Performance evaluation of a motor-imagery-based EEG-Brain computer interface using a combined cue with heterogeneous training data in BCI-Naive subjects

Data augmentation strategies for EEG-based motor imagery decoding

The wide use of motor imagery as a paradigm for brain-computer interfacing (BCI) points to its characteristic ability to generate discriminatory signals for communication and control. In recent times, deep learning techniques have increasingly been explored, in motor imagery decoding. While deep learning techniques are promising, a major challenge limiting their wide adoption is the amount of data available for decoding. To combat this challenge, data augmentation can be performed, to enhance decoding performance. In this study, we performed data augmentation by synthesizing motor imagery (MI) electroencephalography (EEG) trials, following six approaches. Data generated using these methods were evaluated based on four criteria, namely – the accuracy of prediction, the Frechet Inception distance (FID), the t-distributed Stochastic Neighbour Embedding (t-SNE) plots and topographic head plots. We show, based on these, that the synthesized data exhibit similar characteristics with real data, gaining up to 3% and 12% increases in mean accuracies across two public datasets. Finally, we believe these approaches should be utilized in applying deep learning techniques, as they not only have the potential to improve prediction performances, but also to save time spent on subject data collection.

On the (in)efficacy of motor imagery training without feedback and event-related desynchronizations considerations

Motor imagery (MI) constitutes a recurrent strategy for signals generation in brain-computer interfaces (BCIs) - systems that aim to control external devices by directly associating brain responses to distinct commands. Although great improvement has been achieved in MI-BCIs performance over recent years, they still suffer from inter- and intra-subject variability issues. As an attempt to cope with this, some studies have suggested that MI training should aid users to appropriately modulate their response for BCI usage: generally, this training is performed based on the sensorimotor rhythms' modulation over the primary sensorimotor cortex (PMC), with the signal being feedbacked to the user. Nonetheless, recent studies have revisited the actual involvement of the PMC into MI, and little to no attention has been devoted to understanding the participation of other cortical areas into training protocols. Therefore, in this work, our aim was to analyze the response induced by hands MI of 10 healthy subjects in the form of event-related desynchronizations (ERDs) and to assess whether features from beyond the PMC might be useful for hands MI classification. We investigated how this response occurs for distinct frequency intervals between 7-30 Hz, and ex0plored changes in their evocation pattern across 12 MI training sessions without feedback. Overall, we found that ERD patterns occur differently for the frequencies encompassed by the μ and β bands, with its evocation being favored for the first band. Over time, the no-feedback approach was inefficient to aid in enhancing ERD evocation (EO). Moreover, to some extent, EO tends to decrease over blocks within a given run, and runs within an MI session, but remains stable within an MI block. We also found that the C3/C4 pair is not necessarily optimal for data classification, and both spectral and spatial subjects' specificities should be considered when designing training protocols.

Hybrid brain-computer interface for biomedical cyber-physical system application using wireless embedded EEG systems

Background

One of the key challenges of the biomedical cyber-physical system is to combine cognitive neuroscience with the integration of physical systems to assist people with disabilities. Electroencephalography (EEG) has been explored as a non-invasive method of providing assistive technology by using brain electrical signals.

Methods

This paper presents a unique prototype of a hybrid brain computer interface (BCI) which senses a combination classification of mental task, steady state visual evoked potential (SSVEP) and eyes closed detection using only two EEG channels. In addition, a microcontroller based head-mounted battery-operated wireless EEG sensor combined with a separate embedded system is used to enhance portability, convenience and cost effectiveness. This experiment has been conducted with five healthy participants and five patients with tetraplegia.

Results

Generally, the results show comparable classification accuracies between healthy subjects and tetraplegia patients. For the offline artificial neural network classification for the target group of patients with tetraplegia, the hybrid BCI system combines three mental tasks, three SSVEP frequencies and eyes closed, with average classification accuracy at 74% and average information transfer rate (ITR) of the system of 27 bits/min. For the real-time testing of the intentional signal on patients with tetraplegia, the average success rate of detection is 70% and the speed of detection varies from 2 to 4 s.

Bayesian spatial filters for source signal extraction: a study in the peripheral nerve

The ability to extract physiological source signals to control various prosthetics offer tremendous therapeutic potential to improve the quality of life for patients suffering from motor disabilities. Regardless of the modality, recordings of physiological source signals are contaminated with noise and interference along with crosstalk between the sources. These impediments render the task of isolating potential physiological source signals for control difficult. In this paper, a novel Bayesian Source Filter for signal Extraction (BSFE) algorithm for extracting physiological source signals for control is presented. The BSFE algorithm is based on the source localization method Champagne and constructs spatial filters using Bayesian methods that simultaneously maximize the signal to noise ratio of the recovered source signal of interest while minimizing crosstalk interference between sources. When evaluated over peripheral nerve recordings obtained in vivo, the algorithm achieved the highest signal to noise interference ratio ( 7.00 ±3.45 dB) amongst the group of methodologies compared with average correlation between the extracted source signal and the original source signal R = 0.93. The results support the efficacy of the BSFE algorithm for extracting source signals from the peripheral nerve.