scholarly journals EEG-Based Emotion Recognition by Convolutional Neural Network with Multi-Scale Kernels

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5092
Author(s):  
Tran-Dac-Thinh Phan ◽  
Soo-Hyung Kim ◽  
Hyung-Jeong Yang ◽  
Guee-Sang Lee
Keyword(s):  
Emotion Recognition ◽  
Binary Classification ◽  
Eeg Signals ◽  
Frequency Bands ◽  
Multi Scale ◽  
Eeg Data ◽  
Crucial Information ◽  
The Time Domain ◽  
Electroencephalogram Eeg ◽  
Selection Of

Besides facial or gesture-based emotion recognition, Electroencephalogram (EEG) data have been drawing attention thanks to their capability in countering the effect of deceptive external expressions of humans, like faces or speeches. Emotion recognition based on EEG signals heavily relies on the features and their delineation, which requires the selection of feature categories converted from the raw signals and types of expressions that could display the intrinsic properties of an individual signal or a group of them. Moreover, the correlation or interaction among channels and frequency bands also contain crucial information for emotional state prediction, and it is commonly disregarded in conventional approaches. Therefore, in our method, the correlation between 32 channels and frequency bands were put into use to enhance the emotion prediction performance. The extracted features chosen from the time domain were arranged into feature-homogeneous matrices, with their positions following the corresponding electrodes placed on the scalp. Based on this 3D representation of EEG signals, the model must have the ability to learn the local and global patterns that describe the short and long-range relations of EEG channels, along with the embedded features. To deal with this problem, we proposed the 2D CNN with different kernel-size of convolutional layers assembled into a convolution block, combining features that were distributed in small and large regions. Ten-fold cross validation was conducted on the DEAP dataset to prove the effectiveness of our approach. We achieved the average accuracies of 98.27% and 98.36% for arousal and valence binary classification, respectively.

scholarly journals Multi-Scale Frequency Bands Ensemble Learning for EEG-Based Emotion Recognition

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1262
Author(s):  
Fangyao Shen ◽  
Yong Peng ◽  
Wanzeng Kong ◽  
Guojun Dai
Keyword(s):  
Emotion Recognition ◽  
Ensemble Learning ◽  
Global Scale ◽  
Learning Method ◽  
Eeg Signals ◽  
Data Set ◽  
Frequency Bands ◽  
Multi Scale ◽  
Adaptive Weight ◽  
Weight Learning

Emotion recognition has a wide range of potential applications in the real world. Among the emotion recognition data sources, electroencephalography (EEG) signals can record the neural activities across the human brain, providing us a reliable way to recognize the emotional states. Most of existing EEG-based emotion recognition studies directly concatenated features extracted from all EEG frequency bands for emotion classification. This way assumes that all frequency bands share the same importance by default; however, it cannot always obtain the optimal performance. In this paper, we present a novel multi-scale frequency bands ensemble learning (MSFBEL) method to perform emotion recognition from EEG signals. Concretely, we first re-organize all frequency bands into several local scales and one global scale. Then we train a base classifier on each scale. Finally we fuse the results of all scales by designing an adaptive weight learning method which automatically assigns larger weights to more important scales to further improve the performance. The proposed method is validated on two public data sets. For the “SEED IV” data set, MSFBEL achieves average accuracies of 82.75%, 87.87%, and 78.27% on the three sessions under the within-session experimental paradigm. For the “DEAP” data set, it obtains average accuracy of 74.22% for four-category classification under 5-fold cross validation. The experimental results demonstrate that the scale of frequency bands influences the emotion recognition rate, while the global scale that directly concatenating all frequency bands cannot always guarantee to obtain the best emotion recognition performance. Different scales provide complementary information to each other, and the proposed adaptive weight learning method can effectively fuse them to further enhance the performance.

scholarly journals The challenges of emotion recognition methods based on electroencephalogram signals: a literature review

2022 ◽  
Vol 12 (2) ◽  
pp. 1508
Author(s):  
I Made Agus Wirawan ◽  
Retantyo Wardoyo ◽  
Danang Lelono
Keyword(s):  
Emotion Recognition ◽  
Eeg Signals ◽  
Data Set ◽  
Eeg Data ◽  
Network Method ◽  
Reduction Methods ◽  
Research Questions ◽  
Participant Characteristics ◽  
Diverse Data ◽  
Electroencephalogram Eeg

