1
|
Pfeffer MA, Ling SSH, Wong JKW. Exploring the frontier: Transformer-based models in EEG signal analysis for brain-computer interfaces. Comput Biol Med 2024; 178:108705. [PMID: 38865781 DOI: 10.1016/j.compbiomed.2024.108705] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/11/2023] [Revised: 05/01/2024] [Accepted: 06/01/2024] [Indexed: 06/14/2024]
Abstract
This review systematically explores the application of transformer-based models in EEG signal processing and brain-computer interface (BCI) development, with a distinct focus on ensuring methodological rigour and adhering to empirical validations within the existing literature. By examining various transformer architectures, such as the Temporal Spatial Transformer Network (TSTN) and EEG Conformer, this review delineates their capabilities in mitigating challenges intrinsic to EEG data, such as noise and artifacts, and their subsequent implications on decoding and classification accuracies across disparate mental tasks. The analytical scope extends to a meticulous examination of attention mechanisms within transformer models, delineating their role in illuminating critical temporal and spatial EEG features and facilitating interpretability in model decision-making processes. The discourse additionally encapsulates emerging works that substantiate the efficacy of transformer models in noise reduction of EEG signals and diversifying applications beyond the conventional motor imagery paradigm. Furthermore, this review elucidates evident gaps and propounds exploratory avenues in the applications of pre-trained transformers in EEG analysis and the potential expansion into real-time and multi-task BCI applications. Collectively, this review distils extant knowledge, navigates through the empirical findings, and puts forward a structured synthesis, thereby serving as a conduit for informed future research endeavours in transformer-enhanced, EEG-based BCI systems.
Collapse
Affiliation(s)
- Maximilian Achim Pfeffer
- Faculty of Engineering and Information Technology, University of Technology Sydney, CB11 81-113, Broadway, Ultimo, 2007, New South Wales, Australia.
| | - Steve Sai Ho Ling
- Faculty of Engineering and Information Technology, University of Technology Sydney, CB11 81-113, Broadway, Ultimo, 2007, New South Wales, Australia.
| | - Johnny Kwok Wai Wong
- Faculty of Design, Architecture and Building, University of Technology Sydney, 15 Broadway, Ultimo, 2007, New South Wales, Australia.
| |
Collapse
|
2
|
Bai Z, Zhang S. Effects of different natural soundscapes on human psychophysiology in national forest park. Sci Rep 2024; 14:17462. [PMID: 39075109 PMCID: PMC11286786 DOI: 10.1038/s41598-024-67812-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/22/2023] [Accepted: 07/16/2024] [Indexed: 07/31/2024] Open
Abstract
Most of the current soundscape research content is limited to the discussion of the restoration effect of single-element soundscapes, but it is the combination of sounds that is common in outdoor activities, and there is no evidence that the restoration of natural soundscapes is better with multi-element combinations. In this study, the Zhangjiajie National Forest Park in China was used as the research object, and the physiological indices of the subjects were collected through electroencephalogram signals, and the POMS short-form psychological scale was used to understand the subjective psychological responses of the subjects to the soundscape. The results showed that (1) The psychophysiological restorative ability of the natural soundscape of the National Forest Park was confirmed, and the subjects' psychological and physiological indices changed significantly and positively after listening to each section of the natural soundscape (p = 0.001). (2) The restorative effect of the multi-natural sound combination was ranked first in the overall ranking of the five natural soundscapes, and the multi-natural sound combination did indeed provide better restorative effects than the single-element sounds. (3) Gender does not usually have a significant effect on the restoration effect, and only Windy Sound among the four single-element nature sound landscapes and one multi-element combination of nature sound landscapes showed a significant gender difference, so in general, the effect of gender on the restoration effect of nature sound landscapes is not significant. In terms of research methodology, this study used cluster analysis to cluster the five types of natural soundscapes according to psychological and physiological recovery ability, and used ridge regression to construct mathematical models of the psychological and physiological recovery of each of the four natural soundscapes. The study of human physiological and psychological recovery from different types of natural soundscapes in China's national forest parks will provide a basis for soundscape planning, design, and policy formulation in national forest parks.
