A Review on Machine Learning Algorithms in Handling EEG Artifacts (original) (raw)

Recent Artifacts Handling Algorithms in Electroencephalogram

International Journal of Advanced Science and Technology, 2020

The Electroencephalogram (EEG) is mostly used to measure electrical activity of brain. EEG tells us how the Brain reacts to various stimuli at that instant of time or in various conditions arrived at any instance of time. These are even used for Brain-Computer-Interface (BCI) activities, where the activities of brain are communicated to brain directly. While doing so the Artifacts produced in EEG data by various activities of a human being is a common problem and a research going on in this stream. Therefore, Artifact handling and removal at a very early stage is a prior research going on. This paper focuses on various artifacts involved and recent algorithms which give higher accuracy EEG signals. Mainly EEG artifacts handling and removals using various algorithms are presented based on their performances. We have studied various EEG Artifacts handling techniques by retrieving methods from past 12 years. These methods are categorized to handle various artifacts including EEG, EOG and EMG. Study found that instead of using one single algorithm; hybrid combination of it gives superior results as compared to single algorithm. Results found by using Hybrid model of handling artifacts that usage of individual machine learning algorithm had few limitations, whereas combination of algorithms could give better accuracy and sensitivity. Mostly Independent Component Analysis (ICA) and Support Vector Machine (SVM) where found to be used by various authors more times and these gave better performances as compared to other algorithms individually and combinational

Automated EEG artifact elimination by applying machine learning algorithms to ICA-based features

Journal of Neural Engineering

Objective. Biological and non-biological artifacts cause severe problems when dealing with electroencephalogram (EEG) recordings. Independent component analysis (ICA) is a widely used method for eliminating various artifacts from recordings. However, evaluating and classifying the calculated independent components (IC) as artifact or EEG is not fully automated at present. Approach. In this study, we propose a new approach for automated artifact elimination, which applies machine learning algorithms to ICA-based features. Main results. We compared the performance of our classifiers with the visual classification results given by experts. The best result with an accuracy rate of 95% was achieved using features obtained by range filtering of the topoplots and IC power spectra combined with an artificial neural network. Significance. Compared with the existing automated solutions, our proposed method is not limited to specific types of artifacts, electrode configurations, or number of EEG channels. The main advantages of the proposed method is that it provides an automatic, reliable, real-time capable, and practical tool, which avoids the need for the time-consuming manual selection of ICs during artifact removal.

Automatic reduction of artifacts in EEG-signals

2007

Electroencephalograms (EEG) are often contaminated with high amplitude artifacts limiting the usability of data. Methods that reduce these artifacts are often restricted to certain types of artifacts, require manual interaction or large training data sets. Within this paper we introduce a novel method, which is able to eliminate many different types of artifacts without manual intervention. The algorithm first decomposes the signal into different sub-band signals in order to isolate different types of artifacts into specific frequency bands. After signal decomposition with principal component analysis (PCA) an adaptive threshold is applied to eliminate components with high variance corresponding to the dominant artifact activity. Our results show that the algorithm is able to significantly reduce artifacts while preserving the EEG activity. Parameters for the algorithm do not have to be identified for every patient individually making the method a good candidate for preprocessing in automatic seizure detection and prediction algorithms.

EEG Artifacts Removal: A Detailed Review Based Study

International Journal For Multidisciplinary Research

Electroencephalography, or EEG for short, is a technique used to record the electrical activity of the brain. This EEG detects errors that affect how the human brain functions. This method is the most commonly used for recording the brain in laboratory research, clinical investigations, patient health monitoring, diagnostics, and a variety of other applications due to its non-invasiveness and cost-benefit ratio. Most EEG recordings are contaminated by a variety of irregularities, including those caused by electrode displacement, motion, ocular, and muscular activity related EMG anomalies. These unwanted artifacts may make it difficult to distinguish genuine information from them, in addition to confusing the brain's information processing that supports them. EEG signal artifacts can be removed in a variety of ways. The top and most popular artifact reduction techniques are listed on this page as PCA, pure EEG, and wavelet transform. The study provides a thorough evaluation of cu...

Study of Various Automatic EEG Artifact Removal Techniques

International Journal for Research in Applied Science and Engineering Technology, 2017

In recent researches, Electroencephalography (EEG) gains a widespread popularity. There is maximum probability of artifact with EEG signal because of physical and experimental problems therefore artifact elimination is a central issue during encephalogram recordings. Although many researchers have doing research in this area and developed their own method for artifact elimination like independent component analysis (ICA), average artifact subtraction (AAS), real time independent component analysis (ICA), Recursive Least Squares (RLS) adaptive filter, Spatially Constrained Independent Component Analysis(SCICA), Blind Source Separation and Wavelet Denoising, still visual examination by experts is needed. Finding the artifacts and eliminating them from real EEG signal by the use of competent algorithm assists researchers and doctors. This paper discusses the various methods along with limitations of automatic EEG artifact removal techniques.