Electroencephalogram (EEG) signals in recognizing emotions have several advantages. Still, the success of this study, however, is strongly influenced by: i) the distribution of the data used, ii) consider of differences in participant characteristics, and iii) consider the characteristics of the EEG signals. In response to these issues, this study will examine three important points that affect the success of emotion recognition packaged in several research questions: i) What factors need to be considered to generate and distribute EEG data?, ii) How can EEG signals be generated with consideration of differences in participant characteristics?, and iii) How do EEG signals with characteristics exist among its features for emotion recognition? The results, therefore, indicate some important challenges to be studied further in EEG signals-based emotion recognition research. These include i) determine robust methods for imbalanced EEG signals data, ii) determine the appropriate smoothing method to eliminate disturbances on the baseline signals, iii) determine the best baseline reduction methods to reduce the differences in the characteristics of the participants on the EEG signals, iv) determine the robust architecture of the capsule network method to overcome the loss of knowledge information and apply it in more diverse data set.

Sparse Bayesian Learning for Obtaining Sparsity of EEG Frequency Bands Based Feature Vectors in Motor Imagery Classification

2016 ◽  
Vol 27 (02) ◽  
pp. 1650032 ◽  
Author(s):  
Yu Zhang ◽  
Yu Wang ◽  
Jing Jin ◽  
Xingyu Wang
Keyword(s):  
Motor Imagery ◽  
Bayesian Learning ◽  
Sparse Bayesian Learning ◽  
Common Spatial Pattern ◽  
Linear Discriminant ◽  
Frequency Bands ◽  
Eeg Data ◽  
Eeg Recordings ◽  
Electroencephalogram Eeg ◽  
Selection Of

Effective common spatial pattern (CSP) feature extraction for motor imagery (MI) electroencephalogram (EEG) recordings usually depends on the filter band selection to a large extent. Subband optimization has been suggested to enhance classification accuracy of MI. Accordingly, this study introduces a new method that implements sparse Bayesian learning of frequency bands (named SBLFB) from EEG for MI classification. CSP features are extracted on a set of signals that are generated by a filter bank with multiple overlapping subbands from raw EEG data. Sparse Bayesian learning is then exploited to implement selection of significant features with a linear discriminant criterion for classification. The effectiveness of SBLFB is demonstrated on the BCI Competition IV IIb dataset, in comparison with several other competing methods. Experimental results indicate that the SBLFB method is promising for development of an effective classifier to improve MI classification.

scholarly journals Biological Computation Indexes of Brain Oscillations in Unattended Facial Expression Processing Based on Event-Related Synchronization/Desynchronization

2016 ◽  
Vol 2016 ◽  
pp. 1-17 ◽  
Author(s):  
Bo Yu ◽  
Lin Ma ◽  
Haifeng Li ◽  
Lun Zhao ◽  
Hongjian Bo ◽  
...  
Keyword(s):  
Facial Expressions ◽  
Emotional Experience ◽  
Vital Role ◽  
Brain Oscillations ◽  
Eeg Signals ◽  
Frequency Bands ◽  
Set Up ◽  
Biological Computation ◽  
Expression Processing ◽  
Electroencephalogram Eeg

Estimation of human emotions from Electroencephalogram (EEG) signals plays a vital role in affective Brain Computer Interface (BCI). The present study investigated the different event-related synchronization (ERS) and event-related desynchronization (ERD) of typical brain oscillations in processing Facial Expressions under nonattentional condition. The results show that the lower-frequency bands are mainly used to update Facial Expressions and distinguish the deviant stimuli from the standard ones, whereas the higher-frequency bands are relevant to automatically processing different Facial Expressions. Accordingly, we set up the relations between each brain oscillation and processing unattended Facial Expressions by the measures of ERD and ERS. This research first reveals the contributions of each frequency band for comprehension of Facial Expressions in preattentive stage. It also evidences that participants have emotional experience under nonattentional condition. Therefore, the user’s emotional state under nonattentional condition can be recognized in real time by the ERD/ERS computation indexes of different frequency bands of brain oscillations, which can be used in affective BCI to provide the user with more natural and friendly ways.

scholarly journals Detection of Burst Suppression Patterns in EEG Using Recurrence Rate

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Zhenhu Liang ◽  
Yinghua Wang ◽  
Yongshao Ren ◽  
Duan Li ◽  
Logan Voss ◽  
...  
Keyword(s):  
Recurrence Rate ◽  
Brain Activity ◽  
Approximate Entropy ◽  
Energy Operator ◽  
Burst Suppression ◽  
Multiple Comparison ◽  
Eeg Data ◽  
Electroencephalogram Eeg ◽  
Nonlinear Energy ◽  
Selection Of