Collapse
Affiliation(s)
- Zhengkang Bai
- National Park and Tourism College, Central South University of Forestry and Technology, Changsha, 410004, China
| | - Shuangquan Zhang
- National Park and Tourism College, Central South University of Forestry and Technology, Changsha, 410004, China.
| |
Collapse
|
3
|
Tigga NP, Garg S. Efficacy of novel attention-based gated recurrent units transformer for depression detection using electroencephalogram signals. Health Inf Sci Syst 2023; 11:1. [PMID: 36590874 PMCID: PMC9800680 DOI: 10.1007/s13755-022-00205-8] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/20/2022] [Accepted: 12/13/2022] [Indexed: 12/30/2022] Open
Abstract
Purpose Depression is a global challenge causing psychological and intellectual problems that require efficient diagnosis. Electroencephalogram (EEG) signals represent the functional state of the human brain and can help build an accurate and viable technique for the early prediction and treatment of depression. Methods An attention-based gated recurrent units transformer (AttGRUT) time-series model is proposed to efficiently identify EEG perturbations in depressive patients. Statistical, spectral and wavelet features were first extracted from the 60-channel EEG signal data. Then, two feature selection techniques, recursive feature elimination and the Boruta algorithm, both with Shapley additive explanations, were utilised for selecting essential features. Results The proposed model outperformed the two baseline and two hybrid time-series models-long short-term memory (LSTM), gated recurrent units (GRU), convolutional neural network-LSTM (CNN-LSTM), and CNN-GRU-achieving an accuracy of up to 98.67%. Feature selection considerably increased the performance across all time-series models. Conclusion Based on the obtained results, novel feature selection greatly affected the results of the baseline and hybrid time-series models. The proposed AttGRUT can be implemented and tested in other domains by using different modalities for prediction. Supplementary Information The online version contains supplementary material available at 10.1007/s13755-022-00205-8.
Collapse
Affiliation(s)
| | - Shruti Garg
- Birla Institute of Technology, Mesra, Ranchi, India
| |
Collapse
|
4
|
Mumenin KM, Biswas P, Khan MAM, Alammary AS, Nahid AA. A Modified Aquila-Based Optimized XGBoost Framework for Detecting Probable Seizure Status in Neonates. SENSORS (BASEL, SWITZERLAND) 2023; 23:7037. [PMID: 37631573 PMCID: PMC10458382 DOI: 10.3390/s23167037] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/19/2023] [Revised: 07/29/2023] [Accepted: 08/01/2023] [Indexed: 08/27/2023]
Abstract
Electroencephalography (EEG) is increasingly being used in pediatric neurology and provides opportunities to diagnose various brain illnesses more accurately and precisely. It is thought to be one of the most effective tools for identifying newborn seizures, especially in Neonatal Intensive Care Units (NICUs). However, EEG interpretation is time-consuming and requires specialists with extensive training. It can be challenging and time-consuming to distinguish between seizures since they might have a wide range of clinical characteristics and etiologies. Technological advancements such as the Machine Learning (ML) approach for the rapid and automated diagnosis of newborn seizures have increased in recent years. This work proposes a novel optimized ML framework to eradicate the constraints of conventional seizure detection techniques. Moreover, we modified a novel meta-heuristic optimization algorithm (MHOA), named Aquila Optimization (AO), to develop an optimized model to make our proposed framework more efficient and robust. To conduct a comparison-based study, we also examined the performance of our optimized model with that of other classifiers, including the Decision Tree (DT), Random Forest (RF), and Gradient Boosting Classifier (GBC). This framework was validated on a public dataset of Helsinki University Hospital, where EEG signals were collected from 79 neonates. Our proposed model acquired encouraging results showing a 93.38% Accuracy Score, 93.9% Area Under the Curve (AUC), 92.72% F1 score, 65.17% Kappa, 93.38% sensitivity, and 77.52% specificity. Thus, it outperforms most of the present shallow ML architectures by showing improvements in accuracy and AUC scores. We believe that these results indicate a major advance in the detection of newborn seizures, which will benefit the medical community by increasing the reliability of the detection process.
