China
Africa
Explore chapters and articles related to this topic
Machine Learning Methods for Electroencephalogram (EEG) Big Data in the Context of MIOT Smart Systems
Published in Nishu Gupta, Srinivas Kiran Gottapu, Rakesh Nayak, Anil Kumar Gupta, Mohammad Derawi, Jayden Khakurel, Human-Machine Interaction and IoT Applications for a Smarter World, 2023
Aileni Raluca Maria, Pasca Sever
The EEG represents the recording of signals from neurons [1–3] of the cerebral cortex [4], a set of fluctuating field potentials, produced by the simultaneous activity of a large number of neurons and captured by electrodes located on the scalp. EEG is used in the diagnosis of epilepsy, encephalopathy, in monitoring brain activity during anesthesia, in patients with coma, and in determining brain death. On the entire surface of the skin of the head are arranged 10–20 metal electrodes connected by wires to the recording device. It measures the electrical potential detected by each electrode and compares the electrodes two by two, each comparison translating through a path called a bypass. Electroencephalographic reactivity is evaluated using simple tests such as eye-opening, hyperpnea (slow and full breathing), and intermittent light stimulation obtained with short and intense light discharges whose frequency is gradually increased. The EEG assessment takes approximately 20 minutes and does not require hospitalization.
A Review on Biomedical Signals with Fundamentals of Digital Signal Processing
Published in Mitul Kumar Ahirwal, Narendra D. Londhe, Anil Kumar, Artificial Intelligence Applications for Health Care, 2022
Mangesh Ramaji Kose, Mitul Kumar Ahirwal, Mithilesh Atulkar
Human brain is a complex and an important organ. It consists of millions of interconnected neurons sharing information to generate cognitive activity. The working of human brain is not yet completely explored. EEG signal presents non-invasive, cheaper, and easier neuroimaging techniques used to analyze brain functionality. EEG signal presents electrical functionality of human brain, which gets recorded using electrodes. There are many other techniques available for analysis of brain functionality, such as Functional Magnetic Resonance Imaging (FMRI) and Computer Tomography (CT), but they are costlier than EEG. EEG signals are non-stationary and non-linear by nature, because their generation is complex [15–17]. The electrical potential is generated due to transfer of collective or combined neural information between the groups of neurons. EEG electrodes are used to record these potential by placing them on the scalp [17]. The visualization of multichannel EEG is presented in Figure 2.8. The EEG signal corresponding to channel number 1, 2, 3, and 32 from total 32-channel EEG data is plotted.
Application of Machine-Learning Techniques in Electroencephalography Signals
Published in Mridu Sahu, G. R. Sinha, Brain and Behavior Computing, 2021
Arun Sasidharan, Kusumika Krori Dutta
The EEG can give millisecond resolution information of brain activity, which many other neuroimaging tools (like functional magnetic resonance imaging) fail to capture. As a noninvasive, affordable medical device, EEG has been in clinical use for the last several years. From a clinical context, the main use of EEG is to diagnose brain activity abnormalities, mainly with respect to epileptic seizures and sleep disorders. Other clinical uses of EEG are in the diagnosis of coma, brain death, encephalopathy, etc. From a research context, the analysis of EEG signals has found innumerable applications such as understanding the mechanism behind several mental disorders, emotional processing/regulation, substance abuse, altered consciousness (like coma), mental training (like mediation and music), the effect of cognitive training, lie detection, neuromarketing, so on and so forth. Thus, EEG patterns capture many aspects of mental processes like cognition, behavior, and emotions, across the illness to wellness spectrum.
Driver drowsiness detection methods using EEG signals: a systematic review
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2023
Raed Mohammed Hussein, Firas Sabar Miften, Loay E. George
Electroencephalography (EEG) is a neuroimaging method that measures the brain's electrical activity. It enables vital medical diagnostic and brain research investigations. Despite its sensitivity to noise, EEG is the most effective method for recording and analyzing brain activity because it is non-invasive, portable, cost-effective, relatively simple to use, and has an exceptional temporal resolution of less than one millisecond (Gevins et al. 1999). EEG utilizes electrodes placed on the scalp to measure the brain's electrical activity. The recorded signal waves contain valuable information about the brain's health. EEG records electric potential differences of tens of microvolts (μV) reaching the scalp when pyramidal neurons generate tiny excitatory postsynaptic potentials in the brain's cortical layers. Numerous electrode positioning systems are typically utilized for EEG signal recording. The EEG signal processing and analysis consist of four steps:Preprocessing the raw signals with filtering or other techniquesExtracting the essential information in the form of featuresApplying feature selection methods for more optimized resultsAnalyzing the results
Correlation and Relief Attribute Rank-based Feature Selection Methods for Detection of Alcoholic Disorder Using Electroencephalogram Signals
Published in IETE Journal of Research, 2022
Nandini Kumari, Shamama Anwar, Vandana Bhattacharjee
Electroencephalogram (EEG) is a physiological test that acquires brain signals to model behavioural, characteristics, diagnostic, prognostic and therapeutic analysis of neuropsychological conditions. It is a methodology that records the brain’s electrical activity by positioning electrodes in a standard order using an EEG headset. The electroencephalogram is a reliable rumination of the many physiological factors harmonizing the brain. Using EEG has divergent merits as compared to other imaging techniques or pure behavioural observations. The most paramount merit of using EEG is its excellent time resolution. An EEG headset can capture hundreds to thousands of snapshots of electrical activity, within a single second, across the different sensors. Howbeit, a concern with EEG signals is that they are immensely dynamic in most cases. This affects the accurate measurements of the survey using EEG signals as they vary between different cases. Consequently, handling the intrinsic features of EEG signals is continued a challenging research problem in clinical diagnosis. Since the EEG signals are dynamic, multiple analysis approaches have evolved for various EEG based applications such as epileptic diagnosis [1], emotion recognition [2], sleep stage identification [3], diagnosis of diseases like Alzheimer [4], autistic spectrum disorder [5] and study of attention deficit hyperactivity disorder [6], depression [7] and detection of alcoholism [8].
Investigation of an EEG-based Indicator of Skill Acquisition as Novice Participants Practice a Lifeboat Maneuvering Task in a Simulator
Published in International Journal of Human–Computer Interaction, 2020
Rifat Biswas, Brian Veitch, Sarah D. Power
One challenge of EEG data is its low signal-to-noise. The EEG signal can be contaminated with other electrophysiological artifacts including EOG (electrooculography, from eye movements) and EMG (electromyography, from muscle activity), as well as electrical noise from the environment, or simply movement of the electrodes. Small segments of EEG data that were contaminated by EMG or other movement-related noise were manually removed. High and low frequency components were removed using a digital Chebyshev Type 2 bandpass filter (1–40 Hz). Blinking and saccade artifacts were removed using Independent Component Analysis (ICA) (Delorme & Makeig, 2004; Di Flumeri, Aricó, Borghini, Colosimo, & Babiloni, 2016). The data were then baseline normalized using the Z-transform. Each lifeboat trial was normalized using the eyes-open baseline recorded immediately before it. This was done to account for any task-unrelated changes in brain activity that could potentially occur over the course of the session, for example due to fatigue.