1
|
Degirmenci M, Yuce YK, Perc M, Isler Y. EEG-based finger movement classification with intrinsic time-scale decomposition. Front Hum Neurosci 2024; 18:1362135. [PMID: 38505099 PMCID: PMC10948500 DOI: 10.3389/fnhum.2024.1362135] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/27/2023] [Accepted: 02/15/2024] [Indexed: 03/21/2024] Open
Abstract
Introduction Brain-computer interfaces (BCIs) are systems that acquire the brain's electrical activity and provide control of external devices. Since electroencephalography (EEG) is the simplest non-invasive method to capture the brain's electrical activity, EEG-based BCIs are very popular designs. Aside from classifying the extremity movements, recent BCI studies have focused on the accurate coding of the finger movements on the same hand through their classification by employing machine learning techniques. State-of-the-art studies were interested in coding five finger movements by neglecting the brain's idle case (i.e., the state that brain is not performing any mental tasks). This may easily cause more false positives and degrade the classification performances dramatically, thus, the performance of BCIs. This study aims to propose a more realistic system to decode the movements of five fingers and the no mental task (NoMT) case from EEG signals. Methods In this study, a novel praxis for feature extraction is utilized. Using Proper Rotational Components (PRCs) computed through Intrinsic Time Scale Decomposition (ITD), which has been successfully applied in different biomedical signals recently, features for classification are extracted. Subsequently, these features were applied to the inputs of well-known classifiers and their different implementations to discriminate between these six classes. The highest classifier performances obtained in both subject-independent and subject-dependent cases were reported. In addition, the ANOVA-based feature selection was examined to determine whether statistically significant features have an impact on the classifier performances or not. Results As a result, the Ensemble Learning classifier achieved the highest accuracy of 55.0% among the tested classifiers, and ANOVA-based feature selection increases the performance of classifiers on five-finger movement determination in EEG-based BCI systems. Discussion When compared with similar studies, proposed praxis achieved a modest yet significant improvement in classification performance although the number of classes was incremented by one (i.e., NoMT).
Collapse
Affiliation(s)
- Murside Degirmenci
- Department of Biomedical Technologies, Izmir Katip Celebi University, Izmir, Türkiye
| | - Yilmaz Kemal Yuce
- Department of Computer Engineering, Alanya Alaaddin Keykubat University, Alanya, Antalya, Türkiye
| | - Matjaž Perc
- Faculty of Natural Sciences and Mathematics, University of Maribor, Maribor, Slovenia
- Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan
- Complexity Science Hub Vienna, Vienna, Austria
- Department of Physics, Kyung Hee University, Seoul, Republic of Korea
| | - Yalcin Isler
- Department of Biomedical Engineering, Izmir Katip Celebi University, Izmir, Türkiye
| |
Collapse
|
2
|
Park S, Kim M, Nam H, Kwon J, Im CH. In-Car Environment Control Using an SSVEP-Based Brain-Computer Interface with Visual Stimuli Presented on Head-Up Display: Performance Comparison with a Button-Press Interface. Sensors (Basel) 2024; 24:545. [PMID: 38257638 PMCID: PMC10819861 DOI: 10.3390/s24020545] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/16/2023] [Revised: 01/04/2024] [Accepted: 01/12/2024] [Indexed: 01/24/2024]
Abstract
Controlling the in-car environment, including temperature and ventilation, is necessary for a comfortable driving experience. However, it often distracts the driver's attention, potentially causing critical car accidents. In the present study, we implemented an in-car environment control system utilizing a brain-computer interface (BCI) based on steady-state visual evoked potential (SSVEP). In the experiment, four visual stimuli were displayed on a laboratory-made head-up display (HUD). This allowed the participants to control the in-car environment by simply staring at a target visual stimulus, i.e., without pressing a button or averting their eyes from the front. The driving performances in two realistic driving tests-obstacle avoidance and car-following tests-were then compared between the manual control condition and SSVEP-BCI control condition using a driving simulator. In the obstacle avoidance driving test, where participants needed to stop the car when obstacles suddenly appeared, the participants showed significantly shorter response time (1.42 ± 0.26 s) in the SSVEP-BCI control condition than in the manual control condition (1.79 ± 0.27 s). No-response rate, defined as the ratio of obstacles that the participants did not react to, was also significantly lower in the SSVEP-BCI control condition (4.6 ± 14.7%) than in the manual control condition (20.5 ± 25.2%). In the car-following driving test, where the participants were instructed to follow a preceding car that runs at a sinusoidally changing speed, the participants showed significantly lower speed difference with the preceding car in the SSVEP-BCI control condition (15.65 ± 7.04 km/h) than in the manual control condition (19.54 ± 11.51 km/h). The in-car environment control system using SSVEP-based BCI showed a possibility that might contribute to safer driving by keeping the driver's focus on the front and thereby enhancing the overall driving performance.
