Global Cortical Network Distinguishes Motor Imagination of the Left and Right Foot

Phang, Chun Ren; Ko, Li-Wei

doi:10.1109/access.2020.2999133

Cited by 22 publications

(13 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The brain-wide connectivity estimated from temporal EEG signals with reduced frontoparietal connections would proceed to the classification step. A linear support vector machine (SVM) was shown to be able to classify connectivity features generated from lower limb MI with promising accuracy [28]. Hence, the machine learning classifier used in this system was linear SVM.…”

Section: Frontoparietal Connections and Classifiermentioning

confidence: 99%

“…A recent review [26] has discussed the applications of EEG-based functional connectivity and effective connectivity computed by Pearson's correlation, magnitude squared coherence, phase-locking value, transfer entropy, multivariate autoregression, directed transfer function, and partial directed coherence in the domain of upper limb MI-BCI. Moreover, given the fact that the cortices of the lower limb are anatomically close to each other [12], functional connectivity derived from EEG has shown superiority in distinguishing both upper and lower limb MIs compared to localized features such as common spatial pattern and band power [27], [28].…”

Section: Introductionmentioning

confidence: 99%

“…This study is an extension of our previous study. In [28], we successfully classified the functional connectivities of healthy individuals in 1-second bipedal motor imagery tasks with promising accuracy. Hence, in this current study, we further evaluate the distinguishability of functional connectivities during bipedal MP and MI tasks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Frontoparietal Dysconnection in Covert Bipedal Activity for Enhancing the Performance of the Motor Preparation-Based Brain–Computer Interface

Phang

Chen

Cheng

et al. 2023

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Motor-based brain-computer interfaces (BCIs) were developed from the brain signals during motor imagery (MI), motor preparation (MP), and motor execution (ME). Motorbased BCIs provide an active rehabilitation scheme for poststroke patients. However, BCI based solely on MP was rarely investigated. Since MP is the precedence phase before MI or ME, MP-BCI could potentially detect brain commands at an earlier state. This study proposes a bipedal MP-BCI system, which is actuated by the reduction in frontoparietal connectivity strength. Three substudies, including bipedal classification, neurofeedback, and post-stroke analysis, were performed to validate the performance of our proposed model. In bipedal classification, functional connectivity was extracted by Pearson's correlation model from electroencephalogram (EEG) signals recorded while the subjects were performing MP and MI. The binary classification of MP achieved short-lived peak accuracy of 73.73(±7.99)% around 200-400 ms post-cue. The peak accuracy was found synchronized to the MP-related potential and the decrement in frontoparietal connection strength. The connection strengths of the right frontal and left parietal lobes in the alpha range were found negatively correlated to the classification accuracy. In the subjective neurofeedback study, the majority of subjects reported that motor preparation instead of the motor imagery activated the frontoparietal dysconnection. Post-stroke C.-R. Phang, and L.-W. Ko are with the International Ph.D. Program in Interdisciplinary Neuroscience (UST), and Center for Intelligent Drug Systems and Smart Bio-devices (IDS

show abstract

Section: Frontoparietal Connections and Classifiermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Frontoparietal Dysconnection in Covert Bipedal Activity for Enhancing the Performance of the Motor Preparation-Based Brain–Computer Interface

Phang

Chen

Cheng

et al. 2023

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…In Shahbakhti et al [36], the authors investigated the detection and elimination of eye blinks from EEG trials but performed no classification. Other works by Phang and Ko [37] and Mwata-Velu et al [38] performed classification, though with limited number of classes. Phang and Ko [37] focused on left-and right-foot distinction using CSP, band power and Pearson's correlation-based connectivity features with traditional SOTA algorithms -SVM, LDA and KNN.…”

Section: Ii) Crops Yielded Better Performance In Shallow Network Comp...mentioning

confidence: 99%

“…Other works by Phang and Ko [37] and Mwata-Velu et al [38] performed classification, though with limited number of classes. Phang and Ko [37] focused on left-and right-foot distinction using CSP, band power and Pearson's correlation-based connectivity features with traditional SOTA algorithms -SVM, LDA and KNN. While they reported plausible results for the best-performing method (86.26 ± 9.95%), it should be noted that their result is based on a binary decoding task.…”

Section: Ii) Crops Yielded Better Performance In Shallow Network Comp...mentioning

confidence: 99%

State-of-the-Art Versus Deep Learning: A Comparative Study of Motor Imagery Decoding Techniques

et al. 2022

View full text Add to dashboard Cite

State-of-the-art techniques (SOTA) for motor imagery decoding have largely involved the use of common spatial patterns (CSP) and power spectral density (PSD), for feature extraction. Other frequency transforms, such as wavelets and empirical mode decomposition (EMD) have also been used but the aforementioned two have been the most popular. For classification, linear discriminant analysis (LDA) and support vector machines (SVM) have been mostly used. It is, however, worth investigating other approaches, such as deep learning, which offer a potential for improvement, but are not yet mainstream. Deep learning techniques based on neural networks (NNs) have been underexplored in motor imagery processing. Considering their success in other fields, which speaks to their potential for obtaining improved results over the SOTA, they should be explored for motor imagery decoding. This study takes a comparative approach in the use of deep learning as compared with the SOTA. From our findings, we infer that neural networks are suitable for motor imagery decoding and might be preferable over the SOTA. The use of specific feature extraction is also not as necessary as seen with SOTA approaches, though it might offer some gains in performance. Our results show a statistically significant improvement in decoding accuracies, up to 20% increase, with the use of NNs as compared with the SOTA. Also, we conclude that the use of crops for data augmentation might yield better results with shallow architectures as against deeper ones and that there might be other factors affecting the effectiveness of crops, needing further investigation.

show abstract

Emerging Non-invasive Brain–Computer Interface Technologies and Their Clinical Applications

Stevenson

Chang

et al. 2022

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Global Cortical Network Distinguishes Motor Imagination of the Left and Right Foot

Cited by 22 publications

References 64 publications

Frontoparietal Dysconnection in Covert Bipedal Activity for Enhancing the Performance of the Motor Preparation-Based Brain–Computer Interface

Frontoparietal Dysconnection in Covert Bipedal Activity for Enhancing the Performance of the Motor Preparation-Based Brain–Computer Interface

State-of-the-Art Versus Deep Learning: A Comparative Study of Motor Imagery Decoding Techniques

Emerging Non-invasive Brain–Computer Interface Technologies and Their Clinical Applications

Contact Info

Product

Resources

About