Arao Funase

Spectrum based feature extraction using spectrum intensity ratio for SSVEP detection

Recent years, a Steady-State Visual Evoked Potential (SSVEP) is used as a basis for Brain Computer Interface (BCI)[1]. Various feature extraction and classification techniques are proposed to achieve BCI based on SSVEP. The feature extraction of SSVEP is developed in the frequency domain regardless of the limitation in flickering frequency of visual stimulus caused by hardware architecture. We introduce here the feature extraction using a spectrum intensity ratio. Results show that the detection ratio reaches 84% by using a spectrum intensity ratio with unsupervised classification. It also indicates the SSVEP is enhanced by proposed feature extraction with second harmonic.

Feature extraction of EEG spectrum for Steady-State Visual Evoked Potentials detection

Recent years, a Steady-State Visual Evoked Potential (SSVEP) is adopted as a basis for Brain Computer Interface (BCI)[1]. Various feature extractions and classification techniques are proposed to achieve BCI based on SSVEP. The feature extraction of SSVEP is developed in the frequency domain regardless of the limitation in hardware architecture, i.e. a low power and simple calculation. We introduce here the feature extraction using a spectrum intensity ratio. Results show that the detection ratio reaches 90% by using a spectrum intensity ratio with unsupervised classification. It also indicates the SSVEP is enhanced by proposed feature extraction with second harmonic.

SSVEP by checkerboard related to grid size and board size

The steady-state visual evoked potential (SSVEP) is used for input signals of brain computer interfaces (BCIs). There are two types of stimulus for SSVEP. One is flushing visual stimuli and the other is flipping checker board patterns. We have been studying SSVEPs with checker board patterns. There are few studies described relationship between SSVEP and property of flipping checkerboard patterns. In this study, We pay my attention to the size of the grid squares in the checkerboard and the length on the checker board pattern.

The saccadic EEG analysis in slow cortical potential before the eye movement

The saccade-related electroencephalogram (saccadic EEG) has researched to achieve a brain computer interface (BCI) based on intention to eye movement. The spike signal before eye movements has been confirmed as the characteristic of saccadic EEG. However, the detection ratio of spike signal is not enough to achieve the BCI. In this paper, we focus on a slow cortical potential before a saccadic eye movement. Results show the possibility that the slow cortical potential is related to the bereitschaftspotential (BP) of saccadic eye movement.

Study on analysis of movement-related cortical potentials included in saccade-related EEG

Attention is currently being paid to the saccade-related electroencephalogram (saccadic EEG) signal to investigate the brain function in saccadic eye movement. The saccadic EEG has researched to achieve a brain computer interface (BCI) based on intention to eye movement. The spike EEG just before eye movements has been confirmed as the characteristic of saccadic EEG. However, other features have not been reported. In this paper, we focus on a slow cortical potential before a saccadic eye movement. Results show the possibility that the slow cortical potential is related to the bereitschafts potential (BP) of saccadic eye movement.

Simple SSVEP detection using spectrum intensity ratio and threshold

In recent years, Steady-State Visual Evoked Potential (SSVEP) is often used as a basis for Brain Computer Interface (BCI) [1]. The feature extraction is significant problem to achieve the SSVEP based BCI. Various signal processing and classification techniques are proposed to extract SSVEP from Electroencephalograph (EEG). We introduced a spectrum intensity ratio as a simple characterization and separation of SSVEP. However, it is difficult to classify an unseeing state of subjects. In addition, the comparison with conventional method is not tried yet. In this paper, we adopt a classification using threshold to reject the unseeing state.