Collapse
Affiliation(s)
- Seonghun Park
- Department of Electronic Engineering, Hanyang University, Seoul 04763, Republic of Korea; (S.P.); (J.K.)
| | - Minsu Kim
- Department of Artificial Intelligence, Hanyang University, Seoul 04763, Republic of Korea; (M.K.); (H.N.)
| | - Hyerin Nam
- Department of Artificial Intelligence, Hanyang University, Seoul 04763, Republic of Korea; (M.K.); (H.N.)
| | - Jinuk Kwon
- Department of Electronic Engineering, Hanyang University, Seoul 04763, Republic of Korea; (S.P.); (J.K.)
| | - Chang-Hwan Im
- Department of Electronic Engineering, Hanyang University, Seoul 04763, Republic of Korea; (S.P.); (J.K.)
- Department of Artificial Intelligence, Hanyang University, Seoul 04763, Republic of Korea; (M.K.); (H.N.)
- Department of Biomedical Engineering, Hanyang University, Seoul 04763, Republic of Korea
| |
Collapse
|
3
|
Chen D, Huang H, Bao X, Pan J, Li Y. An EEG-based attention recognition method: fusion of time domain, frequency domain, and non-linear dynamics features. Front Neurosci 2023; 17:1194554. [PMID: 37502681 PMCID: PMC10368951 DOI: 10.3389/fnins.2023.1194554] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2023] [Accepted: 06/22/2023] [Indexed: 07/29/2023] Open
Abstract
Introduction Attention is a complex cognitive function of human brain that plays a vital role in our daily lives. Electroencephalogram (EEG) is used to measure and analyze attention due to its high temporal resolution. Although several attention recognition brain-computer interfaces (BCIs) have been proposed, there is a scarcity of studies with a sufficient number of subjects, valid paradigms, and reliable recognition analysis across subjects. Methods In this study, we proposed a novel attention paradigm and feature fusion method to extract features, which fused time domain features, frequency domain features and nonlinear dynamics features. We then constructed an attention recognition framework for 85 subjects. Results and discussion We achieved an intra-subject average classification accuracy of 85.05% ± 6.87% and an inter-subject average classification accuracy of 81.60% ± 9.93%, respectively. We further explored the neural patterns in attention recognition, where attention states showed less activation than non-attention states in the prefrontal and occipital areas in α, β and θ bands. The research explores, for the first time, the fusion of time domain features, frequency domain features and nonlinear dynamics features for attention recognition, providing a new understanding of attention recognition.
