Richard M. G. Tello

A comparison of techniques and technologies for SSVEP classification

This paper presents the evaluation of seven techniques of feature extraction (PSD, F-Test, EMD, MCE, CCA, LASSO and MSI) for gaze-target detections in a SSVEP-based BCI. Two type of technologies for visual stimulation were used (LCD and LEDs). Five differents windows lengths (1, 2, 4, 5 and 10 s) were used and seven volunteers participated in this study. The highest accuracy obtained in all cases was 93.57% using LEDs and the highest ITR was 36.90 bits/min for LCD. The technique based on MSI shows the highest success rate in both cases (LCD or LED) and is even more noticeable when the window size is increased.

Comparison of the influence of stimuli color on Steady-State Visual Evoked Potentials

IntroductionThe main idea of a traditional Steady State Visually Evoked Potentials (SSVEP)-BCI is the activation of commands through gaze control. For this purpose, the retina of the eye is excited by a stimulus at a certain frequency. Several studies have shown effects related to different kind of stimuli, frequencies, window lengths, techniques of feature extraction and even classification. So far, none of the previous studies has performed a comparison of performance of stimuli colors through LED technology. This study addresses precisely this important aspect and would be a great contribution to the topic of SSVEP-BCIs. Additionally, the performance of different colors at different frequencies and the visual comfort were evaluated in each case.MethodsLEDs of four different colors (red, green, blue and yellow) flickering at four distinct frequencies (8, 11, 13 and 15 Hz) were used. Twenty subjects were distributed in two groups performing different protocols. Multivariate Synchronization Index (MSI) was the technique adopted as feature extractor.ResultsThe accuracy was gradually enhanced with the increase of the time window. From our observations, the red color provides, in most frequencies, both highest rates of accuracy and Information Transfer Rate (ITR) for detection of SSVEP.ConclusionAlthough the red color has presented higher ITR, this color was turned in the less comfortable one and can even elicit epileptic responses according to the literature. For this reason, the green color is suggested as the best choice according to the proposed rules. In addition, this color has shown to be safe and accurate for an SSVEP-BCI.

Towards semg classification based on Bayesian and k-NN to control a prosthetic hand

This paper presents the classification of motor tasks, using surface electromyography (sEMG) to control a prosthetic hand for rehabilitation of amputees. Two types of classifiers are compared: k-Nearest Neighbor (k-NN) and Bayesian (Discriminant Analysis). Motor tasks are divided into four groups correlated. The volunteers were healthy people (without amputation) and several analyzes of each of the signals were conducted. The online simulations use the sliding window technique and for feature extraction RMS (Root Mean Square), VAR (Variance) and WL (Waveform Length) values were used. A model is proposed for reclassification using cross-validation in order to validate the classification, and a visualization in Sammon Maps is provided in order to observe the separation of the classes for each set of motor tasks. Finally, the proposed method can be implemented in a computer interface providing a visual feedback through a artificial prosthetic developed in Visual C++ and MATLAB commands.

Comparison of new techniques based on EMD for control of a SSVEP-BCI

This paper presents the comparation of three different feature extraction techniques based on the Empirical Mode Decomposition (EMD) for a SSVEP-BCI. This approach based on the characterization of the signal by EMD, is proposed as a novel alternative to other techniques and it was demonstrated that it exceeds both in accuracy rate and Information Transfer Rate (ITR). The experiments were performed in an offline way, and seven volunteers participated of the study. The stimulis were generated both by LCD and LEDs. The frequencies used were 8, 11, 13 and 15 Hz. The results here reported such represent the average of the seven participants, achieving a success rate of 81% and ITR of 23.32 bits/min of the total set of cases analyzed. It is further confirmed that the highest success rates and ITRs were obtained for stimulation by LEDs.

Comparison Between Wire and Wireless EEG Acquisition Systems Based on SSVEP in an Independent-BCI

This paper presents a comparison between two different technologies of acquisition systems (BrainNet36 and Emotiv Epoc) for an Independent-BCI based on Steady-State Visual Evoked Potential (SSVEP). Two stimuli separated by a viewing angle <; 1° were used. Multivariate Synchronization Index (MSI) technique was used as feature extractor and five subjects participated in the experiments. The class is obtained through a criterion of maxima. The left and right flicker stimuli were modulated at frequencies of 8.0 and 13.0 Hz, respectively. Acquisition via BrainNet system showed better results, obtaining the highest value for accuracy (100%) and the highest ITR (35.18 bits/min). This Independent-BCI is based on covert attention.