Spectrum Intensity Ratio and Thresholding Based SSVEP Detection

Brain Computer Interface BCI is a powerful tool to control a computer or machine without body movement. There has been great interest in using Steady-State Visual Evoked Potential SSVEP for BCI [1]. Various signal processing and classification techniques are proposed to extract SSVEP from Electroencephalograph EEG. The feature extraction of SSVEP is developed in the frequency domain regardless of the limitation in hardware architecture, i.e. a low power and simple calculation. We introduced a spectrum intensity ratio as a simple characterization and separation of SSVEP. However, it is difficult to classify an unseeing state of subjects. In addition, we only tried the wide band flickering frequency as visual stimuli. In this paper, we adopt a classification using a simple calculation with threshold to detect the unseeing state from SSVEP in a narrow frequency band.

Non-linear filter based outer product expansion with reference signal for EEG analysis

This paper addresses a saccade-related electrooculogram (EOG) reduction using outer product expansion with nonlinear filter. The saccade-related electroencephalogram (EEG) signal produced by the saccadic eye movement is adopted to analyze relationship between a brain function and a human activity. Eye movement origin EOG artifacts denoising is important task to analyze the relationship between the saccade and the EEG. The tensor product expansion with absolute error (TPE-AE), which calculates two terms of outer product using reference signal, was proposed to reduce EOG artifacts. However, this TPE-AE has a significant problem corresponding to a calculation cost. In this paper, we propose and apply the median based outer product expansion with reference signal to estimate EOG component accurately. Results show that the proposed TPE-AE is effective to separate the EOG component and other noises.

Research on relationship between saccade-related EEG signals and selection of electrode position by independent component analysis

Our goal is to develop a novel BCI based on an eye movements system employing EEG signals on-line. Most of the analysis on EEG signals has been performed using ensemble averaging approaches. However, in signal processing methods for BCI, raw EEG signals are analyzed. n nIn order to process raw EEG signals, we used independent component analysis(ICA). n nPrevious paper presented extraction rate of saccade-related EEG signals by five ICA algorithms and eight window size. n nHowever, three ICA algorithms, the FastICA, the NG-FICA and the JADE algorithms, are based on 4th order statistic and AMUSE algorithm has an improved algorithm named the SOBI. Therefore, we must reselect ICA algorithms. n nIn this paper, Firstly, we add new algorithms; the SOBI and the MILCA. Using the Fast ICA, the JADE, the AMUSE, the SOBI, and the MILCA. The SOBI is an improved algorithm based on the AMUSE and uses at least two covariance matrices at different time steps. The MILCA use the independency based on mutual information. We extract saccade-related EEG signals and check extracting rates. n nSecondly, we check relationship between window sizes of EEG signals to be analyzed and extracting rates. n nThirdly, we researched on relationship between Saccade-related EEG signals and selection of electrode position by ICA. In order to develop the BCI, it is important to use a few electrode. In previous studies, we analyzed EEG signals using by 19 electrodes. In this study, we checked various combination of electrode.

Suitable ICA Algorithm for Extracting Saccade-Related EEG Signals

Our goal is to develop a novel BCI based on an eye movements system employing EEG signals on-line. Most of the analysis on EEG signals has been performed using ensemble averaging approaches. However,It is suitable to analyze raw EEG signals in signal processing methods for BCI. n nIn order to process raw EEG signals, we used independent component analysis(ICA). However, we do not know which ICA algorithms have good performance. It is important to check which ICA algorithms have good performance to develop BCIs. Previous paper presented extraction rate of saccade-related EEG signals by five ICA algorithms and eight window size. n nHowever, three ICA algorithms, the FastICA, the NG-FICA and the JADE algorithms, are based on 4th order statistic and AMUSE algorithm has an improved algorithm named SOBI.Therefore, we must re-select ICA algorithms. n nIn this paper, we add new algorithms; the SOBI and the MILCA. The SOBI is an improved algorithm based on the AMUSE and uses at least two covariance matrices at different time steps. The MILCA use the independency based on mutual information. Using the Fast ICA, the JADE, the AMUSE, the SOBI, and the MILCA, we extract saccade-related EEG signals and check extracting rates. n nSecondly, in order to get more robustness against EOG noise, we use improved FastICA with reference signals and check extracting rates.