Collapse
Affiliation(s)
- Khondoker Mirazul Mumenin
- Electronics and Communication Engineering (ECE) Discipline, Khulna University (KU), Khulna 9208, Bangladesh; (K.M.M.); (P.B.)
| | - Prapti Biswas
- Electronics and Communication Engineering (ECE) Discipline, Khulna University (KU), Khulna 9208, Bangladesh; (K.M.M.); (P.B.)
| | - Md. Al-Masrur Khan
- Department of ICT Integrated Ocean Smart Cities Engineering, Dong-A University, Busan 49315, Republic of Korea;
| | - Ali Saleh Alammary
- College of Computing and Informatics, Saudi Electronic University, Riyadh 11673, Saudi Arabia
| | - Abdullah-Al Nahid
- Electronics and Communication Engineering (ECE) Discipline, Khulna University (KU), Khulna 9208, Bangladesh; (K.M.M.); (P.B.)
| |
Collapse
|
5
|
Classification of EEG Signals Using Neural Network for Predicting Consumer Choices. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2022; 2022:5872401. [PMID: 35909868 PMCID: PMC9328993 DOI: 10.1155/2022/5872401] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/19/2022] [Revised: 02/10/2022] [Accepted: 03/23/2022] [Indexed: 11/18/2022]
Abstract
EEG, or Electroencephalogram, is an instrument that examines the brain’s functions while it is executing any activity. EEG signals to aid in the identification of brain processes and movements and are thus useful in the detection of neurobiological illnesses. Pulses have a very weak magnitude and are recorded from peak to peak, with pulse width ranging from 0.5 to 100 V, which is around 100 times below than ECG signals. As a result, many types of noise can easily influence them. Because EEG signals are so important in detecting brain illnesses, it is critical to preprocess them for accurate assessment and detection. The crown of your head The EEG is a weighted combination of the signals generated by the different small locations beneath the electrodes on the cortical plate. The rhythm of electrical impulses is useful for evaluating a broad range of brain diseases. Hypertension, Alzheimer, and brain damage are all possibilities. We can compare and distinguish the brainwaves for different emotions and illnesses linked with the brain by studying the EEG signal. Multiple research studies and methodologies for preprocessing, extraction of features, and evaluation of EEG data have recently been created. The use of EEG in human-computer communication could be a novel and demanding field that has acquired traction in recent years. We present predictive modeling for analyzing the customer’s preference of likes and dislikes via EEG signal in our report. The impulses were obtained when clients used the Internet to seek for multiple items. The studies were carried out on a dataset that included a variety of consumer goods.
Collapse
|
6
|
A Cumulants-Based Human Brain Decoding. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2022; 2022:6474515. [PMID: 35860640 PMCID: PMC9293498 DOI: 10.1155/2022/6474515] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 04/25/2022] [Revised: 06/12/2022] [Accepted: 06/23/2022] [Indexed: 11/17/2022]
Abstract
Human cognition is influenced by the way the nervous system processes information and is linked to this mechanical explanation of the human body’s cognitive function. Accuracy is the key emphasis in neuroscience which may be enhanced by utilising new hardware, mathematical, statistical, and computational methodologies. Feature extraction and feature selection also play a crucial function in gaining improved accuracy since the proper characteristics can identify brain states efficiently. However, both feature extraction and selection procedures are dependent on mathematical and statistical techniques which implies that mathematical and statistical techniques have a direct or indirect influence on prediction accuracy. The forthcoming challenges of the brain-computer interface necessitate a thorough critical understanding of the complicated structure and uncertain behavior of the brain. It is impossible to upgrade hardware periodically, and thus, an option is necessary to collect maximum information from the brain against varied actions. The mathematical and statistical combination could be the ideal answer for neuroscientists which can be utilised for feature extraction, feature selection, and classification. That is why in this research a statistical technique is offered together with specialised feature extraction and selection methods to increase the accuracy. A score fusion function is changed utilising an enhanced cumulants-driven likelihood ratio test employing multivariate pattern analysis. Functional MRI data were acquired from 12 patients versus a visual test that comprises of pictures from five distinct categories. After cleaning the data, feature extraction and selection were done using mathematical approaches, and lastly, the best match of the projected class was established using the likelihood ratio test. To validate the suggested approach, it is compared with the current methods reported in recent research.