A new approach for SSVEP detection using PARAFAC and canonical correlation analysis.

This paper presents a new way for automatic detection of SSVEPs through correlation analysis between tensor models. 3-way EEG tensor of channel × frequency × time is decomposed into constituting factor matrices using PARAFAC model. PARAFAC analysis of EEG tensor enables us to decompose multichannel EEG into constituting temporal, spectral and spatial signatures. SSVEPs characterized with localized spectral and spatial signatures are then detected exploiting a correlation analysis between extracted signatures of the EEG tensor and the corresponding simulated signatures of all target SSVEP signals. The SSVEP that has the highest correlation is selected as the intended target. Two flickers blinking at 8 and 13 Hz were used as visual stimuli and the detection was performed based on data packets of 1 second without overlapping. Five subjects participated in the experiments and the highest classification rate of 83.34% was achieved, leading to the Information Transfer Rate (ITR) of 21.01 bits/min.

Capture protocol of forearm sEMG signals with four channels in healthy and amputee people

One of the biggest concerns for control of robotic devices for rehabilitation of amputees is related to the sEMG signal quality. Depending on how clean is the signal, more efficient and effective is the response of the control in relation to the user needs. This article proposes a protocol for sEMG signal capture with the least amount of crosstalk without the use of filters on the forearm with four channels in healthy and amputee people. Extensor digitorum muscle (Channel 1), flexor digitorum superficilis (Channel 2), flexor carpi ulnaris (Channel 3) and flexor pollicislongus (Channel 4) were used. This protocol is used in a series of ten isometric motor tasks related to the movement of the fingers and wrist on the hand. The attenuates 90% of the noise generated by the grid at 59.97 to 60.05 Hz and their harmonics, together with identified unusual noise frequency of 258.1 Hz which was isolated in 75% for the four channels during all motor tasks. This allows the recognition of motor defects with the use of the signal obtained without the use of filters, allowing a lower computational overhead for processing the signal to control a myoelectric hand prosthesis.

An Evaluation of Performance for an Independent SSVEP-BCI Based on Compressive Sensing System

This paper aims a comparison of performance on classification between two feature extractors after of a novel process of compression and reconstruction of signals proposed. In this way, Multivariate Synchronization Index (MSI) and Canonical Correlation Analysis (CCA) techniques of feature extraction were used to get the frequency used from visual stimuli. This system is based on detection of visual attention in SSVEP-BCIs for people with severe motor disabilities. Five male subjects (29.8 ± 2.17 years) participated in this study. According to the results, the MSI technique showed better results in terms of accuracy compared to CCA. It was demon- strated that the proposed system based in MSI technique can offer acceptable performance for a high compression ratio compared to CCA technique. Consequently, the power- consumption in wireless systems can be significantly reduced.

Feature extraction and classification of sEMG signals applied to a virtual hand prosthesis

This paper presents the classification of motor tasks, using surface electromyography (sEMG) to control a virtual prosthetic hand for rehabilitation of amputees. Two types of classifiers are compared: k-Nearest Neighbor (k-NN) and Bayesian (Discriminant Analysis). Motor tasks are divided into four groups correlated. The volunteers were people without amputation and several analyzes of each of the signals were conducted. The online simulations use the sliding window technique and for feature extraction RMS (Root Mean Square), VAR (Variance) and WL (Waveform Length) values were used. A model is proposed for reclassification using cross-validation in order to validate the classification, and a visualization in Sammon Maps is provided in order to observe the separation of the classes for each set of motor tasks. Finally, the proposed method can be implemented in a computer interface providing a visual feedback through an virtual hand prosthetic developed in Visual C++ and MATLAB commands.

Real-time control of finger and wrist movements in a virtual hand using traditional features of semg and Bayesian classifier

In this study, we present a real-time system to control a virtual hand using traditional features of surface electromyography (sEMG). The sEMG signal was recorded while performing simple finger and wrist movements related to the day-to-day activities. Traditional features of sEMG: RMS (Root Mean Square), VAR (Variance) and WL (Waveform Length) were computed using the sliding window technique. These features were classified using two types of classifiers: k-Nearest Neighbor (k-NN) and Bayesian (Discriminant Analysis). These classified patterns were used to control the designed virtual hand. This proposed system for controlling virtual hand can provide a better training and visual feedback to people with disability and for amputees.