Dataset Information

A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery.

ABSTRACT: This work describes a generalized method for classifying motor-related neural signals for a brain-computer interface (BCI), based on a stochastic machine learning method. The method differs from the various feature extraction and selection techniques employed in many other BCI systems. The classifier does not use extensive a-priori information, resulting in reduced reliance on highly specific domain knowledge. Instead of pre-defining features, the time-domain signal is input to a population of multi-layer perceptrons (MLPs) in order to perform a stochastic search for the best structure. The results showed that the average performance of the new algorithm outperformed other published methods using the Berlin BCI IV (2008) competition dataset and was comparable to the best results in the Berlin BCI II (2002-3) competition dataset. The new method was also applied to electroencephalography (EEG) data recorded from five subjects undertaking a hand squeeze task and demonstrated high levels of accuracy with a mean classification accuracy of 78.9% after five-fold cross-validation. Our new approach has been shown to give accurate results across different motor tasks and signal types as well as between subjects.

SUBMITTER: Nurse ES

PROVIDER: S-EPMC4482677 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery.

Nurse Ewan S ES Karoly Philippa J PJ Grayden David B DB Freestone Dean R DR

PloS one 20150626 6

This work describes a generalized method for classifying motor-related neural signals for a brain-computer interface (BCI), based on a stochastic machine learning method. The method differs from the various feature extraction and selection techniques employed in many other BCI systems. The classifier does not use extensive a-priori information, resulting in reduced reliance on highly specific domain knowledge. Instead of pre-defining features, the time-domain signal is input to a population of m ...[more]

PMID: 26114954

Similar Datasets

Project description:BACKGROUND AND PURPOSE:Stroke is a leading cause of long-term motor disability among adults. Present rehabilitative interventions are largely unsuccessful in improving the most severe cases of motor impairment, particularly in relation to hand function. Here we tested the hypothesis that patients experiencing hand plegia as a result of a single, unilateral subcortical, cortical or mixed stroke occurring at least 1 year previously, could be trained to operate a mechanical hand orthosis through a brain-computer interface (BCI). METHODS:Eight patients with chronic hand plegia resulting from stroke (residual finger extension function rated on the Medical Research Council scale=0/5) were recruited from the Stroke Neurorehabilitation Clinic, Human Cortical Physiology Section of the National Institute for Neurological Disorders and Stroke (NINDS) (n=5) and the Clinic of Neurology of the University of Tübingen (n=3). Diagnostic MRIs revealed single, unilateral subcortical, cortical or mixed lesions in all patients. A magnetoencephalography-based BCI system was used for this study. Patients participated in between 13 to 22 training sessions geared to volitionally modulate micro rhythm amplitude originating in sensorimotor areas of the cortex, which in turn raised or lowered a screen cursor in the direction of a target displayed on the screen through the BCI interface. Performance feedback was provided visually in real-time. Successful trials (in which the cursor made contact with the target) resulted in opening/closing of an orthosis attached to the paralyzed hand. RESULTS:Training resulted in successful BCI control in 6 of 8 patients. This control was associated with increased range and specificity of mu rhythm modulation as recorded from sensors overlying central ipsilesional (4 patients) or contralesional (2 patients) regions of the array. Clinical scales used to rate hand function showed no significant improvement after training. CONCLUSIONS:These results suggest that volitional control of neuromagnetic activity features recorded over central scalp regions can be achieved with BCI training after stroke, and used to control grasping actions through a mechanical hand orthosis.

