EEG Artifact Removal by Bayesian Deep Learning &amp; ICA

Lee, Sangmin S.; Lee, Ki-Won; Kang, Guiyeom

doi:10.1109/embc44109.2020.9175785

Cited by 17 publications

(12 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The paper, however, did not compare the gained results with other existing methods. Lee et al [92] adopted BDL for artifact removal from EEG signals. ICA was used to extract independent components from EEG signals and other information that was not part of independent components were treated as artifacts and removed.…”

Section: B Medical Signal Processingmentioning

confidence: 99%

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

2022

View full text Add to dashboard Cite

In the last decade, Deep Learning (DL) has revolutionized the use of artificial intelligence, and it has been deployed in different fields of healthcare applications such as image processing, natural language processing, and signal processing. DL models have also been intensely used in different tasks of healthcare such as disease diagnostics and treatments. Deep learning techniques have surpassed other machine learning algorithms and proved to be the ultimate tools for many state-of-the-art applications. Despite all that success, classical deep learning has limitations and their models tend to be very confident about their predicted decisions because it does not know when it makes mistake. For the healthcare field, this limitation can have a negative impact on models predictions since almost all decisions regarding patients and diseases are sensitive. Therefore, Bayesian deep learning (BDL) has been developed to overcome these limitations. Unlike classical DL, BDL uses probability distributions for the model parameters, which makes it possible to estimate the whole uncertainties associated with the predicted outputs. In this regard, BDL offers a rigorous framework to quantify all sources of uncertainties in the model. This study reviews popular techniques of using Bayesian deep learning with their benefits and limitations. It also reviewed recent deep learning architecture such as Convolutional Neural Networks and Recurrent Neural Networks. In particular, the applications of Bayesian deep learning in healthcare have been discussed such as its use in medical imaging tasks, clinical signal processing, medical natural language processing, and electronic health records. Furthermore, this paper has covered the deployment of Bayesian deep learning for some of the widespread diseases. This paper has also discussed the fundamental research challenges and highlighted some research gaps in both the Bayesian deep learning and healthcare perspective.

show abstract

Section: B Medical Signal Processingmentioning

confidence: 99%

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

2022

View full text Add to dashboard Cite

show abstract

“…In [18], the author established a deep learning method using Bayesian and attention modules to improve the performance of the classifier. Here, after the filtering process, to remove line noise, the artifact subspace reform (ASR) technique [19] was revised to remove an artifact that is dispersed throughout the entire scalp with a huge variance.…”

Section: Literature Reviewmentioning

confidence: 99%

Detection Of Event-Related Potential Artifacts Of Oddball Paradigm By Unsupervised Machine Learning Algorithm

Akhter¹,

Beyette²

2022

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Electroencephalography (EEG) is one of the most common and benign methods for analyzing and identifying abnormalities in the human brain. EEG is an incessant measure of the activities of the human brain. In contrast, when the measurement of EEG is bounded by time and the EEG is synchronized to an exterior stimulus, is known as Event-Related Potential (ERP). ERP has the capability to perceive and explore the human brain's responses to specific sensitive, cognitive, or motor events in real time with high temporal resolution. Among the various techniques, the oddball paradigm is very famous in EEG studies. In an oddball paradigm experiment, brain responses to frequent and infrequent stimuli are measured. However, the success of ERP research is very much dependent on the analysis of clean data sets and unfortunately, EEG is a combination of both neural and nonneural activities which introduce significant sources of noise that are not related to the brain's response to the external stimulus. These unrelated non-EEG components are acknowledged as artifacts and due to these, the quality of the EEG may damage by decreasing SNR (signal-to-noise ratio). In addition, these artifacts may mislead the actual information in the study. Addressing this problem, the purpose of this research is to introduce a machine learning algorithm (ML) that can screen EEG/ERP data to remove data epochs that are disrupted by artifacts and thus produce a clean data set. Overall, three unsupervised ML algorithms are applied to identify noisy epochs and it is found that the DBScan method performs best with 93.43% accuracy. Finally, the success of this study will allow the ERP study to have a cleaner ERP data set in normal laboratory conditions with less complexity in the ERP studies.

show abstract

“…One group consists of linear regression [11]- [13], adaptive filters [14]- [16], Wiener filters [17], and Kalman filters [18], [19]. A second group of methods relies on EEG decomposition such as wavelet transform [20]- [22], empirical mode decomposition [23]- [25], and blind source separation [26]- [42]. Recently deep neural networks (DNNs) were also suggested for artifact removal [43]- [45].…”

Section: Introductionmentioning

confidence: 99%

“…IC classifiers have been created using temporal, spectral and spatial handcrafted features extracted from the time-series, PSD and topographic map of each IC, respectively [27], [29]- [34], [36], [39]. Recently, new approaches based on DNNs were formulated [40]- [42]. Croce et al [40] developed convolutional neural networks (CNNs) using the PSDs and topographic maps.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ensemble Deep Neural Network for Automatic Classification of EEG Independent Components

Lopes

Leal

Medeiros

et al. 2022

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Objective: Independent component analysis (ICA) is commonly used to remove noisy artifacts from multi-channel scalp electroencephalogram (EEG) signals. ICA decomposes EEG into different independent components (ICs) and then, experts remove the noisy ones. This process is highly time-consuming and experts are not always available. To surpass this drawback, research is going on to develop models to automatically conduct IC classification. Current state-of-the-art models use power spectrum densities (PSDs) and topoplots to classify ICs. The performance of these methods may be limited by disregarding the IC time-series that would contribute to fully simulate the visual inspection performed by experts. Methods: We present a novel ensemble deep neural network that combines time-series, PSDs, and topoplots to classify ICs. Moreover, we study the ability to use our model in transfer learning approaches. Results: Experimental results showed that using time-series improves IC classification. Results also indicated that transfer learning obtained higher performance than simply training a new model from scratch. Conclusion: Researchers should develop IC classifiers using the three sources of information. Moreover, transfer learning approaches should be considered when producing new deep learning models. Significance: This work improves IC classification, enhancing the automatic removal of EEG artifacts. Additionally, since labelled ICs are generally not publicly available, the possibility of using our model in transfer learning studies may motivate other researchers to develop their own classifiers.

show abstract

EEG Artifact Removal by Bayesian Deep Learning & ICA

Cited by 17 publications

References 12 publications

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

Detection Of Event-Related Potential Artifacts Of Oddball Paradigm By Unsupervised Machine Learning Algorithm

Ensemble Deep Neural Network for Automatic Classification of EEG Independent Components

Contact Info

Product

Resources

About