Graph variational auto-encoder for deriving EEG-based graph embedding

Behrouzi, Tina; Hatzinakos, Dimitrios

doi:10.1016/j.patcog.2021.108202

Cited by 24 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To classify mental arithmetic vs. rest, the stacked long-short term memory (LSTM) was applied to raw 1D EEG signals, resulting in an accuracy of 93.59% in the study by Ganguly et al (2020). Behrouzi and Hatzinakos, (2022) elaborated the benefits of deep learningbased graph variational auto-encoder on 2D representation to detect the task of mental arithmetic attaining a peak mean performance of 95%. Statistical features represented in 1D were input to k-NN for EEG baseline classification (eyes open vs. eyes closed) by Gopan et al (2016b) and a peak mean accuracy of 77.92% was obtained.…”

Section: Introductionmentioning

confidence: 99%

An investigation of the multi-dimensional (1D vs. 2D vs. 3D) analyses of EEG signals using traditional methods and deep learning-based methods

Shah

Sinha

2022

Front. Signal Process.

View full text Add to dashboard Cite

Electroencephalographic (EEG) signals are electrical signals generated in the brain due to cognitive activities. They are non-invasive and are widely used to assess neurodegenerative conditions, mental load, and sleep patterns. In this work, we explore the utility of representing the inherently single dimensional time-series in different dimensions such as 1D-feature vector, 2D-feature maps, and 3D-videos. The proposed methodology is applied to four diverse datasets: 1) EEG baseline, 2) mental arithmetic, 3) Parkinson’s disease, and 4) emotion dataset. For a 1D analysis, popular 1D features hand-crafted from the time-series are utilized for classification. This performance is compared against the data-driven approach of using raw time-series as the input to the deep learning framework. To assess the efficacy of 2D representation, 2D feature maps that utilize a combination of the Feature Pyramid Network (FPN) and Atrous Spatial Pyramid Pooling (ASPP) is proposed. This is compared against an approach utilizing a composite feature set consisting of 2D feature maps and 1D features. However, these approaches do not exploit spatial, spectral, and temporal characteristics simultaneously. To address this, 3D EEG videos are created by stacking spectral feature maps obtained from each sub-band per time frame in a temporal domain. The EEG videos are the input to a combination of the Convolution Neural Network (CNN) and Long–Short Term Memory (LSTM) for classification. Performances obtained using the proposed methodologies have surpassed the state-of-the-art for three of the classification scenarios considered in this work, namely, EEG baselines, mental arithmetic, and Parkinson’s disease. The video analysis resulted in 92.5% and 98.81% peak mean accuracies for the EEG baseline and EEG mental arithmetic, respectively. On the other hand, for distinguishing Parkinson’s disease from controls, a peak mean accuracy of 88.51% is achieved using traditional methods on 1D feature vectors. This illustrates that 3D and 2D feature representations are effective for those EEG data where topographical changes in brain activation regions are observed. However, in scenarios where topographical changes are not consistent across subjects of the same class, these methodologies fail. On the other hand, the 1D analysis proves to be significantly effective in the case involving changes in the overall activation of the brain due to varying degrees of deterioration.

show abstract

Section: Introductionmentioning

confidence: 99%

An investigation of the multi-dimensional (1D vs. 2D vs. 3D) analyses of EEG signals using traditional methods and deep learning-based methods

Shah

Sinha

2022

Front. Signal Process.