Collapse
Affiliation(s)
- Di Chen
- School of Automation Science and Engineering, South China University of Technology, Guangzhou, China
- Research Center for Brain-Computer Interface, Pazhou Laboratory, Guangzhou, China
| | - Haiyun Huang
- Research Center for Brain-Computer Interface, Pazhou Laboratory, Guangzhou, China
- School of Software, South China Normal University, Foshan, China
| | - Xiaoyu Bao
- School of Automation Science and Engineering, South China University of Technology, Guangzhou, China
- Research Center for Brain-Computer Interface, Pazhou Laboratory, Guangzhou, China
| | - Jiahui Pan
- Research Center for Brain-Computer Interface, Pazhou Laboratory, Guangzhou, China
- School of Software, South China Normal University, Foshan, China
| | - Yuanqing Li
- School of Automation Science and Engineering, South China University of Technology, Guangzhou, China
- Research Center for Brain-Computer Interface, Pazhou Laboratory, Guangzhou, China
| |
Collapse
|
4
|
Degirmenci M, Yuce YK, Perc M, Isler Y. Statistically significant features improve binary and multiple Motor Imagery task predictions from EEGs. Front Hum Neurosci 2023; 17:1223307. [PMID: 37497042 PMCID: PMC10366537 DOI: 10.3389/fnhum.2023.1223307] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/15/2023] [Accepted: 06/23/2023] [Indexed: 07/28/2023] Open
Abstract
In recent studies, in the field of Brain-Computer Interface (BCI), researchers have focused on Motor Imagery tasks. Motor Imagery-based electroencephalogram (EEG) signals provide the interaction and communication between the paralyzed patients and the outside world for moving and controlling external devices such as wheelchair and moving cursors. However, current approaches in the Motor Imagery-BCI system design require effective feature extraction methods and classification algorithms to acquire discriminative features from EEG signals due to the non-linear and non-stationary structure of EEG signals. This study investigates the effect of statistical significance-based feature selection on binary and multi-class Motor Imagery EEG signal classifications. In the feature extraction process performed 24 different time-domain features, 15 different frequency-domain features which are energy, variance, and entropy of Fourier transform within five EEG frequency subbands, 15 different time-frequency domain features which are energy, variance, and entropy of Wavelet transform based on five EEG frequency subbands, and 4 different Poincare plot-based non-linear parameters are extracted from each EEG channel. A total of 1,364 Motor Imagery EEG features are supplied from 22 channel EEG signals for each input EEG data. In the statistical significance-based feature selection process, the best one among all possible combinations of these features is tried to be determined using the independent t-test and one-way analysis of variance (ANOVA) test on binary and multi-class Motor Imagery EEG signal classifications, respectively. The whole extracted feature set and the feature set that contain statistically significant features only are classified in this study. We implemented 6 and 7 different classifiers in multi-class and binary (two-class) classification tasks, respectively. The classification process is evaluated using the five-fold cross-validation method, and each classification algorithm is tested 10 times. These repeated tests provide to check the repeatability of the results. The maximum of 61.86 and 47.36% for the two-class and four-class scenarios, respectively, are obtained with Ensemble Subspace Discriminant among all these classifiers using selected features including only statistically significant features. The results reveal that the introduced statistical significance-based feature selection approach improves the classifier performances by achieving higher classifier performances with fewer relevant components in Motor Imagery task classification. In conclusion, the main contribution of the presented study is two-fold evaluation of non-linear parameters as an alternative to the commonly used features and the prediction of multiple Motor Imagery tasks using statistically significant features.
Collapse
Affiliation(s)
- Murside Degirmenci
- Department of Biomedical Technologies, Izmir Katip Celebi University, İzmir, Türkiye
| | - Yilmaz Kemal Yuce
- Department of Computer Engineering, Alanya Alaaddin Keykubat University, Antalya, Türkiye
| | - Matjaž Perc
- Faculty of Natural Sciences and Mathematics, University of Maribor, Maribor, Slovenia
- Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan
- Alma Mater Europaea, Maribor, Slovenia
- Complexity Science Hub Vienna, Vienna, Austria
- Department of Physics, Kyung Hee University, Seoul, Republic of Korea
| | - Yalcin Isler
- Department of Biomedical Engineering, Izmir Katip Celebi University, İzmir, Türkiye
| |
Collapse
|
5
|
Horowitz AJ, Guger C, Korostenskaja M. What Internal Variables Affect Sensorimotor Rhythm Brain-Computer Interface (SMR-BCI) Performance? HCA Healthc J Med 2021; 2:163-179. [PMID: 37427003 PMCID: PMC10324829 DOI: 10.36518/2689-0216.1196] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/11/2023]
Abstract
Description In this review article, we aimed to create a summary of the effects of internal variables on the performance of sensorimotor rhythm-based brain computer interfaces (SMR-BCIs). SMR-BCIs can be potentially used for interfacing between the brain and devices, bypassing usual central nervous system output, such as muscle activity. The careful consideration of internal factors, affecting SMR-BCI performance, can maximize BCI application in both healthy and disabled people. Internal variables may be generalized as descriptors of the processes mainly dependent on the BCI user and/or originating within the user. The current review aimed to critically evaluate and summarize the currently accumulated body of knowledge regarding the effect of internal variables on SMR-BCI performance. The examples of such internal variables include motor imagery, hand coordination, attention, motivation, quality of life, mood and neurophysiological signals other than SMR. We will conclude our review with the discussion about the future developments regarding the research on the effects of internal variables on SMR-BCI performance. The end-goal of this review paper is to provide current BCI users and researchers with the reference guide that can help them optimize the SMR-BCI performance by accounting for possible influences of various internal factors.
