Jessica Cantillo-Negrete

An approach to improve the performance of subject-independent BCIs-based on motor imagery allocating subjects by gender

BackgroundOne of the difficulties for the implementation of Brain-Computer Interface (BCI) systems for motor impaired patients is the time consumed in the system design process, since patients do not have the adequate physical nor psychological conditions to complete the process. For this reason most of BCIs are designed in a subject-dependent approach using data of healthy subjects. The developing of subject-independent systems is an option to decrease the required training sessions to design a BCI with patient functionality. This paper presents a proof-of-concept study to evaluate subject-independent system based on hand motor imagery taking gender into account.MethodsSubject-Independent BCIs are proposed using Common Spatial Patterns and log variance features of two groups of healthy subjects; one of the groups was composed by people of male gender and the other one by people of female gender. The performance of the developed gender-specific BCI designs was evaluated with respect to a subject-independent BCI designed without taking gender into account, and afterwards its performance was evaluated with data of two healthy subjects that were not included in the initial sample. As an additional test to probe the potential use for subcortical stroke patients we applied the methodology to two patients with right hand weakness. T-test was employed to determine the significance of the difference between traditional approach and the proposed gender-specific approach.ResultsFor most of the tested conditions, the gender-specific BCIs have a statistically significant better performance than those that did not take gender into account. It was also observed that with a BCI designed with log-variance features in the alpha and beta band of healthy subjects’ data, it was possible to classify hand motor imagery of subcortical stroke patients above the practical level of chance.ConclusionsA larger subjects’ sample test may be necessary to improve the performances of the gender-specific BCIs and to further test this methodology on different patients. The reduction of complexity in the implementation of BCI systems could bring these systems closer to applications such as controlling devices for the motor rehabilitation of stroke patients, and therefore, contribute to a more effective neurological rehabilitation.

Classification of motor imagery electroencephalography signals using spiking neurons with different input encoding strategies

Motor imagery-based brain–computer interfaces decode users’ intentions from the electroencephalogram; however, poor spatial resolution makes automatic recognition of these intentions a challenging task. New classification approaches with low computational costs and high classification performances need to be developed in order to increase the number of users benefitted by these systems. On the other hand, spiking neuron models, which are mathematical abstractions of real neurons, have shown good performances in several classification tasks, making these models suitable for motor imagery classification. In this work, two different encoding strategies for spiking neuron models, applied to the classification of motor imagery time–frequency features of stroke patients and healthy subjects, were evaluated. Classification performances and computational costs of spiking neuron models were compared against those of linear discriminant analysis, support vector machines and artificial neural networks. Results showed that a time-varying encoding strategy is more suitable for motor imagery classification, and its implementation computational cost is low. Therefore, a spiking neuron model with a time-varying encoding strategy could increase the number of potential users of brain–computer interfaces.

Module to present and identify motor imagery tasks in electroencephalography

This work presents a module that aims to facilitate the acquisition of motor imagery tasks in electroencephalography (EEG) research. The device components are: a microcontroller which sets the time intervals of the events of a Graz type paradigm, and sends markers to an EEG acquisition system; a software that presents the visual and auditory clues of the paradigm in a personal computer (PC) for the test subject to perform the motor imagery tasks; and an algorithm aimed to extract the EEG information related to the motor imagery tasks. In the module validation, a delay of 1 ± 0.5 ms between the time in which the microcontroller marks the event in the EEG amplifier and the time in which the event is showed to the subject in the computers monitor was measured. Furthermore, an average difference of 167 μs was obtained between the time intervals theoretically set for every event and the time intervals obtained. The module was tested in the EEG acquisition of motor imagery tasks in a BCI research protocol involving thirty healthy subjects. The module was successful at presenting the paradigm in all the trials and in indentifying the events of each trial in the signals that were recorded.

Time-frequency analysis of EEG signals from healthy subjects allocated by gender for a subject-independent BCI-based on motor imagery

Most of the recent brain-computer interfaces (BCI) based on motor imagery are designed by taking into account a single user. A BCI system designed for multiple users without the need of extensive training sessions could be a viable solution for the systems implementation outside research centers. The present work explores the design of an independent-subject BCI. In order to accomplish this, Linear Discriminant Analysis classifiers were designed with the data of a sample of 30 healthy volunteers as a whole group and separated by gender. Three different methods were employed to compute power spectrum features from the volunteers electroencephalographic recordings. The results show that it is possible to design an independent-subject BCI for the classification of right or left hand motor imagery with respect of a reference interval with classification accuracies above 70%. The female gender could benefit more from a subject-independent classifier, than the male gender.

Spiking Neural Networks Trained with Particle Swarm Optimization for Motor Imagery Classification

Spiking neural networks (SNN) have been successfully applied in pattern classification problems. However, their performance for solving complex problems such as electroencephalography (EEG) classification has not been widely assessed. It is necessary to consider new approaches to select relevant information and for training SNN in order to improve their accuracy when applied to complex data classification. In this paper, we present a novel channel selection and classification method based on SNN trained with Particle Swarm Optimization (PSO) for the classification of EEG signals associated to motor imagery. The proposed method was able to correctly identify the most relevant channels for different motor imagery tasks. The SNN trained with PSO achieved good classification performances for a well-studied public database using a minimal number of EEG channels, showing advantages against other approaches, regarding both performance and system requirements.

Control signal for a mechatronic hand orthosis aimed for neurorehabilitation

Individuals with stroke and other central nervous damage, which may cause paresis, are unable to move their affected limb or the movements are inefficient and clumsy. Brain-computer interfaces coupled with robotic assistive technologies such as robotic hand orthosis have the potential to provide rehabilitation strategies that promote brain plasticity for these patients. This paper presents the design of a control signal based on EEG signal processed using common spatial patterns and linear discriminant analysis to identify hand motor imagery. The control signal is implemented on a robotic hand orthosis so that it activates when a healthy subject performs motor imagery of her/his right hand, simulating an online signal acquisition. The mechatronic orthosis performance was always as indicated by the control signal, and the systems online performance for detecting motor imagery was of nearly 80% of correct classification. The system may be improved by using other classification algorithms however results show that it is ready to be tested with motor impaired patients.

Mechanical structure prototype and control unit for an active orthosis for a human had

In this work, a structure for an active mechatronics orthosis of right hand is presented. The mechanism was designed to help to patients with neuromuscular injuries who cannot grasp heavy objects. The Integration Definition for Function Modeling provides activities, functional requirements and anthropomorphic parameters to define the materials and components. Currently, the first prototype allows holding objects up 500 g.