Collapse
|
7
|
Liang H, Maedono S, Yu Y, Liu C, Ueda N, Li P, Zhu C. Exploring Neurofeedback Training for BMI Power Augmentation of Upper Limbs: A Pilot Study. ENTROPY (BASEL, SWITZERLAND) 2021; 23:443. [PMID: 33918833 PMCID: PMC8068929 DOI: 10.3390/e23040443] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 02/25/2021] [Revised: 03/25/2021] [Accepted: 04/06/2021] [Indexed: 11/16/2022]
Abstract
Electroencephalography neurofeedback (EEG-NFB) training can induce changes in the power of targeted EEG bands. The objective of this study is to enhance and evaluate the specific changes of EEG power spectral density that the brain-machine interface (BMI) users can reliably generate for power augmentation through EEG-NFB training. First, we constructed an EEG-NFB training system for power augmentation. Then, three subjects were assigned to three NFB training stages, based on a 6-day consecutive training session as one stage. The subjects received real-time feedback from their EEG signals by a robotic arm while conducting flexion and extension movement with their elbow and shoulder joints, respectively. EEG signals were compared with each NFB training stage. The training results showed that EEG beta (12-40 Hz) power increased after the NFB training for both the elbow and the shoulder joints' movements. EEG beta power showed sustained improvements during the 3-stage training, which revealed that even the short-term training could improve EEG signals significantly. Moreover, the training effect of the shoulder joints was more obvious than that of the elbow joints. These results suggest that NFB training can improve EEG signals and clarify the specific EEG changes during the movement. Our results may even provide insights into how the neural effects of NFB can be better applied to the BMI power augmentation system and improve the performance of healthy individuals.
Collapse
Affiliation(s)
- Hongbo Liang
- Maebashi Institute of Technology, Center for Regional Collaboration, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan
| | - Shota Maedono
- Department of Systems Life Engineering, Graduate School of Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan; (S.M.); (Y.Y.)
| | - Yingxin Yu
- Department of Systems Life Engineering, Graduate School of Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan; (S.M.); (Y.Y.)
| | - Chang Liu
- Department of Environment and Life Engineering, Graduate School of Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan; (C.L.); (N.U.); (P.L.)
| | - Naoya Ueda
- Department of Environment and Life Engineering, Graduate School of Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan; (C.L.); (N.U.); (P.L.)
| | - Peirang Li
- Department of Environment and Life Engineering, Graduate School of Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan; (C.L.); (N.U.); (P.L.)
| | - Chi Zhu
- Department of Systems Life Engineering, Maebashi Institute of Technology, 460-1 Kamisadori, Maebashi, Gunma 371-0816, Japan
| |
Collapse
|
8
|
A Concealed Information Test System Based on Functional Brain Connectivity and Signal Entropy of Audio–Visual ERP. IEEE Trans Cogn Dev Syst 2020. [DOI: 10.1109/tcds.2020.2991359] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
|
9
|
A novel approach based on wavelet analysis and arithmetic coding for automated detection and diagnosis of epileptic seizure in EEG signals using machine learning techniques. Biomed Signal Process Control 2020. [DOI: 10.1016/j.bspc.2019.101707] [Citation(s) in RCA: 38] [Impact Index Per Article: 9.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/24/2022]
|
10
|
Chang W, Wang H, Hua C, Wang Q, Yuan Y. Comparison of different functional connectives based on EEG during concealed information test. Biomed Signal Process Control 2019. [DOI: 10.1016/j.bspc.2018.12.008] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/27/2022]
|
11
|
Vega-Gualán E, Vargas A, Becerra M, Umaquinga A, Riascos JA, Peluffo D. Exploring the Characterization and Classification of EEG Signals for a Computer-Aided Epilepsy Diagnosis System. Brain Inform 2019. [DOI: 10.1007/978-3-030-37078-7_19] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/25/2022] Open
|
12
|
Akbarian B, Erfanian A. Automatic Seizure Detection Based on Nonlinear Dynamical Analysis of EEG Signals and Mutual Information. Basic Clin Neurosci 2018; 9:227-240. [PMID: 30519381 PMCID: PMC6276534 DOI: 10.32598/bcn.9.4.227] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/03/2017] [Revised: 04/10/2017] [Accepted: 10/04/2017] [Indexed: 11/20/2022] Open
Abstract
Introduction: In this paper, nonlinear dynamical analysis based on Recurrence Quantification Analysis (RQA) is employed to characterize the nonlinear EEG dynamics. RQA can provide useful quantitative information on the regular, chaotic, or stochastic property of the underlying dynamics. Methods: We use the RQA-based measures as the quantitative features of the nonlinear EEG dynamics. Mutual Information (MI) was used to find the most relevant feature subset out of RQA-based features. The selected features were fed into an artificial neural network for grouping of EEG recordings to detect ictal, interictal, and healthy states. The performance of the proposed procedure was evaluated using a database for different classification cases. Results: The combination of five selected features based on MI achieved 100% accuracy, which demonstrates the superiority of the proposed method. Conclusion: The results showed that the nonlinear dynamical analysis based on Rcurrence Quantification Analysis (RQA) can be employed as a suitable approach for characterizing the nonlinear EEG dynamics and detecting the seizure.