Collapse
Affiliation(s)
- Alex J. Horowitz
- Functional Brain Mapping and Brain Computer Interface Lab, Neuroscience Institute, AdventHealth Orlando, Orlando, FL,
USA
- University of Central Florida/HCA Healthcare GME Consortium, Gainesville, Florida
| | | | - Milena Korostenskaja
- Functional Brain Mapping and Brain Computer Interface Lab, Neuroscience Institute, AdventHealth Orlando, Orlando, FL,
USA
- MEG Lab, AdventHealth for Children, Orlando, FL,
USA
- Department of Psychology, College of Arts and Sciences, University of North Florida, Jacksonville, FL,
USA
- Comprehensive Epilepsy Center, AdventHealth Orlando, Orlando, FL,
USA
| |
Collapse
|
6
|
Horowitz AJ, Guger C, Korostenskaja M. What External Variables Affect Sensorimotor Rhythm Brain-Computer Interface (SMR-BCI) Performance? HCA Healthc J Med 2021; 2:143-162. [PMID: 37427002 PMCID: PMC10324824 DOI: 10.36518/2689-0216.1188] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/11/2023]
Abstract
Description Sensorimotor rhythm-based brain-computer interfaces (SMR-BCIs) are used for the acquisition and translation of motor imagery-related brain signals into machine control commands, bypassing the usual central nervous system output. The selection of optimal external variable configuration can maximize SMR-BCI performance in both healthy and disabled people. This performance is especially important now when the BCI is targeted for everyday use in the environment beyond strictly regulated laboratory settings. In this review article, we summarize and critically evaluate the current body of knowledge pertaining to the effect of the external variables on SMR-BCI performance. When assessing the relationship between SMR-BCI performance and external variables, we broadly characterize them as elements that are less dependent on the BCI user and originate from beyond the user. These elements include such factors as BCI type, distractors, training, visual and auditory feedback, virtual reality and magneto electric feedback, proprioceptive and haptic feedback, carefulness of electroencephalography (EEG) system assembling and positioning of EEG electrodes as well as recording-related artifacts. At the end of this review paper, future developments are proposed regarding the research into the effects of external variables on SMR-BCI performance. We believe that our critical review will be of value for academic BCI scientists and developers and clinical professionals working in the field of BCIs as well as for SMR-BCI users.
Collapse
Affiliation(s)
- Alex J. Horowitz
- Functional Brain Mapping and Brain Computer Interface Lab, Neuroscience Institute, AdventHealth Orlando, Orlando, FL,
USA
- University of Central Florida/HCA Healthcare GME Consortium, Gainesville, Florida
| | | | - Milena Korostenskaja
- Functional Brain Mapping and Brain Computer Interface Lab, Neuroscience Institute, AdventHealth Orlando, Orlando, FL,
USA
- MEG Lab, AdventHealth for Children, Orlando, FL,
USA
- Department of Psychology, College of Arts and Sciences, University of North Florida, Jacksonville, FL,
USA
- Comprehensive Epilepsy Center, AdventHealth Orlando, Orlando, FL,
USA
| |
Collapse
|
7
|
Park Y, Chung W. Selective Feature Generation Method Based on Time Domain Parameters and Correlation Coefficients for Filter-Bank-CSP BCI Systems. Sensors (Basel) 2019; 19:E3769. [PMID: 31480390 DOI: 10.3390/s19173769] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 07/22/2019] [Revised: 08/21/2019] [Accepted: 08/29/2019] [Indexed: 11/17/2022]
Abstract
This paper presents a novel motor imagery (MI) classification algorithm using filter-bank common spatial pattern (FBCSP) features based on MI-relevant channel selection. In contrast to existing channel selection methods based on global CSP features, the proposed algorithm utilizes the Fisher ratio of time domain parameters (TDPs) and correlation coefficients: the channel with the highest Fisher ratio of TDPs, named principle channel, is selected and a supporting channel set for the principle channel that consists of highly correlated channels to the principle channel is generated. The proposed algorithm using the FBCSP features generated from the supporting channel set for the principle channel significantly improved the classification performance. The performance of the proposed method was evaluated using BCI Competition III Dataset IVa (18 channels) and BCI Competition IV Dataset I (59 channels).
Collapse
|