A new blind source separation approach based on dynamical similarity and its application on epileptic seizure prediction

Niknazar, Hamid; Nasrabadi, Ali Motie; Shamsollahi, Mohammad Bagher

doi:10.1016/j.sigpro.2021.108045

Cited by 11 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the performance of the improved BSS approach based on generalized Jaccarcd similarity with that based on cosine similarity [25], cross-talking error [31] and RMSE are adopted as the measurement indexes, which have the following expressions respectively:…”

Section: Performance Comparisonmentioning

confidence: 99%

“…The quantification of dynamic similarity in signals is perceived as a challenging work in the BSS problem with Gaussian sources. Recently, Niknazar et al [25] proposed the fuzzy statistical behavior of local extreme (FSBLE) method based on cosine similarity to measure the dynamic similarity in signals, which solved the problem of isolating the linear chaotic Gaussian source signals and applied it to epileptic seizure prediction. Whereas, the cosine similarity suffers from the partial information loss since it is unable to make a distinction among the similarity vectors and unable to amplify the important part of the data object [12].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An improved blind Gaussian source separation approach based on generalized Jaccard similarity

Jianwei

2022

Preprint

View full text Add to dashboard Cite

Blind source separation (BSS) consists of recovering the independent source signals from their linear mixtures with unknown mixing channel. As a momentous technical means of signal processing and analysis, BSS has been widely utilized in image recognition, feature extraction, biomedical engineering, etcetera. The existing BSS approaches rely on the fundamental assumption: the source signals are non-Gaussian, this limited the use of BSS seriously. To overcome this problem and the weakness of cosine index in measuring the dynamic similarity between signals, this study proposes the fuzzy statistical behavior of local extremum (FSBLE) based on generalized Jaccard similarity as the signal's feature to implement the separation of nonlinear chaotic Gaussian source signals. In particular, the imperialist competition algorithm is introduced to minimize the cost function which jointly considers the stationarity factor describing the dynamical similarity of each source signal separately and the independency factor describing the dynamical similarity between source signals. Simulation experiments on synthetic data verify the effectiveness and superiority of the improved BSS approach in terms of cross-talking error and root mean square error (RMSE) criterions.

show abstract

Section: Performance Comparisonmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An improved blind Gaussian source separation approach based on generalized Jaccard similarity

Jianwei

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The simplest method for diagnosing epilepsy is to utilize non-invasive EEG to record the voltage of brain fluctuations [12]. EEG monitors continuous brain activity by inserting many electrodes at various locations across the brain and detecting voltage variations [13]. Even for highly skilled neurologists, proper diagnosis of EEG is time-consuming and challenging despite its great temporal resolution.…”

Section: Introductionmentioning

confidence: 99%

Prediction of the epileptic seizure through deep learning techniques using electroencephalography

Sunkara,

Rajakumari

2023

IJEECS

View full text Add to dashboard Cite

<span>Electroencephalography (EEG) is a widely used and significant technique for aiding in epilepsy diagnosis and investigating the electrical patterns of the human brain. Due to the non-stationary nature of EEG signals, seizure patterns will vary across different recording sessions for individual patients. In this study, a new deep learning long short-term memory (LSTM) model is implemented for the detection of brain tumors and seizures. The process consists of four key steps: EEG signal pre-processing, preictal feature extraction, hyper optimization using grey wolf optimization (GWO), and LSTM-based classification. The evaluation makes use of long-term EEG recordings from the EEG and ABIDE fMRI datasets. By experimenting with various modules and layers of memory units, a pre-analysis is first conducted to determine the best LSTM network architecture. The LSTM model makes use of numerous retrieved features, including temporal and frequency domain information between EEG channels that were extracted before classification. The discovery of the implemented methodology revealed significant advantages in accurately predicting seizures while minimizing false alarms. The implemented LSTM method achieves a 99% accuracy rate, 98% precision, 99% recall, and 98% f1-measure which is better when compared with cross sub-pattern correlation-based principal component analysis (SUBXPCA) and gradient-boosting decision tree (GBDT) methods.</span>

show abstract

“…However, its applicability is limited by the need to perform the computationally expensive singular value decomposition (SVD) multiple times [19]. In addition, there are many other mixing models, such as blind source separation, independent component analysis, sparse component analysis, and non-negative matrix factorisation (NMF) [31][32][33]. In recent years, NMF has been applied to many research areas.…”

Section: Introductionmentioning

confidence: 99%

Robust estimation of highly corrupted low‐rank matrix via alternating direction method of multiplier

Yimamu

Eisaka

2022

IET Signal Processing

View full text Add to dashboard Cite

Low‐rank matrices play a central role in modelling and computational methods for signal processing and large‐scale data analysis. Real‐world observed data are often sampled from low‐dimensional subspaces, but with sample‐specific corruptions (i.e. outliers) or random noises. In many applications where low‐rank matrices arise, these matrices cannot be fully sampled or directly observed, and one encounters the problem of recovering the matrix given only incomplete and indirect observations. The authors aim to recover a low‐rank component from incomplete and indirect observations and correct the possible errors. A new low‐rank matrix recovery formula based on generalised Tikhonov regularisation and its solution algorithm are proposed. The proposed method determines the low‐rank component for performing matrix recovery from highly corrupted observations. The authors’ recommended algorithm reduces not only the outliers but also random corruptions in the recovering task. The experimental results obtained using both synthetic and real application data demonstrate the efficacy of the proposed method.

show abstract

A new blind source separation approach based on dynamical similarity and its application on epileptic seizure prediction

Cited by 11 publications

References 41 publications

An improved blind Gaussian source separation approach based on generalized Jaccard similarity

An improved blind Gaussian source separation approach based on generalized Jaccard similarity

Prediction of the epileptic seizure through deep learning techniques using electroencephalography

Robust estimation of highly corrupted low‐rank matrix via alternating direction method of multiplier

Contact Info

Product

Resources

About