Alessandro B. Benevides

Robotic wheelchair commanded by SSVEP, motor imagery and word generation

This work presents a robotic wheelchair that can be commanded by a Brain Computer Interface (BCI) through Steady-State Visual Evoked Potential (SSVEP), Motor Imagery and Word Generation. When using SSVEP, a statistical test is used to extract the evoked response and a decision tree is used to discriminate the stimulus frequency, allowing volunteers to online operate the BCI, with hit rates varying from 60% to 100%, and guide a robotic wheelchair through an indoor environment. When using motor imagery and word generation, three mental task are used: imagination of left or right hand, and imagination of generation of words starting with the same random letter. Linear Discriminant Analysis is used to recognize the mental tasks, and the feature extraction uses Power Spectral Density. The choice of EEG channel and frequency uses the Kullback-Leibler symmetric divergence and a reclassification model is proposed to stabilize the classifier.

Proposal of Brain-Computer Interface architecture to command a robotic wheelchair

This paper presents a Brain-Computer Interface architecture that is being implemented in a robotic wheelchair. The interface uses electroencephalographic signals and works with three mental tasks, which are the imagination of right or left hand movements and generation of words beginning with the same random letter. This research uses a data set to perform a simulation of real-time classification, which is the pseudo-online technique, in order to have a preliminary view of the performance of the proposed BCI architecture. Linear Discriminant Analysis is used to recognize the mental tasks. The feature extraction uses the Power Spectral Density and the choice of EEG channel and frequency uses the Kullback-Leibler symmetric divergence. A reclassification model is proposed to stabilize the classifier, and the Sammon map is used to visualize the class separation.

Towards an architecture of a hybrid BCI based on SSVEP-BCI and passive-BCI.

Recent decades have seen BCI applications as a novel and promising new channel of communication, control and entertainment for disabled and healthy people. However, BCI technology can be prone to errors due to the basic emotional state of the user: the performance of reactive and active BCIs decrease when user becomes stressed or bored, for example. Passive-BCI is a recent approach that fuses BCI technology with cognitive monitoring, providing valuable information about the users intentions, the situational interpretations and mainly the emotional state. In this work, an architecture composed by passive-BCI co-working with SSVEP-BCI is proposed, with the aim of improving the performance of the reactive-BCI. The possibility of adjusting recognition characteristics of SSVEP-BCIs using a passive-BCI output is evaluated. In this sense, two ways to recover the accuracy of SSVEP are presented in this paper: 1) Adjusting of Amplitude of the SSVEP and 2) Adjusting of Frequency of the SSVEP response. The results are promising, because accuracy of SSVEP-BCI can be recovered in the case that it was reduced by the BCI users emotional state.

Comparative muscle study fatigue with sEMG signals during the isotonic and isometric tasks for diagnostics purposes

The study of fatigue is an important tool for diagnostics of disease, sports, ergonomics and robotics areas. This work deals with the analysis of sEMG most important fatigue muscle indicators with use of signal processing in isometric and isotonic tasks with the propose of standardizing fatigue protocol to select the data acquisition and processing with diagnostic proposes. As a result, the slope of the RMS, ARV and MNF indicators were successful to describe the fatigue behavior expected. Whereas that, MDF and AIF indicators failed in the description of fatigue. Similarly, the use of a constant load for sEMG data acquisition was the best strategy in both tasks.

Adaptive BCI based on software agents

The selection of features is generally the most difficult field to model in BCIs. Therefore, time and effort are invested in individual feature selection prior to data set training. Another great difficulty regarding the model of the BCI topology is the brain signal variability between users. How should this topology be in order to implement a system that can be used by large number of users with an optimal set of features? The proposal presented in this paper allows for obtaining feature reduction and classifier selection based on software agents. The software agents contain Genetic Algorithms (GA) and a cost function. GA used entropy and mutual information to choose the number of features. For the classifier selection a cost function was defined. Success rate and Cohens Kappa coefficient are used as parameters to evaluate the classifiers performance. The obtained results allow finding a topology represented as a neural model for an adaptive BCI, where the number of the channels, features and the classifier are interrelated. The minimal subset of features and the optimal classifier were obtained with the adaptive BCI. Only three EEG channels were needed to obtain a success rate of 93% for the BCI competition III data set IVa.

An Ethernet sniffer for On-line acquisition of EEG with the BrainNet36® device applied to a BCI

This paper presents an Ethernet sniffer programmed in ANSI C to export in an on-line way the electroencephalographic (EEG) data acquired with the device BrainNet36®, which is a Brazilian EEG device for clinical and polysomnography purposes. The sniffer proved to be useful for off-line analysis of the EEG and also for on-line applications, as Brain Computer-Interfaces (BCIs). Despite its limitations and packet losses at around 1.7% due to noise, the off-line analysis of the EEG successfully replicated results of the literature regarding the Event-Related (De)Synchronization (ERD/ERS) and Evoked Potentials (EPs) calculated for mental tasks. For on-line applications the sniffer was used to program a single-switch BCI for on-line classification of motor and no motor mental tasks with high success rate. Nowadays, the BrainNet36® device is being used for EEG research at many Brazilian universities, therefore, we hope that this article may encourage on-line applications. Finally, as the sniffer operation is explained here with examples, this text may serve as a reference guide for potential users.

Using Brain-Computer Interface to control an avatar in a Virtual Reality Environment

The proposal of this research is to present the development of a tool that might be useful in rehabilitation, for subjects with disability, that suffer from some kind of limbs movement limitation. This tool carries a 3D Virtual Reality Environment (VRE), which emulates the movement of a healthy person, using the immersion of the subject through an avatar. To do so, and test its feasibility, pre acquired motor imagery signals were used to test the VRE as an off-line Brain Computer Interface (BCI) feedback. The subjects brain waves were captured by an Electroencephalography (EEG) equipment. For training the classifier, 45 trials, 25 seconds long, were used, and 15 trials for its validation. Five mental tasks were tested with the BCI, and the one with the best results (imagination of the manipulation of a cube) was used to move the avatar through a virtual room.

Evaluation of ERD/ERS caused by unpleasant sounds to be applied in BCIs

This paper presents the off-line evaluation of scalp EEG signals registered while a potential brain-computer-interface user was hearing unpleasant sounds. The EEG patterns of event-related desynchronization/synchronization are studied during the human perception of sound stimuli, which was given by scraping a sharp knife along the surface of a ridged metal bottle sound. This evaluation was conducted with the purpose of search the correlating EEG signals with emotional state caused by unpleasant stimuli and their influence of brain-computer-interface commands, as motor imagery commands, in order to increase the efficiency of this kind of human-machine-interface performance. It was observed that waveforms in μ and ß frequency bands of Fpl and Fp2 channels present variations that could been caused by neural activity due to the unpleasant sound stimuli.

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.

A pseudo-online Brain-Computer Interface with automatic choice for EEG channel and frequency

This paper presents the classification of three mental tasks, using the electroencephalographic signal and simulating a real-time process, that is, the pseudo-online technique. Linear Discriminant Analysis is used to recognize the mental tasks, and the feature extraction uses the Power Spectral Density. The choice of EEG channel and frequency uses the Kullback-Leibler symmetric divergence and a reclassification model is proposed to stabilize the classifier. Finally, it is expected that the proposed method can be implemented in a Brain-Computer Interface associated with a robotic wheelchair.