EEG Signal Processing for Alzheimer’s Disorders Using Discrete Wavelet Transform and Machine Learning Approaches

AlSharabi, Khalil; Salamah, Yasser Bin; Abdurraqeeb, Akram M.; Aljalal, Majid; Alturki, Fahd A.

doi:10.1109/access.2022.3198988

Cited by 47 publications

(18 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…KNN: This method was published in. 44 The features calculated in this study were logarithmic band power, standard deviation, variance, kurtosis, average energy, root mean square, and norm. A KNN was chosen to be the classifier.…”

Section: Resultsmentioning

confidence: 99%

“…Even though the structure of the EEG_CNN 43 was easy, this model showed better performance than both SVM 31 and KNN. 44 As some special convolution operations used in models, like the depthwise convolution, the separable convolution in EEGNet, 45 dynamic routing in CapsNet, 46 graph convolution in DGCNN, 47 and auxiliary outputs in our D-Unet, the performance had been improved effectively.…”

Section: The Comparison Of Different Methodsmentioning

confidence: 99%

See 1 more Smart Citation

D‐Unet: A symmetric architecture of convolutional neural network with two auxiliary outputs for dementia recognition

Li,

Xia,

Zou

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Dementia‐associated disorders cause damage to the brains of patients and bring huge burdens to individuals and families. Electroencephalogram (EEG) monitoring is friendly to patients on account of low cost, non‐invasion, and objective. Event‐related potential (ERP) is a component of EEG that has huge potential to evaluate the cognitive function of the brain. In this study, we recorded the ERP from patients with dementia and healthy people, then proposed an ERP‐based deep learning method to realize dementia recognition via the model Dementia‐Unet (D‐Unet). To improve the performance of the model, on the base of the decoder and the primary classifier, we added new structures including a symmetric decoder and two auxiliary outputs. One of the auxiliary outputs was input reconstruction, and the other one was aimed at working like the primary classifier with the same task. The results of the experiment of four‐fold cross‐validation demonstrated the two auxiliary outputs improved the performance of the model effectively. When compared with some other machine learning methods and deep learning methods, our model obtained the best performance with an accuracy of 0.815, a precision of 0.829, a recall of 0.797, and an f1‐score of 0.812. Besides, we put up a complex training strategy with all outputs involved, but a simple testing strategy with only a primary classifier working to keep high performance but cut down the complexity burden during testing.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: The Comparison Of Different Methodsmentioning

confidence: 99%

D‐Unet: A symmetric architecture of convolutional neural network with two auxiliary outputs for dementia recognition

Li,

Xia,

Zou

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…Among many wavelet families, the Daubechies wavelets have the properties of orthogonality and efficiency in filter implementation and were used in EEG spectral analysis(Subasi, 2007; Zarjam et al, 2015). Specifically, Daubechies wavelet with four vanishing moments (Db4) was adopted due to its suitability for analyzing EEG signals in AD in several studies (AlSharabi et al, 2022; Fiscon et al, 2018; Ghorbanian et al, 2013; Petrosian et al, 2001; Polikar et al, 2007). The Wavelet Toolbox of MATLAB (The MathWorks, Natick, Massachusetts) was employed in this study.…”

Section: Methodsmentioning

confidence: 99%

Machine learning based on event-related oscillations of working memory differentiates between preclinical Alzheimer’s disease and neurotypical aging

Liao,

Martin,

Fakorede

et al. 2023

Preprint

View full text Add to dashboard Cite

There is increasing evidence of the usefulness of electroencephalography (EEG) as an early neurophysiological marker of preclinical AD. Our objective was to apply machine learning approaches on event-related oscillations to discriminate preclinical AD from neurotypical controls. Twenty-two preclinical AD participants who were cognitively normal with elevated amyloid and 21 cognitively normal with no elevated amyloid controls completed n-back working memory tasks (n= 0, 1, 2). EEG signals were recorded through a high-density sensor net. The event-related spectral changes were extracted using the discrete wavelet transform in the delta, theta, alpha, and beta bands. The support vector machine (SVM) machine learning method was employed to classify participants, and classification performance was assessed using the Area Under the Curve (AUC) metric. The relative power of the beta and delta bands outperformed other frequency bands with higher AUC values. The 2-back task obtained higher AUC values than the 0 and 1-back tasks. The highest AUC values were from the 2-back task beta band (AUC = 0.86) and delta bands (AUC = 0.85) nontarget data. This study demonstrates the promise of using machine learning on EEG event-related oscillations from working memory tasks to detect preclinical AD.

show abstract

“…In [12], a computer-centered diagnosis system was developed to diagnose Alzheimer's disease using EEG signals.…”

Section: Related Workmentioning

confidence: 99%

Explainable Data Poison Attacks on Human Emotion Evaluation Systems Based on EEG Signals

Zhang

Umar

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The major aim of this paper is to explain the data poisoning attacks using label-flipping during the training stage of the electroencephalogram (EEG) signal-based human emotion evaluation systems deploying Machine Learning models from the attackers' perspective. Human emotion evaluation using EEG signals has consistently attracted a lot of research attention. The identification of human emotional states based on EEG signals is effective to detect potential internal threats caused by insider individuals. Nevertheless, EEG signal-based human emotion evaluation systems have shown several vulnerabilities to data poison attacks. Besides, due to the instability and complexity of the EEG signals, it is challenging to explain and analyze how data poison attacks influence the decision process of EEG signal-based human emotion evaluation systems. In this paper, from the attackers' side, data poison attacks occurring in the training phases of six different Machine Learning models including Random Forest, Adaptive Boosting (AdaBoost), Extra Trees, XGBoost, Multilayer Perceptron (MLP), and K-Nearest Neighbors (KNN) intrude on the EEG signal-based human emotion evaluation systems using these Machine Learning models. This seeks to reduce the performance of the aforementioned Machine Learning models with regard to the classification task of 4 different human emotions using EEG signals. The findings of the experiments demonstrate that the suggested data poison assaults are model-independently successful, although various models exhibit varying levels of resilience to the attacks. In addition, the data poison attacks on the EEG signal-based human emotion evaluation systems are explained with several Explainable Artificial Intelligence (XAI) methods including Shapley Additive Explanation (SHAP) values, Local Interpretable Model-agnostic Explanations (LIME), and Generated Decision Trees. And the codes of this paper are publicly available on GitHub.

show abstract

EEG Signal Processing for Alzheimer’s Disorders Using Discrete Wavelet Transform and Machine Learning Approaches

Cited by 47 publications

References 35 publications

D‐Unet: A symmetric architecture of convolutional neural network with two auxiliary outputs for dementia recognition

D‐Unet: A symmetric architecture of convolutional neural network with two auxiliary outputs for dementia recognition

Machine learning based on event-related oscillations of working memory differentiates between preclinical Alzheimer’s disease and neurotypical aging

Explainable Data Poison Attacks on Human Emotion Evaluation Systems Based on EEG Signals

Contact Info

Product

Resources

About