View full text Add to dashboard Cite

show abstract

“…A Gaussian distribution is often used in practical implementations, so that the loss function can be written as:

, where the regularization term pushes the distribution generated by the latent representations to approach a standard normal distribution. Variational AEs have been widely used for learning representations from biosignals to enhance biometric recognition performance [ 114 , 128 , 129 , 130 ]. In these studies, the variational AEs were integrated with other constructs such as autoregressive layers [ 114 ] and graph neural networks [ 128 ] to capture dynamics from EEG timeseries for diverse classification tasks [ 114 ] and to learn graph embedding from EEG functional connectivity input [ 128 ].…”

Section: Representation Learning In Cognitive Biometricsmentioning

confidence: 99%

“…Variational AEs have been widely used for learning representations from biosignals to enhance biometric recognition performance [ 114 , 128 , 129 , 130 ]. In these studies, the variational AEs were integrated with other constructs such as autoregressive layers [ 114 ] and graph neural networks [ 128 ] to capture dynamics from EEG timeseries for diverse classification tasks [ 114 ] and to learn graph embedding from EEG functional connectivity input [ 128 ]. Latent representations learned by variational AEs also show promising results in EEG-based emotion recognition [ 130 ] and ERP-based driver–vehicle interface systems [ 129 ].…”

Section: Representation Learning In Cognitive Biometricsmentioning

confidence: 99%

Representation Learning and Pattern Recognition in Cognitive Biometrics: A Survey

Wang

Yin

Zhu

et al. 2022

Sensors

View full text Add to dashboard Cite

Cognitive biometrics is an emerging branch of biometric technology. Recent research has demonstrated great potential for using cognitive biometrics in versatile applications, including biometric recognition and cognitive and emotional state recognition. There is a major need to summarize the latest developments in this field. Existing surveys have mainly focused on a small subset of cognitive biometric modalities, such as EEG and ECG. This article provides a comprehensive review of cognitive biometrics, covering all the major biosignal modalities and applications. A taxonomy is designed to structure the corresponding knowledge and guide the survey from signal acquisition and pre-processing to representation learning and pattern recognition. We provide a unified view of the methodological advances in these four aspects across various biosignals and applications, facilitating interdisciplinary research and knowledge transfer across fields. Furthermore, this article discusses open research directions in cognitive biometrics and proposes future prospects for developing reliable and secure cognitive biometric systems.

show abstract

“…In previous studies, EEG-FC has been shown to have characterizing properties for different individuals [ 8 ]. Commonly used FC indicators in identity recognition are Pearson’s correlation coefficient ( COR ), coherence ( COH ) [ 9 , 10 , 11 ], phase lock value ( PLV ) [ 4 , 12 , 13 , 14 , 15 , 16 ], mutual information index ( MI ) [ 9 ], and Granger causality index (GC) [ 8 ], etc. These features are often used in isolation, which limits the accuracy improvement.…”

Section: Introductionmentioning

confidence: 99%

Applying Multiple Functional Connectivity Features in GCN for EEG-Based Human Identification

Tian

et al. 2022

Brain Sciences

View full text Add to dashboard Cite

EEG-based human identification has gained a wide range of attention due to the further increase in demand for security. How to improve the accuracy of the human identification system is an issue worthy of attention. Using more features in the human identification system is a potential solution. However, too many features may cause overfitting, resulting in the decline of system accuracy. In this work, the graph convolutional neural network (GCN) was adopted for classification. Multiple features were combined and utilized as the structure matrix of the GCN. Because of the constant signal matrix, the training parameters would not increase as the structure matrix grows. We evaluated the classification accuracy on a classic public dataset. The results showed that utilizing multiple features of functional connectivity (FC) can improve the accuracy of the identity authentication system, the best results of which are at 98.56%. In addition, our methods showed less sensitivity to channel reduction. The method proposed in this paper combines different FCs and reaches high classification accuracy for unpreprocessed data, which inspires reducing the system cost in the actual human identification system.

show abstract

Graph variational auto-encoder for deriving EEG-based graph embedding

Cited by 24 publications

References 24 publications

An investigation of the multi-dimensional (1D vs. 2D vs. 3D) analyses of EEG signals using traditional methods and deep learning-based methods

An investigation of the multi-dimensional (1D vs. 2D vs. 3D) analyses of EEG signals using traditional methods and deep learning-based methods

Representation Learning and Pattern Recognition in Cognitive Biometrics: A Survey

Applying Multiple Functional Connectivity Features in GCN for EEG-Based Human Identification

Contact Info

Product

Resources

About