A Machine Learning Approach for Artifact Removal from Brain Signal

Computer Systems Science and Engineering

Electroencephalography (EEG), helps to analyze the neuronal activity of a human brain in the form of electrical signals with high temporal resolution in the millisecond range. To extract clean clinical information from EEG signals, it is essential to remove unwanted artifacts that are due to different causes including at the time of acquisition. In this piece of work, the authors considered the EEG signal contaminated with Electrocardiogram (ECG) artifacts that occurs mostly in cardiac patients. The clean EEG is taken from the openly available Mendeley database whereas the ECG signal is collected from the Physionet database to create artifacts in the EEG signal and verify the proposed algorithm. Being the artifactual signal is non-linear and non-stationary the Random Vector Functional Link Network (RVFLN) model is used in this case. The Machine Learning approach has taken a leading role in every field of current research and RVFLN is one of them. For the proof of adaptive nature, the model is designed with EEG as a reference and artifactual EEG as input. The peaks of ECG signals are evaluated for artifact estimation as the amplitude is higher than the EEG signal. To vary the weight and reduce the error, an exponentially weighted Recursive Least Square (RLS) algorithm is used to design the adaptive filter with the novel RVFLN model. The random vectors are considered in this model with a radial basis function to satisfy the required signal experimentation. It is found that the result is excellent in terms of Mean Square Error (MSE), Normalized Mean Square Error (NMSE), Relative Error (RE), Gain in Signal to Artifact Ratio (GSAR), Signal Noise Ratio (SNR), Information Quantity (IQ), and Improvement in Normalized Power Spectrum (INPS). Also, the proposed method is compared with the earlier methods to show its efficacy.

A Hybrid Approach for Artifacts Removal from EEG Recordings

The electroencephalogram (EEG) is a widely used traditional procedure for diagnosing, monitoring and managing neurological disorders. Many artifact types that often contaminate EEG remain a key challenge for precise diagnosis of brain dysfunctions and neurological disorders. Hence, artifact removal is intuitively required for accurate EEG analysis and treatment. This paper presents a new extensive method that can remove a wide variety of EEG artifacts based mainly on Template Matching approach including multiple signal-processing tools. The method was evaluated and validated on real EEG data, giving promising results that offer better capabilities to neurophysiologists in routine EEG examinations and diagnosis.

Unsupervised EEG Artifact Detection and Correction

Frontiers in Digital Health, 2021

Electroencephalography (EEG) is used in the diagnosis, monitoring, and prognostication of many neurological ailments including seizure, coma, sleep disorders, brain injury, and behavioral abnormalities. One of the primary challenges of EEG data is its sensitivity to a breadth of non-stationary noises caused by physiological-, movement-, and equipment-related artifacts. Existing solutions to artifactdetectionare deficient because they require experts to manually explore and annotate data for artifact segments. Existing solutions to artifactcorrectionor removal are deficient because they assume that the incidence and specific characteristics of artifacts are similar across both subjects and tasks (i.e., “one-size-fits-all”). In this paper, we describe a novel EEG noise-reduction method that uses representation learning to perform patient- and task-specific artifact detection and correction. More specifically, our method extracts 58 clinically relevant features and applies an ensemble ...

Automated Artifact Removal From the Electroencephalogram: A Comparative Study

Clinical EEG and neuroscience : official journal of the EEG and Clinical Neuroscience Society (ENCS), 2013

Contamination of the electroencephalogram (EEG) by artifacts greatly reduces the quality of the recorded signals. There is an increased need for automated artifact removal methods. However such methods are rarely evaluated against one another via rigorous criteria, with results often presented based upon visual inspection alone.

EEG Artifact Removal Techniques: A Comparative Study

2020

An Electroencephalograph (EEG) is widely used to study the working of brain. The main challenge in dealing with such signals is its non-stationary nature and high dimensionality. The propagation of signals from one electrode to other gives rise to presence of artifacts in the data. This can lead to generation of false results while working on the data. Ocular, Muscular, and Cardiac are some of the very common artifacts for a dataset dedicated for motor imagery classification. This paper deals with the techniques and their comparative analysis for removal of ocular artifact from EEG signals. On the basis of dataset, the paper shows a study of how Linear Regression, Filtering, and Independent Component Analysis(ICA) works on the removal of ocular artifact in this data. To verify the effect of artifact removal technique features were extracted from the data, before and after applying different techniques and further fed to the classifier. An improvement in the accuracy of motor imagery...