Collapse
Affiliation(s)
- Behnaz Akbarian
- Iran Neural Technology Research Centre, Iran University of Science and Technology, Tehran, Iran
| | - Abbas Erfanian
- Department of Bioelectrical Engineering, School of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran
| |
Collapse
|
13
|
Brain–computer interface connected to telemedicine and telecommunication in virtual reality applications. TELEMATICS AND INFORMATICS 2017. [DOI: 10.1016/j.tele.2016.01.003] [Citation(s) in RCA: 17] [Impact Index Per Article: 2.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022]
|
14
|
Pereira LAM, Papa JP, Coelho ALV, Lima CAM, Pereira DR, de Albuquerque VHC. Automatic identification of epileptic EEG signals through binary magnetic optimization algorithms. Neural Comput Appl 2017. [DOI: 10.1007/s00521-017-3124-3] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/19/2022]
|
15
|
Biogeography based hybrid scheme for automatic detection of epileptic seizures from EEG signatures. Appl Soft Comput 2017. [DOI: 10.1016/j.asoc.2016.12.009] [Citation(s) in RCA: 15] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
|
16
|
Mihandoost S, Chehel Amirani M. Cyclic spectral analysis of electrocardiogram signals based on GARCH model. Biomed Signal Process Control 2017. [DOI: 10.1016/j.bspc.2016.07.012] [Citation(s) in RCA: 10] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
|
17
|
Multi-category EEG signal classification developing time-frequency texture features based Fisher Vector encoding method. Neurocomputing 2016. [DOI: 10.1016/j.neucom.2016.08.050] [Citation(s) in RCA: 38] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
18
|
Hsu WY. Assembling A Multi-Feature EEG Classifier for Left–Right Motor Imagery Data Using Wavelet-Based Fuzzy Approximate Entropy for Improved Accuracy. Int J Neural Syst 2015; 25:1550037. [DOI: 10.1142/s0129065715500379] [Citation(s) in RCA: 55] [Impact Index Per Article: 6.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
Abstract
An EEG classifier is proposed for application in the analysis of motor imagery (MI) EEG data from a brain–computer interface (BCI) competition in this study. Applying subject-action-related brainwave data acquired from the sensorimotor cortices, the system primarily consists of artifact and background removal, feature extraction, feature selection and classification. In addition to background noise, the electrooculographic (EOG) artifacts are also automatically removed to further improve the analysis of EEG signals. Several potential features, including amplitude modulation, spectral power and asymmetry ratio, adaptive autoregressive model, and wavelet fuzzy approximate entropy (wfApEn) that can measure and quantify the complexity or irregularity of EEG signals, are then extracted for subsequent classification. Finally, the significant sub-features are selected from feature combination by quantum-behaved particle swarm optimization and then classified by support vector machine (SVM). Compared with feature extraction without wfApEn on MI data from two data sets for nine subjects, the results indicate that the proposed system including wfApEn obtains better performance in average classification accuracy of 88.2% and average number of commands per minute of 12.1, which is promising in the BCI work applications.
