Deep Learning With Convolutional Neural Networks for Motor Brain-Computer Interfaces Based on Stereo-Electroencephalography (SEEG)

Wu, Xiaolong; Jiang, Shize; Li, Guangye; Liu, Shengjie; Metcalfe, Benjamin; Chen, Liang; Zhang, Dingguo

doi:10.1109/jbhi.2023.3242262

Cited by 6 publications

(5 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As another transformer-based GAN study showed that the transformer alone led to obvious highfrequency artefacts in the generated data [75], an extra filter layer was added to the transformer model to further regulate the generated data in the spectral domain. On the other hand, it has been demonstrated that a CNN can act as a filter in EEG decoding tasks [73,80]. This has been validated with an ablation study by checking the spectral contents of data generated with and without the filter layer.…”

Section: The Extra Filter Layermentioning

confidence: 86%

“…Therefore, before conducting the experiment using the proposed and baseline methods, a sliding-window strategy was adopted to augment the SEEG data as a starting point [17]. The window and sliding step were, respectively, set to 500 ms and 100 ms (sampling rate of 1000 Hz) [80]. This windowing process split the original trials (10 s) into multiple shorter subtrials (500 ms).…”

Section: Da Model Training Processmentioning

confidence: 99%

See 1 more Smart Citation

Data augmentation for invasive brain–computer interfaces based on stereo-electroencephalography (SEEG)

Wu,

Zhang,

et al. 2024

J. Neural Eng.

Self Cite

View full text Add to dashboard Cite

Abstract—Objective: : Deep learning is increasingly used for Brain-computer interfaces (BCIs). However, the available brain data is sparse, especially for invasive BCIs, which can dramatically deteriorate deep learning performance. Data augmentation methods (DA), such as generative models, can help to address this issue. However, existing studies on brain signals relied on convolutional neural networks (CNNs) and ignored the temporal dependence. This paper tried to enhance the generative model by capturing the temporal relationship from a time-series perspective.Methods: A conditional generative network (cTGAN) based on the transformer model is proposed, and tested on the Stereo- electroencephalography (SEEG) data which was recorded from eight epileptic patients performing five different movements. Three other commonly-used DA methods were also implemented: noise injection (NI), variational autoencoder (VAE) and conditional Wasserstein GAN with gradient penalty (cWGANGP). Artificial SEEG data was generated, and various metrics were used to compare the data quality, including visual inspection, Cosine similarity (CS), Jensen-Shannon distance (JSD) and the effect on the performance of a deep learning-based classifier.Result: Both the proposed cTGAN and the cWGANGP methods were able to generate realistic data, while NI and VAE output inferior samples when visualised as raw sequences and in a lower dimensional space. The cTGAN generated the best samples in terms of cosine similarity and Jensen-Shannon distance and outperformed cWGANGP significantly in enhancing the performance of a deep learning-based classifier (each of them yielding a significant improvement of 6% and 3.4%, respectively).Conclusion: This paper demonstrated that a generative model that preserves temporal dependence is superior in data generation and boosting deep learning performance for SEEG signals.Significance: This is the first time that DA methods are applied to invasive BCIs based on SEEG. In addition, this study demonstrated the advantages of the model that preserves the temporal dependence from a time-series perspective.

show abstract

Section: The Extra Filter Layermentioning

confidence: 86%

Section: Da Model Training Processmentioning

confidence: 99%

Data augmentation for invasive brain–computer interfaces based on stereo-electroencephalography (SEEG)

Wu,

Zhang,

et al. 2024

J. Neural Eng.

Self Cite

View full text Add to dashboard Cite

show abstract

“…This process is repeated until the desired number of channels is selected. To extract features during the computation of the MI, we extract spectral features in 0.5-4 Hz, 4-8 Hz, 8-13 Hz, 13-30 Hz, 60-75 Hz, 75-95 Hz, 105-125 Hz, and 125-150 Hz, as in our previous study [35]. Note that these extracted features were only used to select channels, while movement classification was performed using raw SEEG signals of the selection channels.…”

