A novel hybrid kernel function relevance vector machine for multi-task motor imagery EEG classification

Dong, Enzeng; Zhou, Kairui; Tong, Jigang; Du, Shengzhi

doi:10.1016/j.bspc.2020.101991

Cited by 45 publications

(15 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…International Journal of Intelligent Engineering and Systems, Vol. 15 Further improvement might put emphasis on utilizing α and w hyperparameters together with RVs for extracting rules through the finding of decision boundary in form of hyper-rectangles. Moreover, the performance criterion here is restricted to just classification performance.…”

Section: Conflicts Of Interestmentioning

confidence: 99%

“…Compared to the emotional facial-expression recognition, emotion processing in International Journal of Intelligent Engineering and Systems, Vol. 15 the brain appears early with an interval of approximately 180 milliseconds before emotional facial processing arises [5]. Therefore, the motivation for performing emotion recognition based on EEG signals is logical.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, by using RVs from RVM model, it is expected to gain an in-depth understanding behind the prediction making on data samples. Since its appearance, RVM has been implemented in several studies on EEG such as in motor imagery [14,15], mental fatigue detection [16], and many more.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Rules Extraction of Relevance Vector Machine for Predicting Negative Emotions from EEG Signals

2022

IJIES

View full text Add to dashboard Cite

Many studies have reported that patients who are experiencing long-term negative emotions have higher risk of having health deterioration. Therefore, recognition of negative emotions from Electroencephalography (EEG) signals is crucial for monitoring patient conditions. In EEG emotion recognition, clinicians tend to need a clear explanation regarding the rules behind the EEG emotion classification process. Most of the EEG emotions classification use Support Vector Machine (SVM) causing lack of probabilistic prediction which can trigger longer computation time. To address the limitation, we applied Relevance Vector Machine (RVM) with Bayesian inference algorithm to calculate the probability of predicted output. Similar to SVM, RVM was unable to provide transparent rules behind its classification. Therefore, this study attempts to extract rules from RVM by implementing Random Forest algorithm to the relevance vectors. We extract the average energy spectrum in each frequency band as the leading feature of emotions in EEG. Through the resulted rules of RVM_RF, we found that negative emotions of EEG were determined by the average energy spectrum of delta band at fronto-central electrodes (FCZ>23.683, FC4>24.812), theta band at frontal electrode (F5>23.683), and alpha band in parietal-occipital electrode (PO8>20.212). From the evaluation on three sessions of data measurement, it shows that the proposed approach of RVM_RF can predict the negative emotions of EEG with the higher average accuracy of 85.33% and average precision rate of 0.933 compared with other rule-based methods such as RVM_CN2 Rule, RVM_C4.5 Tree, and SVM. All in all, this proposed approach has demonstrated the possibility to identify negative emotions from EEG signal using rules extraction from sparse learning method, RVM.

show abstract

Section: Conflicts Of Interestmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rules Extraction of Relevance Vector Machine for Predicting Negative Emotions from EEG Signals

2022

IJIES

View full text Add to dashboard Cite

show abstract

“…The kernel extensions of ELM offer a sophisticated method to circumvent the estimation of hidden layer output and encode it integrally in a kernel matrix itself. Dong et al [15] proposed a new hybrid kernel function RVM which performs efficient classification at global and local feature levels. Multi task motor imagery EEG classifications are regarded as authenticated for the presented technique.…”

Section: Literature Reviewmentioning

confidence: 99%

Intelligent Machine Learning Based EEG Signal Classification Model

Ali¹,

El-Moghith²,

El-Derini³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

In recent years, Brain-Computer Interface (BCI) system gained much popularity since it aims at establishing the communication between human brain and computer. BCI systems are applied in several research areas such as neuro-rehabilitation, robots, exoeskeletons, etc. Electroencephalography (EEG) is a technique commonly applied in capturing brain signals. It is incorporated in BCI systems since it has attractive features such as noninvasive nature, high time-resolution output, mobility and cost-effective. EEG classification process is highly essential in decision making process and it incorporates different processes namely, feature extraction, feature selection, and classification. With this motivation, the current research paper presents an Intelligent Optimal Fuzzy Support Vector Machine-based EEC recognition (IOFSVM-EEG) model for BCI system. Independent Component Analysis (ICA) technique is applied onto the proposed IOFSVM-EEG model to remove the artefacts that exist in EEG signal and to retain the meaningful EEG information. Besides, Common Spatial Pattern (CSP)-based feature extraction technique is utilized to derive a helpful set of feature vectors from the preprocessed EEG signals. Moreover, OFSVM method is applied in the classification of EEG signals, in which the parameters involved in FSVM are optimally tuned using Grasshopper Optimization Algorithm (GOA). In order to validate the enhanced EEG recognition outcomes of the proposed IOFSVM-EEG model, an extensive set of experiments was conducted. The outcomes were examined under distinct aspects. The experimental results highlighted the enhanced performance of the presented IOFSVM-EEG model over other state-of-the-art methods.

show abstract

“…Furthermore, Dong et al proposed a new hybrid kernel RVM in which fused Gaussian kernel and polynomial kernel together. In MI tasks, by using the OVO-CSP strategy, phase space CSP (PSCSP) features were extracted and fed into the RVM classifier[51]. However, no relevant MK-RVM research has been found in multiple classes of MI tasks based on EEG signals and thus we use the MK-RVM classifier to classify three classes of MI tasks, and the best average accuracy can reach 83.81% (kappa: 0.76).…”

mentioning

confidence: 99%

Diverse Feature Blend Based on Filter-Bank Common Spatial Pattern and Brain Functional Connectivity for Multiple Motor Imagery Detection

Wang

Tang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Motor imagery (MI) based brain-computer interface (BCI) is a research hotspot and has attracted lots of attention. Within this research topic, multiple MI classification is a challenge due to the difficulties caused by time-varying spatial features across different individuals. To deal with this challenge, we tried to fuse brain functional connectivity (BFC) and one-versus-the-rest filter-bank common spatial pattern (OVR-FBCSP) to improve the robustness of classification. The BFC features were extracted by phase locking value (PLV), representing the brain interregional interactions relevant to the MI, whilst the OVR-FBCSP is used to extract the spatial-frequency features related to the MI. These diverse features were then fed into a multi-kernel relevance vector machine (MK-RVM). The dataset with three motor imagery tasks (left hand MI, right hand MI, and feet MI) was used to assess the proposed method. Experimental results not only showed that the cascade structure of diverse feature fusion and MK-RVM achieved satisfactory classification performance (average accuracy: 83.81%, average kappa: 0.76), but also demonstrated that BFC plays a supplementary role in the MI classification. Moreover, the proposed method has a potential to be integrated into multiple MI online detection owing to the advantage of strong timeefficiency of RVM. INDEX TERMS Multiple motor imagery, filter-bank common spatial pattern (FBCSP), phase locking value (PLV), brain functional connectivity (BFC), multi-kernel relevance vector machine (MK-RVM).

show abstract

A novel hybrid kernel function relevance vector machine for multi-task motor imagery EEG classification

Cited by 45 publications

References 39 publications

Rules Extraction of Relevance Vector Machine for Predicting Negative Emotions from EEG Signals

Rules Extraction of Relevance Vector Machine for Predicting Negative Emotions from EEG Signals

Intelligent Machine Learning Based EEG Signal Classification Model

Diverse Feature Blend Based on Filter-Bank Common Spatial Pattern and Brain Functional Connectivity for Multiple Motor Imagery Detection

Contact Info

Product

Resources

About