Collapse
Affiliation(s)
- Wei-Yen Hsu
- Department of Information Management, National Chung Cheng University, No. 168, Sec. 1, University Rd., Min-Hsiung Township, Chiayi County 621, Taiwan
- Advanced Institute of Manufacturing with High-tech Innovations, National Chung Cheng University, No. 168, Sec. 1, University Rd. Min-Hsiung Township, Chiayi County 621, Taiwan
| |
Collapse
|
19
|
EEG signal classification for epilepsy diagnosis via optimum path forest – A systematic assessment. Neurocomputing 2014. [DOI: 10.1016/j.neucom.2014.01.020] [Citation(s) in RCA: 77] [Impact Index Per Article: 7.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
20
|
Genetic algorithms tuned expert model for detection of epileptic seizures from EEG signatures. Appl Soft Comput 2014. [DOI: 10.1016/j.asoc.2014.01.029] [Citation(s) in RCA: 43] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
21
|
Gu JN, Lu HT, Lu BL. An integrated Gaussian mixture model to estimate vigilance level based on EEG recordings. Neurocomputing 2014. [DOI: 10.1016/j.neucom.2012.10.042] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/26/2022]
|
22
|
Methods of EEG signal features extraction using linear analysis in frequency and time-frequency domains. ISRN NEUROSCIENCE 2014; 2014:730218. [PMID: 24967316 PMCID: PMC4045570 DOI: 10.1155/2014/730218] [Citation(s) in RCA: 108] [Impact Index Per Article: 10.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 10/20/2013] [Accepted: 01/09/2014] [Indexed: 11/28/2022]
Abstract
Technically, a feature represents a distinguishing property, a recognizable measurement, and a functional component obtained from a section of a pattern. Extracted features are meant to minimize the loss of important information embedded in the signal. In addition, they also simplify the amount of resources needed to describe a huge set of data accurately. This is necessary to minimize the complexity of implementation, to reduce the cost of information processing, and to cancel the potential need to compress the information. More recently, a variety of methods have been widely used to extract the features from EEG signals, among these methods are time frequency distributions (TFD), fast fourier transform (FFT), eigenvector methods (EM), wavelet transform (WT), and auto regressive method (ARM), and so on. In general, the analysis of EEG signal has been the subject of several studies, because of its ability to yield an objective mode of recording brain stimulation which is widely used in brain-computer interface researches with application in medical diagnosis and rehabilitation engineering. The purposes of this paper, therefore, shall be discussing some conventional methods of EEG feature extraction methods, comparing their performances for specific task, and finally, recommending the most suitable method for feature extraction based on performance.
Collapse
|
23
|
Detection of epileptic seizure event and onset using EEG. BIOMED RESEARCH INTERNATIONAL 2014; 2014:450573. [PMID: 24616892 PMCID: PMC3925519 DOI: 10.1155/2014/450573] [Citation(s) in RCA: 112] [Impact Index Per Article: 11.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 04/29/2013] [Revised: 11/15/2013] [Accepted: 11/17/2013] [Indexed: 11/17/2022]
Abstract
This study proposes a method of automatic detection of epileptic seizure event and onset using wavelet based features and certain statistical features without wavelet decomposition. Normal and epileptic EEG signals were classified using linear classifier. For seizure event detection, Bonn University EEG database has been used. Three types of EEG signals (EEG signal recorded from healthy volunteer with eye open, epilepsy patients in the epileptogenic zone during a seizure-free interval, and epilepsy patients during epileptic seizures) were classified. Important features such as energy, entropy, standard deviation, maximum, minimum, and mean at different subbands were computed and classification was done using linear classifier. The performance of classifier was determined in terms of specificity, sensitivity, and accuracy. The overall accuracy was 84.2%. In the case of seizure onset detection, the database used is CHB-MIT scalp EEG database. Along with wavelet based features, interquartile range (IQR) and mean absolute deviation (MAD) without wavelet decomposition were extracted. Latency was used to study the performance of seizure onset detection. Classifier gave a sensitivity of 98.5% with an average latency of 1.76 seconds.
