Fast Recognition of BCI-Inefficient Users Using Physiological Features from EEG Signals: A Screening Study of Stroke Patients

Emotion assessment in stroke patients gives meaningful information to physiotherapists to identify the appropriate method for treatment. This study was aimed to classify the emotions of stroke patients by applying bispectrum features in electroencephalogram (EEG) signals. EEG signals from three groups of subjects, namely stroke patients with left brain damage (LBD), right brain damage (RBD), and normal control (NC), were analyzed for six different emotional states. The estimated bispectrum mapped in the contour plots show the different appearance of nonlinearity in the EEG signals for different emotional states. Bispectrum features were extracted from the alpha (8–13) Hz, beta (13–30) Hz and gamma (30–49) Hz bands, respectively. The k-nearest neighbor (KNN) and probabilistic neural network (PNN) classifiers were used to classify the six emotions in LBD, RBD and NC. The bispectrum features showed statistical significance for all three groups. The beta frequency band was the best performing EEG frequency-sub band for emotion classification. The combination of alpha to gamma bands provides the highest classification accuracy in both KNN and PNN classifiers. Sadness emotion records the highest classification, which was 65.37% in LBD, 71.48% in RBD and 75.56% in NC groups.

Download Full-text

Hemispheric Coherence Analysis of Wide Band EEG Signals for Characterization of Post-Stroke Patients with Dementia

Traitement du signal ◽

10.18280/ts.380408 ◽

2021 ◽

Vol 38 (4) ◽

pp. 985-992

Author(s):

Sugondo Hadiyoso ◽

Hasballah Zakaria ◽

Paulus Anam Ong ◽

Tati Latifah E.R. Mengko

Keyword(s):

Quantitative Analysis ◽

Wide Band ◽

P Value ◽

Coherence Analysis ◽

Stroke Survivors ◽

Eeg Signals ◽

Stroke Patients ◽

Post Stroke ◽

Significant Difference

Post-stroke dementia (PSD) is a type of vascular dementia (VaD) that might be occurred in post-stroke patients. Memory, language and behavior tests can be used for the analysis of cognitive impairment caused by PSD. Often a supporting clinical examination such as an electroencephalogram (EEG) is used to support the diagnosis or analyze the characteristic changes that occur in the brain. Conventional analysis or visual inspection of EEG signals can be very difficult, since the nature of the signal tends to be non-stationer. Therefore, this study proposes a quantitative analysis for the characterization of EEG signals in stroke survivors with dementia. It is thought that it has different characteristics with the normal subject so that this study can be used as a reference in supporting dementia detection in post-stroke survivors. The quantitative analysis used in this study is coherence analysis. Coherence analysis was performed on EEG signals recorded from six poststroke patients with dementia and then compared with ten normal healthy subjects. Analysis of coherence between brain areas includes inter and intra-hemispheric coherence. Validation was carried out by using the independent t-test where the confidence level was 95%, indicating that the p-value <0.05 had a significant difference. The test results show that in general the coherence of the electrode pairs in patients with dementia is lower than in the normal healthy group. It is notably, i) In interhemispheric, the C3-C4, T3-T4, and T5-T6 pairs generate significant differences, ii) the highest decrease in intrahemispheric coherence was found in C3-T5 with p = 0.0005. The coherence study presented in this paper is expected to be used for early detection of PSD in the future.

Download Full-text

Hurst Exponent Based Brain Behavior Analysis of Stroke Patients Using EEG Signals

Lecture Notes in Electrical Engineering - Proceedings of the 11th National Technical Seminar on Unmanned System Technology 2019 ◽

10.1007/978-981-15-5281-6_66 ◽

2020 ◽

pp. 925-933

Author(s):

Wen Yean Choong ◽

Wan Khairunizam ◽

Murugappan Murugappan ◽

Mohammad Iqbal Omar ◽

Siao Zheng Bong ◽

...

Keyword(s):

Behavior Analysis ◽

Hurst Exponent ◽

Eeg Signals ◽

Stroke Patients ◽

Brain Behavior

Download Full-text

A Multifrequency Brain Network-Based Deep Learning Framework for Motor Imagery Decoding

Neural Plasticity ◽

10.1155/2020/8863223 ◽

2020 ◽

Vol 2020 ◽

pp. 1-11

Author(s):

Juntao Xue ◽

Feiyue Ren ◽

Xinlin Sun ◽

Miaomiao Yin ◽

Jialing Wu ◽

...