Project description:Interventional therapy using brain-computer interface (BCI) technology has shown promise in facilitating motor recovery in stroke survivors; however, the impact of this form of intervention on functional networks outside of the motor network specifically is not well-understood. Here, we investigated resting-state functional connectivity (rs-FC) in stroke participants undergoing BCI therapy across stages, namely pre- and post-intervention, to identify discriminative functional changes using a machine learning classifier with the goal of categorizing participants into one of the two therapy stages. Twenty chronic stroke participants with persistent upper-extremity motor impairment received neuromodulatory training using a closed-loop neurofeedback BCI device, and rs-functional MRI (rs-fMRI) scans were collected at four time points: pre-, mid-, post-, and 1 month post-therapy. To evaluate the peak effects of this intervention, rs-FC was analyzed from two specific stages, namely pre- and post-therapy. In total, 236 seeds spanning both motor and non-motor regions of the brain were computed at each stage. A univariate feature selection was applied to reduce the number of features followed by a principal component-based data transformation used by a linear binary support vector machine (SVM) classifier to classify each participant into a therapy stage. The SVM classifier achieved a cross-validation accuracy of 92.5% using a leave-one-out method. Outside of the motor network, seeds from the fronto-parietal task control, default mode, subcortical, and visual networks emerged as important contributors to the classification. Furthermore, a higher number of functional changes were observed to be strengthening from the pre- to post-therapy stage than the ones weakening, both of which involved motor and non-motor regions of the brain. These findings may provide new evidence to support the potential clinical utility of BCI therapy as a form of stroke rehabilitation that not only benefits motor recovery but also facilitates recovery in other brain networks. Moreover, delineation of stronger and weaker changes may inform more optimal designs of BCI interventional therapy so as to facilitate strengthened and suppress weakened changes in the recovery process.

Project description:The primary goal of this work was to apply data-driven machine learning regression to assess if resting state functional connectivity (rs-FC) could estimate measures of behavioral domains in stroke subjects who completed brain-computer interface (BCI) intervention for motor rehabilitation. The study cohort consisted of 20 chronic-stage stroke subjects exhibiting persistent upper-extremity motor deficits who received the intervention using a closed-loop neurofeedback BCI device. Over the course of this intervention, resting state functional MRI scans were collected at four distinct time points: namely, pre-intervention, mid-intervention, post-intervention and 1-month after completion of intervention. Behavioral assessments were administered outside the scanner at each time-point to collect objective measures such as the Action Research Arm Test, Nine-Hole Peg Test, and Barthel Index as well as subjective measures including the Stroke Impact Scale. The present analysis focused on neuroplasticity and behavioral outcomes measured across pre-intervention, post-intervention and 1-month post-intervention to study immediate and carry-over effects. Rs-FC, changes in rs-FC within the motor network and the behavioral measures at preceding stages were used as input features and behavioral measures and associated changes at succeeding stages were used as outcomes for machine-learning-based support vector regression (SVR) models. Potential clinical confounding factors such as age, gender, lesion hemisphere, and stroke severity were included as additional features in each of the regression models. Sequential forward feature selection procedure narrowed the search for important correlates. Behavioral outcomes at preceding time-points outperformed rs-FC-based correlates. Rs-FC and changes associated with bilateral primary motor areas were found to be important correlates of across several behavioral outcomes and were stable upon inclusion of clinical variables as well. NIH Stroke Scale and motor impairment severity were the most influential clinical variables. Comparatively, linear SVR models aided in evaluation of contribution of individual correlates and seed regions while non-linear SVR models achieved higher performance in prediction of behavioral outcomes.

Project description:Brain-computer interface (BCI) paradigms are usually tested when environmental and biological artifacts are intentionally avoided. In this study, we deliberately introduced different perturbations in order to test the robustness of a steady state visual evoked potential (SSVEP) based BCI. Specifically we investigated to what extent a drop in performance is related to the degraded quality of EEG signals or rather due to increased cognitive load. In the online tasks, subjects focused on one of the four circles and gave feedback on the correctness of the classification under four conditions randomized across subjects: Control (no perturbation), Speaking (counting loudly and repeatedly from one to ten), Thinking (mentally counting repeatedly from one to ten), and Listening (listening to verbal counting from one to ten). Decision tree, Naïve Bayes and K-Nearest Neighbor classifiers were used to evaluate the classification performance using features generated by canonical correlation analysis. During the online condition, Speaking and Thinking decreased moderately the mean classification accuracy compared to Control condition whereas there was no significant difference between Listening and Control conditions across subjects. The performances were sensitive to the classification method and to the perturbation conditions. We have not observed significant artifacts in EEG during perturbations in the frequency range of interest except in theta band. Therefore we concluded that the drop in the performance is likely to have a cognitive origin. During the Listening condition relative alpha power in a broad area including central and temporal regions primarily over the left hemisphere correlated negatively with the performance thus most likely indicating active suppression of the distracting presentation of the playback. This is the first study that systematically evaluates the effects of natural artifacts (i.e. mental, verbal and audio perturbations) on SSVEP-based BCIs. The results can be used to improve individual classification performance taking into account effects of perturbations.

Dataset Information

A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery.

Publications

A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets