Time-Frequency Analysis of Scalp EEG With Hilbert-Huang Transform and Deep Learning

Zheng, Jingyi; Liang, Mingli; Sinha, Sujata; Ge, Linqiang; Yu, Wei; Ekstrom, Arne D.; Hsieh, Fushing

doi:10.1109/jbhi.2021.3110267

Cited by 16 publications

(8 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers recruited a total of 19 adults (7 females, 12 males) from the University of Arizona in this study 20 , 21 . Participants were asked to monitor the distances travelled: short (100 virtual meters) vs long (200 virtual meters) distances while navigating in the virtual reality.…”

Section: Assessment Of Proposed Methodsmentioning

confidence: 99%

CW_ICA: an efficient dimensionality determination method for independent component analysis

Yi,

Billor,

Ekstrom

et al. 2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

Independent component analysis (ICA) is a widely used blind source separation method for signal pre-processing. The determination of the number of independent components (ICs) is crucial for achieving optimal performance, as an incorrect choice can result in either under-decomposition or over-decomposition. In this study, we propose a robust method to automatically determine the optimal number of ICs, named the column-wise independent component analysis (CW_ICA). CW_ICA divides the mixed signals into two blocks and applies ICA separately to each block. A quantitative measure, derived from the rank-based correlation matrix computed from the ICs of the two blocks, is utilized to determine the optimal number of ICs. The proposed method is validated and compared with the existing determination methods using simulation and scalp EEG data. The results demonstrate that CW_ICA is a reliable and robust approach for determining the optimal number of ICs. It offers computational efficiency and can be seamlessly integrated with different ICA methods.

show abstract

Section: Assessment Of Proposed Methodsmentioning

confidence: 99%

CW_ICA: an efficient dimensionality determination method for independent component analysis

Yi,

Billor,

Ekstrom

et al. 2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…, A m ∈ P n along the geodesic that connects two points on the manifold, as illustrated in Figure 1 (B) with four data points as an example. The initial barycenter is set to be the first data point, S (1) := A 1 , and then the barycenter is updated as the middle point along the geodesic that connects S (1) and A 2 , i.e., S (2)…”

Section: Barycenter Estimation With Bw Distancementioning

confidence: 99%

“…The core of BCI system is the classifier that classify the brain signals into one of the commands for the external devices. These brain signals are typically captured through multi-channel Electroencephalography (EEG) 1,2 , functional magnetic resonance imaging (fMRI) 3,4 , and other neuroimaging techniques, which leads to BCI data being mostly spatial and temporal. To capture the spatial and temporal pattern of BCI data, covariance matrices are extensively employed, and BCI classifiers are trained to directly classify covariance matrices in the Riemannian geometry based framework [5][6][7] .…”

Section: Introductionmentioning

confidence: 99%

Towards Analysis of Covariance Matrices through Bures-Wasserstein Distance

Zheng,

Huang,

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Brain-computer interface (BCI) builds a bridge between human brain and external devices by translating brain signals into commands for devices to perform the user’s imagined action. The core of the BCI system is the classifier that labels the brain signals as the user’s imagined action. Among the array of techniques, those directly classifying covariance matrices using Riemannian geometry find broad applications not only in BCI but also in diverse domains like computer vision and biomedical imaging. However, the existing Riemannian-based methods exhibit limitations, including time-intensive computations, susceptibility to disturbances, and convergence challenges in scenarios involving high-dimensional matrices. In this paper, we tackles these issues by introducing the Bures-Wasserstein (BW) distance for analyzing positive semi-definite matrices. Both theoretical and computational aspects of BW distance are investigated, and three algorithms are proposed to estimate the Fréchet Mean (or barycenter) of a set of matrices efficiently and robustly. The BW distance and barycenter are further integrated into the Minimum Distance to Riemannian Mean classifier to explore their performance in classification tasks. Extensive simulations are conducted to evaluate the effectiveness and robustness of the BW distance and barycenter. Additionally, we showcase the superior classification performance and enhanced computational efficiency of BW distance through evaluations on five real datasets.

show abstract

“…Next are feature extraction methods such as principal component analysis (PCA) [12], fourier transform (FT) [13], [14], empirical mode decomposition (EMD) [15]- [18], wavelet transform (WT) [19]- [30] and tunable Q-factor wavelet decomposition (TQWT) [31]- [35]. Feature extraction from the raw EEG signal to train machine learning classification models can identify different states of epilepsy.…”

Section: Introductionmentioning

confidence: 99%

“…However, EMD is a time-frequency decomposition technique from a time domain perspective. The intrinsic oscillatory modes in EMD depend on the characteristics of the time scale, which is widely applied in nonstationary signal recognition [15]- [18]. To obtain sufficient information from different bands of wavelets, another method of EEG data analysis is the wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

Revised Tunable Q-Factor Wavelet Transform for EEG-Based Epileptic Seizure Detection

Liu

Zhu

et al. 2023

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Electroencephalogram (EEG) signals are an essential tool for the detection of epilepsy. Because of the complex time series and frequency features of EEG signals, traditional feature extraction methods have difficulty meeting the requirements of recognition performance. The tunable Q-factor wavelet transform (TQWT), which is a constant-Q transform that is easily invertible and modestly oversampled, has been successfully used for feature extraction of EEG signals. Because the constant-Q is set in advance and cannot be optimized, further applications of the TQWT are restricted. To solve this problem, the revised tunable Q-factor wavelet transform (RTQWT) is proposed in this paper. RTQWT is based on the weighted normalized entropy and overcomes the problems of a nontunable Q-factor and the lack of an optimized tunable criterion. In contrast to the continuous wavelet transform and the raw tunable Q-factor wavelet transform, the wavelet transform corresponding to the revised Q-factor, i.e., RTQWT, is sufficiently better adapted to the nonstationary nature of EEG signals. Therefore, the precise and specific characteristic subspaces obtained can improve the classification accuracy of EEG signals. The classification of the extracted features was performed using the decision tree, linear discriminant, naive Bayes, SVM and KNN classifiers. The performance of the new approach was tested by evaluating the accuracies of five time-frequency distributions: FT, EMD, DWT, CWT and TQWT. The experiments showed that the RTQWT proposed in this paper can be used to extract detailed features more effectively and improve the classification accuracy of EEG signals.

show abstract

Time-Frequency Analysis of Scalp EEG With Hilbert-Huang Transform and Deep Learning

Cited by 16 publications

References 36 publications

CW_ICA: an efficient dimensionality determination method for independent component analysis

CW_ICA: an efficient dimensionality determination method for independent component analysis

Towards Analysis of Covariance Matrices through Bures-Wasserstein Distance

Revised Tunable Q-Factor Wavelet Transform for EEG-Based Epileptic Seizure Detection

Contact Info

Product

Resources

About