How many people can control a motor imagery based BCI using common spatial patterns?

Abstract

EEG based Brain-Computer Interfaces (BCIs) often use evoked potentials (P300), steady state visual evoked potentials (SSVEP) or motor imagery (MI) for control strategies. This study investigated maximum and mean accuracy of a MI based BCI using Common Spatial Patterns (CSP). Twenty healthy people participated in the study and were equipped with 64 active EEG electrodes. They performed a training paradigm with 160 trials by imagining either left or right hand movement to set up a subject specific CSP filter to spatially filter the EEG data. Following that, two real-time runs with 80 trials were performed, which provided feedback to the subject. The real-time accuracy was then calculated for every subject, and finally a grand average accuracy of 80.7% was reached for the 20 subjects. One person reached a perfect classification result of 100%, 30% performed above 90% and one was below 59%. The results show that most people can use a MI based BCI after a brief training time if CSPs with 64 active electrodes are used. The method of CSP yields clearly better classification results compared to a bandpower approach. While more electrodes are needed for classification, this is less of a disadvantage with modern active electrodes.

Complete Locked-in and Locked-in Patients: Command Following Assessment and Communication with Vibro-Tactile P300 and Motor Imagery Brain-Computer Interface Tools

“…In a group study, 20 healthy people attained a mean accuracy of 80.7% after 20 min of training (Ortner et al, 2015). Motor imagery BCIs have also been validated with LIS patients (Kübler et al, 2005; McFarland and Vaughan, 2016).…”

BCI and FES Based Therapy for Stroke Rehabilitation Using VR Facilities

“…The whole AF7 AF3 AF8 AF4 F9 F10 F7 F5 F4 F3 F1 F2 F6 F8 FT9 FT7 FC3 FC10 FC4 FC8 T9 T7 C1 T8 T10 A1 A2 TP9 TP7 CP3 CP4 CP4 P9 CP10 P7 P5 P3 P1 P2 P4 P6 P8 P10 PO7 PO3 PO4 PO8 O1 O2 FC5 FC1 FC2 FC6 C5 C3 C4 system was first tested on 3 healthy people and then some fine tunings were done based on their suggestions in order to get high accuracy and a good repeatability coefficient. All threehealthy people achieved low control error rates, comparable to the ones presented by Ortner and colleagues in [33]. Before starting the tests on patients within clinical environment, this study was approved by the institutional review board of the National Institute of Rehabilitation, Physical Medicine and Balneoclimatology from Bucharest, Romania, and each patient signed informed consent and an authorization for videos and photographs release before starting the study.…”

Assessment and Communication for People with Disorders of Consciousness

“…The sequence of left- and right-hand instructions is randomized. For signal preprocessing, the method of Common Spatial Patterns (CSP) 10 12 13 is used. This method yields a set of spatial filters designed to minimize the variance of one class while maximizing variance for the other class.…”

“…This results in four features, which are classified by the linear discriminant analysis 8 . The whole classification procedure is described in detail in a recent publication, showing a grand average classification accuracy of 80.7% after only 60 min of training in healthy users 13 . The calculation of accuracy is done via cross-validation.…”