Section: B Channel Selection Methodsmentioning

confidence: 99%

“…Then, channels showing high ERS or ERD by visual inspection during the task stage were chosen. The calculation of ERS/ERD and the procedure to sort the channels according to the electrode reactivity was described in our previous work [35].…”

Section: B Channel Selection Methodsmentioning

confidence: 99%

Channel Selection for Stereo- Electroencephalography (SEEG)-Based Invasive Brain-Computer Interfaces Using Deep Learning Methods

Wu,

Li,

Gao

et al. 2024

IEEE Trans. Neural Syst. Rehabil. Eng.

Self Cite

View full text Add to dashboard Cite

Brain-computer interfaces (BCIs) can enable direct communication with assistive devices by recording and decoding signals from the brain. To achieve high performance, many electrodes will be used, such as the recently developed invasive BCIs with channel numbers up to hundreds or even thousands. For those high-throughput BCIs, channel selection is important to reduce signal redundancy and invasiveness while maintaining decoding performance. However, such endeavour is rarely reported for invasive BCIs, especially those using deep learning methods. Two deep learning-based methods, referred to as Gumbel and STG, were proposed in this paper. They were evaluated using the Stereoelectroencephalography (SEEG) signals, and compared with three other methods, including manual selection, mutual information-based method (MI), and all channels (all channels without selection). The task is to classify the SEEG signals into five movements using channels selected by each method. When 10 channels were selected, the mean classification accuracies using Gumbel, STG (referred to as STG-10), manual selection, and MI selection were 65%, 60%, 60%, and 47%, respectively, whilst the accuracy was 59% using all channels (no selection). In addition, an investigation of the selected channels showed that Gumbel and STG have successfully identified the pre-central and post-central areas, which are closely related to motor control. Both Gumbel and STG successfully selected the informative channels in SEEG recordings while maintaining decoding accuracy. This study enables future high-throughput BCIs using deep learning methods, to identify useful channels and reduce computing and wireless transmission pressure.

show abstract

“…Deep neural networks are adept at learning hierarchical representations of features from transactional and user behavior data [52]. CNNs are effective for spatial data, such as images associated with fraud detection [53], while RNNs are valuable for capturing temporal dependencies in sequential data [54]. Transformer models, originally developed for natural language processing, have shown promise in capturing long-range dependencies and contextual information relevant to fraud detection [55,56].…”

Section: Deep Learning Approaches To Fraud Detectionmentioning

confidence: 99%

Artificial Intelligence Techniques for Fraud Detection

Lai

2023

Preprint

View full text Add to dashboard Cite

In the wake of increasing digital fraud, this paper introduces an innovative application of Artificial Intelligence (AI) in detecting fraudulent activities across finance, healthcare, and e-commerce sectors. It presents a detailed analysis of machine learning methodologies, specifically focusing on the advantages of supervised, unsupervised, and deep learning techniques. The paper addresses the challenges such as data imbalance, model interpretability, and ethical implications in AI-based fraud detection. It also discusses the necessity of high-quality datasets and advocates for the integration of traditional and advanced machine learning methods to enhance accuracy and adaptability in fraud identification. However, it acknowledges the limitations including computational demands and overfitting risks. The study underscores the importance of collaborative efforts between AI experts and industry professionals to develop ethical, efficient, and reliable AI solutions for fraud detection.

show abstract

Deep Learning With Convolutional Neural Networks for Motor Brain-Computer Interfaces Based on Stereo-Electroencephalography (SEEG)

Cited by 6 publications

References 38 publications

Data augmentation for invasive brain–computer interfaces based on stereo-electroencephalography (SEEG)

Data augmentation for invasive brain–computer interfaces based on stereo-electroencephalography (SEEG)

Channel Selection for Stereo- Electroencephalography (SEEG)-Based Invasive Brain-Computer Interfaces Using Deep Learning Methods

Artificial Intelligence Techniques for Fraud Detection

Contact Info

Product

Resources

About