935

JO - IEEE Transactions on Biomedical Engineering

T2 - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 10

ER -