3D Input Convolutional Neural Networks for P300 Signal Detection

Oralhan, Zeki

doi:10.1109/access.2020.2968360

Cited by 24 publications

(20 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, the 50 ms data prestimulus is used as the baseline for correction to reduce the slow potential drift and other artifacts. Finally, the data are downsampled to 40 Hz to obtain 28-dimensional feature vector from each channel [39], [40]. This paper adopts 14 channels as the feature channels to yield a 392-dimensional feature.…”

Section: Feature Extraction and Classificationmentioning

confidence: 99%

The Effect of SOA on An Asynchronous ERP and VEP-Based BCI

Yang

Liu

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Our previous study established an asynchronous BCI system by using the oddball paradigm to simultaneously induce event-related potentials (ERPs) and visual evoked potentials (VEPs) (E-V BCIs). We found that stimulus onset asynchrony (SOA) is an important factor for performance since it significantly affects the ERP and VEP. Increasing the SOA increases the ERP, which improves the accuracy of detecting target stimuli. However, a larger SOA leads to a lower VEP frequency, which causes the VEP to have poor accuracy when discriminating between the brain states. How to balance the two potentials and accuracies is a problem. This study established eight SOAs from 100 ms to 375 ms that were composed of different interstimulus intervals and the same stimulus duration of 80 ms. We used a probability-based Fisher linear discriminant analysis (P-FLDA) classifier to calculate the classification accuracies of the ERP-based visual speller, VEP-based brain state discrimination, and E-V BCI. The results show that as the SOA increases, the amplitudes of N200 and P300 increase, and the accuracy also shows an increasing trend. However, the frequency of the VEP and the accuracies of state discrimination show downward trends. The change in accuracies of the E-V BCI system combining these two parts is nonlinear, and the SOA optimal value is 125 ms. The SOA of 125 ms yields the best accuracies of 95.83% and practical bit rate of 57.17 bits/min in the E-V BCI system, which provides a guideline for selecting the SOA to improve the performance.

show abstract

Section: Feature Extraction and Classificationmentioning

confidence: 99%

The Effect of SOA on An Asynchronous ERP and VEP-Based BCI

Yang

Liu

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In particular, the Late Positive Component (LPC), also mentioned as Late Positive Potential (LPP), is a positive-going ERP component that is an instantiation of the domain-general P300 component. [13]- [15]. Investigations consider that LPP reflects the syntactic [16] and semantic information processing [17], [18].…”

Section: Loss Of Visual Input Leads To Significant Morphological and mentioning

confidence: 99%

Perspective-Taking in Blindness: An Event- Related Brain Potentials Study With the Continuous Wavelet Transform

et al. 2020

View full text Add to dashboard Cite

This study presents brain electrophysiological characteristics of perspective-taking for blind individuals ver. sighted ones, aiming to extend our knowledge on social concept formation with visual sensory modality loss. A blind and a sighted group instructed to consider the moral dilemmas therein from a neutral (observer), a morally positive or a negative stance, after the auditory presentation of Aesop's fables. We recorded the basic emotions evoked by every stance in each fable, and the late positive potentials (LPP) component of Event-Related Potential detected with Continuous Wavelet Transform, believed to reflect the sustained attention and stimuli encoding processes. Blind participants demonstrated: i) significantly attenuated LPP amplitudes at distributed brain areas, particularly at leads T3, P4, O2, T6, Cz, Pz, CPz and CP4 (p values 0.026-0.003), ii) prolonged LPP latencies at parietal areas, at leads CP4 (p=0.009), Pz (p=0.019), and P4 (p=0.023), iii) shorter latencies at the frontal area (lead FPz, p=0.026) and iv) significantly shorter latencies at the right posterior centrotemporal area under the negative than the positive condition (lead C4, p=0.009). Both groups exhibited higher LPP amplitudes under the morally positive, and lower under the negative condition located at the parietal areas (P3, p=0.001 and CPz, p=0.013). The relationships between behavior and LPP activation conform to the proposed generation and modulation of LPP. Our findings suggest that the visually deprived cortex may become more responsive to processes associated with perspective-taking, possibly through compensatory plasticity enabling cross-modal reorganization. Thus, the study introduces a novel model for accessing the sustained attention and stimulus-encoding processes underlying perspective-taking. INDEX TERMS Event-related potentials, blindness, late positive potentials, continuous wavelet transform, perspective-taking.

show abstract

“…where i is the index of row or column range in [1,12], y i represents the probability that the signal is P300 while the ith row or column in the matrix is intensificated, i x and i y represent the row and column index with most likely P300 signals. The target character is the one at the intersection of i x -th row and i y -th column.…”

Section: Experiments Of P300-vib-netmentioning

confidence: 99%

“…Lots of approaches were proposed for P300 detection [ 6 , 7 , 8 , 9 ]. Recently, the machine learning based methods achieved excellent performance on P300 detection [ 10 , 11 , 12 ]. For the traditional machine learning methods, feature extraction and classification are two of the key techniques.…”

Section: Introductionmentioning

confidence: 99%

Novel Convolutional Neural Network with Variational Information Bottleneck for P300 Detection

Liao

Xu²,

2020

Entropy

View full text Add to dashboard Cite

In the area of brain-computer interfaces (BCI), the detection of P300 is a very important technique and has a lot of applications. Although this problem has been studied for decades, it is still a tough problem in electroencephalography (EEG) signal processing owing to its high dimension features and low signal-to-noise ratio (SNR). Recently, neural networks, like conventional neural networks (CNN), has shown excellent performance on many applications. However, standard convolutional neural networks suffer from performance degradation on dealing with noisy data or data with too many redundant information. In this paper, we proposed a novel convolutional neural network with variational information bottleneck for P300 detection. Wiht the CNN architecture and information bottleneck, the proposed network termed P300-VIB-Net could remove the redundant information in data effectively. The experimental results on BCI competition data sets show that P300-VIB-Net achieves cutting-edge character recognition performance. Furthermore, the proposed model is capable of restricting the flow of irrelevant information adaptively in the network from perspective of information theory. The experimental results show that P300-VIB-Net is a promising tool for P300 detection.

show abstract

3D Input Convolutional Neural Networks for P300 Signal Detection

Cited by 24 publications

References 47 publications

The Effect of SOA on An Asynchronous ERP and VEP-Based BCI

The Effect of SOA on An Asynchronous ERP and VEP-Based BCI

Perspective-Taking in Blindness: An Event- Related Brain Potentials Study With the Continuous Wavelet Transform

Novel Convolutional Neural Network with Variational Information Bottleneck for P300 Detection

Contact Info

Product

Resources

About