Keyword(s):

Deep Learning ◽

Motor Imagery ◽

Neural System ◽

Brain Network ◽

Eeg Signals ◽

Stroke Patients ◽

Common Spatial Pattern ◽

Convolutional Network ◽

The Public ◽

Learning Framework

Motor imagery (MI) is an important part of brain-computer interface (BCI) research, which could decode the subject’s intention and help remodel the neural system of stroke patients. Therefore, accurate decoding of electroencephalography- (EEG-) based motion imagination has received a lot of attention, especially in the research of rehabilitation training. We propose a novel multifrequency brain network-based deep learning framework for motor imagery decoding. Firstly, a multifrequency brain network is constructed from the multichannel MI-related EEG signals, and each layer corresponds to a specific brain frequency band. The structure of the multifrequency brain network matches the activity profile of the brain properly, which combines the information of channel and multifrequency. The filter bank common spatial pattern (FBCSP) algorithm filters the MI-based EEG signals in the spatial domain to extract features. Further, a multilayer convolutional network model is designed to distinguish different MI tasks accurately, which allows extracting and exploiting the topology in the multifrequency brain network. We use the public BCI competition IV dataset 2a and the public BCI competition III dataset IIIa to evaluate our framework and get state-of-the-art results in the first dataset, i.e., the average accuracy is 83.83% and the value of kappa is 0.784 for the BCI competition IV dataset 2a, and the accuracy is 89.45% and the value of kappa is 0.859 for the BCI competition III dataset IIIa. All these results demonstrate that our framework can classify different MI tasks from multichannel EEG signals effectively and show great potential in the study of remodelling the neural system of stroke patients.

Download Full-text

Comparison of Features for Movement Prediction from Single-Trial Movement-Related Cortical Potentials in Healthy Subjects and Stroke Patients

Computational Intelligence and Neuroscience ◽

10.1155/2015/858015 ◽

2015 ◽

Vol 2015 ◽

pp. 1-8 ◽

Cited By ~ 15

Author(s):

Ernest Nlandu Kamavuako ◽

Mads Jochumsen ◽

Imran Khan Niazi ◽

Kim Dremstrup

Keyword(s):

Template Matching ◽

Spectral Features ◽

Single Trial ◽

Eeg Signals ◽

Stroke Patients ◽

Movement Execution ◽

Linear Discriminant ◽

Temporal Features ◽

Electroencephalogram Eeg ◽

Movement Intention

Detection of movement intention from the movement-related cortical potential (MRCP) derived from the electroencephalogram (EEG) signals has shown to be important in combination with assistive devices for effective neurofeedback in rehabilitation. In this study, we compare time and frequency domain features to detect movement intention from EEG signals prior to movement execution. Data were recoded from 24 able-bodied subjects, 12 performing real movements, and 12 performing imaginary movements. Furthermore, six stroke patients with lower limb paresis were included. Temporal and spectral features were investigated in combination with linear discriminant analysis and compared with template matching. The results showed that spectral features were best suited for differentiating between movement intention and noise across different tasks. The ensemble average across tasks when using spectral features was (error = 3.4 ± 0.8%, sensitivity = 97.2 ± 0.9%, and specificity = 97 ± 1%) significantly better (P<0.01) than temporal features (error = 15 ± 1.4%, sensitivity: 85 ± 1.3%, and specificity: 84 ± 2%). The proposed approach also (error = 3.4 ± 0.8%) outperformed template matching (error = 26.9 ± 2.3%) significantly (P>0.001). Results imply that frequency information is important for detecting movement intention, which is promising for the application of this approach to provide patient-driven real-time neurofeedback.

Download Full-text

Brainwave Classification of Task Performed by Stroke Patients using ANN

Annals of Emerging Technologies in Computing ◽

10.33166/aetic.2021.05.004 ◽

2021 ◽

Vol 5 (5) ◽

pp. 34-40

Author(s):

S.K. Narudin ◽

N.H.M. Nasir ◽

N. Fuad

Keyword(s):

The State ◽

Classification Model ◽

Eeg Signals ◽

Stroke Patients ◽

Data Set ◽

Input Signals ◽

The Subject ◽

Artificial Neural Network Ann ◽

Electroencephalogram Eeg

In this research, 14 stroke patient's brainwave activity with open eyes (OE) and close eyes (CE) sessions are used. This work aims to study and classify 2 activities that validate our data acquisition. The data set of each subject is used to classify the state of the subject during electroencephalogram (EEG) recording. For the classification model, the input signals are alpha, beta, theta, and delta bands. The classification algorithm used in this work is the Artificial Neural Network (ANN). The accuracy value will be obtained from each subject. There are substancial differences between the EEG signals of each patient and hence affecting the accuracy value of the subject. The results obtained from our experiment proved that ANN can be used to classify the state of the subject during data recording.