Burst suppression is a unique electroencephalogram (EEG) pattern commonly seen in cases of severely reduced brain activity such as overdose of general anesthesia. It is important to detect burst suppression reliably during the administration of anesthetic or sedative agents, especially for cerebral-protective treatments in various neurosurgical diseases. This study investigates recurrent plot (RP) analysis for the detection of the burst suppression pattern (BSP) in EEG. The RP analysis is applied to EEG data containing BSPs collected from 14 patients. Firstly we obtain the best selection of parameters for RP analysis. Then, the recurrence rate (RR), determinism (DET), and entropy (ENTR) are calculated. Then RR was selected as the best BSP index one-way analysis of variance (ANOVA) and multiple comparison tests. Finally, the performance of RR analysis is compared with spectral analysis, bispectral analysis, approximate entropy, and the nonlinear energy operator (NLEO). ANOVA and multiple comparison tests showed that the RR could detect BSP and that it was superior to other measures with the highest sensitivity of suppression detection (96.49%, P=0.03). Tracking BSP patterns is essential for clinical monitoring in critically ill and anesthetized patients. The purposed RR may provide an effective burst suppression detector for developing new patient monitoring systems.

A Portable Electroencephalogram Recording System for Rats

2014 ◽  
Vol 577 ◽  
pp. 1236-1240
Author(s):  
Dian Zhang ◽  
Bo Wang ◽  
Qing Liang Qin
Keyword(s):  
Recording System ◽  
Band Pass Filter ◽  
Test Results ◽  
Electronic Components ◽  
Eeg Signals ◽  
Pass Filter ◽  
Wireless Transmitter ◽  
Eeg Data ◽  
Band Pass ◽  
Electroencephalogram Eeg

A wireless portable electroencephalogram (EEG) recording system for animals was designed, manufactured and then tested in rats. The system basically consisted of four modules: 1) EEG collecting module with the wireless transmitter and receiver (designed by NRF24LE1), 2) filter bank consisting of pre-amplifier, band pass filter and 50Hz trapper, 3) power management module and 4) display interface for showing EEG signals. The EEG data were modulated firstly and emitted by the wireless transmitter after being amplified and filtered. The receiver demodulated and displayed the signals in voltage through serial port. The system was designed as surface mount devices (SMD) with small size (20mm×25mm×3mm) and light weight (4g), and was fabricated of electronic components that were commercially available. The test results indicated that in given environment the system could stably record more than 8 hours and transmit EEG signals over a distance of 20m. Our system showed the features of small size, low power consumption and high accuracy which were suitable for EEG telemetry in rats.

scholarly journals Electroencephalogram of Happy Emotional Cognition Based on Complex System of Music and Image Visual and Auditory

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-14
Author(s):  
Lin Gan ◽  
Mu Zhang ◽  
Jiajia Jiang ◽  
Fajie Duan
Keyword(s):  
Complex System ◽  
Cognitive Neuroscience ◽  
Auditory Stimulation ◽  
External Information ◽  
Eeg Signals ◽  
Baseline Drift ◽  
Average Accuracy ◽  
Eeg Data ◽  
Electroencephalogram Eeg ◽  
The Brain

People are ingesting various information from different sense organs all the time to complete different cognitive tasks. The brain integrates and regulates this information. The two significant sensory channels for receiving external information are sight and hearing that have received extensive attention. This paper mainly studies the effect of music and visual-auditory stimulation on electroencephalogram (EEG) of happy emotion recognition based on a complex system. In the experiment, the presentation was used to prepare the experimental stimulation program, and the cognitive neuroscience experimental paradigm of EEG evoked by happy emotion pictures was established. Using 93 videos as natural stimuli, fMRI data were collected. Finally, the collected EEG signals were removed with the eye artifact and baseline drift, and the t-test was used to analyze the significant differences of different lead EEG data. Experimental data shows that, by adjusting the parameters of the convolutional neural network, the highest accuracy of the two-classification algorithm can reach 98.8%, and the average accuracy can reach 83.45%. The results show that the brain source under the combined visual and auditory stimulus is not a simple superposition of the brain source of the single visual and auditory stimulation, but a new interactive source is generated.

