SRI-EEG: State-Based Recurrent Imputation for EEG Artifact Correction

Abstract

Electroencephalogram (EEG) signals are often used as an input modality for Brain Computer Interfaces (BCIs). While EEG signals can be beneficial for numerous types of interaction scenarios in the real world, high levels of noise limits their usage to strictly noise-controlled environments such as a research laboratory. Even in a controlled environment, EEG is susceptible to noise, particularly from user motion, making it highly challenging to use EEG, and consequently BCI, as a ubiquitous user interaction modality. In this work, we address the EEG noise/artifact correction problem. Our goal is to detect physiological artifacts in EEG signal and automatically replace the detected artifacts with imputed values to enable robust EEG sensing overall requiring significantly reduced manual effort than is usual. We present a novel EEG state-based imputation model built upon a recurrent neural network, which we call SRI-EEG, and evaluate the proposed method on three publicly available EEG datasets. From quantitative and qualitative comparisons with six conventional and neural network based approaches, we demonstrate that our method achieves comparable performance to the state-of-the-art methods on the EEG artifact correction task.

Improved Manual Annotation of EEG Signals through Convolutional Neural Network Guidance

“…Automation is mainly achieved through channel referencing ( Schlögl et al, 2007 ), by applying various thresholding mechanisms ( Castellanos and Makarov, 2006 ; Gao et al, 2010 ; Nolan et al, 2010 ; Mognon et al, 2011 ; Akhtar et al, 2012 ; Islam and Tcheslavski, 2016 ; Jas et al, 2017 ), or using feature extraction followed by classification with conventional machine-learning algorithms such as support vector machines ( Shoker et al, 2005 ; Halder et al, 2007 ; Shao et al, 2009 ; Gabard-Durnam et al, 2018 ; Sai et al, 2018 ). In recent years, deep-learning algorithms have gained popularity to address EEG signal denoising ( Wang et al, 2018 ; B Yang et al, 2018 ; Craik et al, 2019 ; Pion-Tonachini et al, 2019 ; Roy et al, 2019 ; Sun et al, 2020 ; Boudaya et al, 2022 ; Jurczak et al, 2022 ; Liu et al, 2022 ), providing a more flexible solution than traditional methods by taking advantage of end-to-end learning, i.e., using a single model to act as both feature extractor and classifier. For example, because of hierarchical feature learning, convolutional neural networks (CNNs; LeCun et al, 1989 , 1998 , 2010 , 2015 ) can recognize complex patterns from minimally preprocessed data.…”

Robin’s Viewer: Using Deep-Learning Predictions to Assist EEG Annotation

“…Examples of machine learning techniques applied in artifact detection are support vector machines (SVMs) (Shoker et al 2005; Shao et al 2008; Bartels et al, 2010; Sai et al, 2017), k-nearest neighbor classifiers (k-NN) (Gao et al, 2010), and independent component analysis (ICA) (Barua and Begum, 2014; Radüntz et al, 2015). Recently, efforts have focused on developing deep-learning solutions for artifact detection (Pardede et al, 2015; Craik et al, 2019; Roy et al, 2019; Sun et al, 2020; Boudaya et al, 2022; Jurczak et al, 2022; Liu et al, 2022; Diachenko, M., Houtman, S. J., Juarez-Martinez, E. L., Ramautar, J. R., Weiler, R., Mansvelder, H. D., Bruining, H., Bloem, P., & Linkenkaer-Hansen K. (2022). Improved manual annotation of EEG signals through convolutional neural network guidance.…”

“…Given the need of neurologists to analyze large amounts of data, efforts have increased towards using various machine learning techniques for handling artifacts in EEG data (Shoker et al 2005; Shao et al 2008; Barua and Begum, 2014; Pardede et al, 2015; Radüntz et al, 2015; Yang et al, 2016, 2018; Sai et al, 2017; Wang et al, 2017; Craik et al, 2019; Roy et al, 2019; Sun et al, 2020; Boudaya et al, 2022; Jurczak et al, 2022; Liu et al, 2022; Diachenko, M., Houtman, S. J., Juarez-Martinez, E. L., Ramautar, J. R., Weiler, R., Mansvelder, H. D., Bruining, H., Bloem, P., & Linkenkaer-Hansen K. (2022). Improved manual annotation of EEG signals through convolutional neural network guidance.…”

“…Unpublished manuscript. ), and recurrent neural networks (RNNs) (Liu et al, 2022; Roy et al, 2019)). Deep-learning solutions, in particular, have increased in popularity for artifact handling (Craig et al, 2019; Roy et al, 2019) due to the minimal preprocessing they require and because of their ability to learn very complex functions between input data and the desired output classification (LeCun et al, 2015).…”

“…Machine-learning techniques trained to detect artifacts in EEG data can address the aforementioned issues, by providing an objective standard for artifact marking, and speeding up the marking process. The most common machine-leaning techniques used for artifact detection in EEG data, are based on support vector machines (SVMs) (Barua and Begum, 2014; Sai et al, 2017; Shao et al 2008), k-nearest neighbor classifiers (k-NN) (Barua and Begum, 2014; Roy, 2019), independent component analysis (ICA) (Barua and Begum, 2014; Radüntz et al, 2015; Sai et al, 2017), and, as of recently, various deep-learning models (e.g., autoencoders (Roy et al, 2019; Yang et al, 2016; 2018), convolutional neural networks (CNNs) (Diachenko et al, 2022; Jurczak et al, 2022; Roy et al, 2019; Sun et al, 2020), and recurrent neural networks (RNNs) (Liu et al, 2022; Roy et al, 2019)). Deep-learning solutions, in particular, have increased in popularity for artifact handling (Craig et al, 2019; Roy et al, 2019) due to the minimal preprocessing they require and because of their ability to learn very complex functions between input data and the desired output classification (LeCun et al, 2015).…”