Download Full-text

The classification of EEG-based winking signals: a transfer learning and random forest pipeline

PeerJ ◽

10.7717/peerj.11182 ◽

2021 ◽

Vol 9 ◽

pp. e11182

Author(s):

Jothi Letchumy Mahendra Kumar ◽

Mamunur Rashid ◽

Rabiu Muazu Musa ◽

Mohd Azraai Mohd Razman ◽

Norizam Sulaiman ◽

...

Keyword(s):

Random Forest ◽

Transfer Learning ◽

Life Quality ◽

Validation Dataset ◽

Eeg Signals ◽

Stroke Patients ◽

Classifier Performance ◽

Search Approach ◽

Bci System

Brain Computer-Interface (BCI) technology plays a considerable role in the control of rehabilitation or peripheral devices for stroke patients. This is particularly due to their inability to control such devices from their inherent physical limitations after such an attack. More often than not, the control of such devices exploits electroencephalogram (EEG) signals. Nonetheless, it is worth noting that the extraction of the features and the classification of the signals is non-trivial for a successful BCI system. The use of Transfer Learning (TL) has been demonstrated to be a powerful tool in the extraction of essential features. However, the employment of such a method towards BCI applications, particularly in regard to EEG signals, are somewhat limited. The present study aims to evaluate the effectiveness of different TL models in extracting features for the classification of wink-based EEG signals. The extracted features are classified by means of fine-tuned Random Forest (RF) classifier. The raw EEG signals are transformed into a scalogram image via Continuous Wavelet Transform (CWT) before it was fed into the TL models, namely InceptionV3, Inception ResNetV2, Xception and MobileNet. The dataset was divided into training, validation, and test datasets, respectively, via a stratified ratio of 60:20:20. The hyperparameters of the RF models were optimised through the grid search approach, in which the five-fold cross-validation technique was adopted. The optimised RF classifier performance was compared with the conventional TL-based CNN classifier performance. It was demonstrated from the study that the best TL model identified is the Inception ResNetV2 along with an optimised RF pipeline, as it was able to yield a classification accuracy of 100% on both the training and validation dataset. Therefore, it could be established from the study that a comparable classification efficacy is attainable via the Inception ResNetV2 with an optimised RF pipeline. It is envisaged that the implementation of the proposed architecture to a BCI system would potentially facilitate post-stroke patients to lead a better life quality.

Download Full-text

Effect of Post-Stroke Motor Training on EEG Movement-Related Cortical Potentials

10.21203/rs.3.rs-50667/v1 ◽

2020 ◽

Author(s):

Maryam Butt ◽

Golshah Naghdy ◽

Fazel Naghdy ◽

Geoffrey Murray ◽

Haiping Du

Keyword(s):

Low Cost ◽

Motor Training ◽

Clinical Test ◽

Kinematic Parameters ◽

Eeg Signals ◽

Hand Rehabilitation ◽

Stroke Patients ◽

Post Stroke ◽

Group A ◽

Group B

Abstract BackgroundRehabilitation of post-stroke patients with motor impairments promotes re-learning of lost motor functions through the brain neuroplasticity. Monitoring of electroencephalogram (EEG) signals has the potential to show neuroplasticity changes that take place during motor training.MethodsIn this study, an EEG-derived time-domain pattern namely movement-related cortical potential (MRCP) was deployed to assess the effect of motor training in seven post-stroke patients. Patients were divided into two groups; group A comprising four subjects with supratentorial lesions and group B consisting of three subjects with infratentorial lesions. Both groups participated in motor training with an AMADEO hand rehabilitation device. During pre and post-training periods, EEG signals at eight selected electrodes were recorded. In addition, hand-kinematic parameters, and clinical tests were measured at the beginning and the end of all training sessions.ResultsThe negative peak of the MRCP signals decreased at all electrodes and reached significance in seven of eight electrodes for group A after 12 training sessions, while it was decreased at all electrodes and reached significance in two of eight electrodes for group B after 24 sessions according to paired t-test (p < 0.05). Moreover, these MRCP changes correlated with improvements in kinematic parameters and clinical test results for both groups.ConclusionsThis study shows that robot-assisted training that improves clinical outcomes is associated with MRCP pattern changes. Subjects with infratentorial strokes improved slower clinically compared to subjects with supratentorial strokes. This was consistent with the longer rehabilitation required for this group of patients to produce significant changes in MRCP. The reduction of negative peaks of the MRCP signal indicates that neurological pathways are established and less cortical resources are needed for motor tasks. This study demonstrates the significance of EEG as a practical and low-cost tool in detecting patterns associated with brain neuroplasticity in the course of motor re-learning. Ethics ApprovalThe procedures performed in this study were approved by the University of Wollongong Ethics Committee (Ethics application number: 2014/400) on 03/07/2017.

Download Full-text