scholarly journals Classification of Motor Imagery Using a Combination of User-Specific Band and Subject-Specific Band for Brain-Computer Interface

2019 ◽  
Vol 9 (23) ◽  
pp. 4990 ◽  
Author(s):  
Jusas ◽  
Samuvel
Keyword(s):  
Feature Extraction ◽  
Time Domain ◽  
Extraction Process ◽  
Computer Interface ◽  
Eeg Signals ◽  
Frequency Bands ◽  
Fisher Ratio ◽  
Frequency Domains ◽  
Subject Specific ◽  
The Time Domain

The essential task of a Brain-Computer Interface (BCI) is to extract the motor imagery features from Electro-Encephalogram (EEG) signals for classifying the thought process. It is necessary to analyse these obtained signals in both the time domain and frequency domains. It is observed that the combination of multiple algorithms increases the performance of the feature extraction process. This paper identifies combinations that have not been attempted previously and improves the accuracy of the overall process, although other authors implemented different combinations of the techniques. The focus is given more on the feature extraction process and frequency bands, which are user-specific and subject-specific frequency bands. In both time and frequency domains, after analysing EEG signals with the time domain parameter, we select the frequency band and the timing while using the Fisher ratio of the time domain parameter (TDP). We used Fisher discriminant analysis (FDA)-type F-score to simultaneously select the frequency band and time segment for multi-class classification. We extracted subject-specific TDP features from the training trials to train the classifier when optimal time-frequency areas were selected for each subject. In this paper, various methods are explored for obtaining the features, which are Time Domain Parameters (TDP), Fast Fourier Transform (FFT), Principal Component Analysis (PCA), R2, Fast Correlation Based Filter (FCBF), Empirical Mode Decomposition (EMD), and Intrinsic time-scale decomposition (ITD). After the extraction process, PCA is used for dimensionality reduction. An efficient result was obtained with the combination of TDP, FFT, and PCA. We used the multi-class Fisher′s linear discriminant analysis (LDA) as the classifier, which was in line with the FDA-type F-score. It is observed that the combination of feature extraction techniques to the frequency bands that were selected by the Fisher ratio and FDA type F-score along with Fisher′s LDA classifier had higher accuracy than the results obtained other researches. A kappa coefficient accuracy of 0.64 is obtained for the proposed technique. Our method leads to better classification performance when compared to state-of-the-art methods. The novelty of the approach is based on the combination of frequency bands and two feature extraction methods.

Effect of Wavelet Packet Log Energy Entropy on Electroencephalogram (EEG) Signals

2015 ◽  
Vol 4 (1) ◽  
pp. 32-43 ◽  
Author(s):  
S. Raghu ◽  
N. Sriraam ◽  
G. Pradeep Kumar
Keyword(s):  
Epileptic Seizure ◽  
Linear Time ◽  
Wavelet Packet ◽  
Research Work ◽  
Eeg Signals ◽  
Time Frequency ◽  
Energy Entropy ◽  
Evoked Response Potential ◽  
Electroencephalogram Eeg ◽  
Selection Of

The scaling behavior of human electroencephalogram (EEG) signals is well exploited by appropriate extraction of time – frequency domain and entropy based features. Such measurable inherently helps understanding the neurophysiological phenomenon of brain as well as its associated cortical activities. Being a non-linear time series, EEG's are assumed to be fragment of fluctuations. Several attempts have been made to study the EEG signals for clinical applications such as epileptic seizure detection, evoked response potential recognition, tumor detection, identification of alcoholics and so on. In all such applications appropriate selection of feature parameter plays an important role in discriminating normal EEG from abnormal. In the recent past one can find the importance of wavelet and wavelet packet towards EEG analysis. This proposed research work investigates the effect of wavelet packet log energy entropy on EEG signals. Entropy being the measure of relative information, the proposed study attempts to discriminate the normal EEGs from abnormal EEG's by employing the log energy entropy features. For better brevity, this study restricts to the analysis of epileptic seizure from normal EEGs. Different decomposition levels from 2 to 5 were considered for wavelet packets with application of Haar, rbio3.1, sym7, dmey wavelets. A one second windowing was introduced for the data segmentation and Shannon's log energy entropy was estimated. Then the statistical non-parametric Wilcoxon model was employed. The result shows that the application of wavelet packet log energy entropy found to be a potential indicator for discriminating epileptic seizure from normal.

Sign in / Sign up

Export Citation Format

Share Document