Collapse
|
24
|
Parkinsonian tremor identification with multiple local field potential feature classification. J Neurosci Methods 2012; 209:320-30. [DOI: 10.1016/j.jneumeth.2012.06.027] [Citation(s) in RCA: 23] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/12/2012] [Revised: 05/31/2012] [Accepted: 06/25/2012] [Indexed: 11/22/2022]
|
25
|
Tang Y, Durand D. A tunable support vector machine assembly classifier for epileptic seizure detection. EXPERT SYSTEMS WITH APPLICATIONS 2012; 39:3925-3938. [PMID: 22563146 PMCID: PMC3341176 DOI: 10.1016/j.eswa.2011.08.088] [Citation(s) in RCA: 25] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/17/2023]
Abstract
Automating the detection of epileptic seizures could reduce the significant human resources necessary for the care of patients suffering from intractable epilepsy and offer improved solutions for closed-loop therapeutic devices such as implantable electrical stimulation systems. While numerous detection algorithms have been published, an effective detector in the clinical setting remains elusive. There are significant challenges facing seizure detection algorithms. The epilepsy EEG morphology can vary widely among the patient population. EEG recordings from the same patient can change over time. EEG recordings can be contaminated with artifacts that often resemble epileptic seizure activity. In order for an epileptic seizure detector to be successful, it must be able to adapt to these different challenges. In this study, a novel detector is proposed based on a support vector machine assembly classifier (SVMA). The SVMA consists of a group of SVMs each trained with a different set of weights between the seizure and non-seizure data and the user can selectively control the output of the SVMA classifier. The algorithm can improve the detection performance compared to traditional methods by providing an effective tuning strategy for specific patients. The proposed algorithm also demonstrates a clear advantage over threshold tuning. When compared with the detection performances reported by other studies using the publicly available epilepsy dataset hosted by the University of BONN, the proposed SVMA detector achieved the best total accuracy of 98.72%. These results demonstrate the efficacy of the proposed SVMA detector and its potential in the clinical setting.
Collapse
Affiliation(s)
- Y Tang
- Neural Engineering Center, Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio 44106
| | | |
Collapse
|
26
|
Lima CAM, Coelho ALV, Eisencraft M. Tackling EEG signal classification with least squares support vector machines: a sensitivity analysis study. Comput Biol Med 2010; 40:705-14. [PMID: 20621291 DOI: 10.1016/j.compbiomed.2010.06.005] [Citation(s) in RCA: 27] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/01/2009] [Revised: 04/19/2010] [Accepted: 06/11/2010] [Indexed: 11/28/2022]
Abstract
The electroencephalogram (EEG) signal captures the electrical activity of the brain and is an important source of information for studying neurological disorders. The proper analysis of this biological signal plays an important role in the domain of brain-computer interface, which aims at the construction of communication channels between human brain and computers. In this paper, we investigate the application of least squares support vector machines (LS-SVM) to the task of epilepsy diagnosis through automatic EEG signal classification. More specifically, we present a sensitivity analysis study by means of which the performance levels exhibited by standard and least squares SVM classifiers are contrasted, taking into account the setting of the kernel function and of its parameter value. Results of experiments conducted over different types of features extracted from a benchmark EEG signal dataset evidence that the sensitivity profiles of the kernel machines are qualitatively similar, both showing notable performance in terms of accuracy and generalization. In addition, the performance accomplished by optimally configured LS-SVM models is also quantitatively contrasted with that obtained by related approaches for the same dataset.
Collapse
Affiliation(s)
- Clodoaldo A M Lima
- Graduate Program in Electrical Engineering, School of Engineering, Mackenzie Presbyterian University, Rua da Consolação, 896, 01302-907 São Paulo, SP, Brazil.
| | | | | |
Collapse
|
27
|
Choi YS, Koenig MA, Jia X, Thakor NV. Quantifying time-varying multiunit neural activity using entropy based measures. IEEE Trans Biomed Eng 2010; 57. [PMID: 20460201 DOI: 10.1109/tbme.2010.2049266] [Citation(s) in RCA: 25] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
Abstract
Modern micro-electrode arrays make it possible to simultaneously record population neural activity. However, methods to analyze multiunit activity (MUA), which reflects the aggregate spiking activity of a population of neurons, have remained underdeveloped in comparison to those used for studying single unit activity (SUA). In scenarios where SUA is hard to record and maintain or is not representative of brains response, MUA is informative in deciphering the brains complex time-varying response to stimuli or to clinical insults. Here, we present two quantitative methods of analysis of the time-varying dynamics of MUA without spike detection. These methods are based on the multiresolution discrete wavelet transform (DWT) of an envelope of MUA followed by information theoretic measures: multiresolution entropy (MRE) and the multiresolution Kullback-Leibler distance (MRKLD). We test the proposed quantifiers on both simulated and experimental MUA recorded from rodent cortex in an experimental model of global hypoxic-ischemic brain injury. First, our results validate the use of the envelope of MUA as an alternative to detecting and analyzing transient and complex spike activity. Second, the MRE and MRKLD are shown to respond to dynamic changes due to the brains response to global injury and to identify the transient changes in the MUA.
Collapse
|