Hybrid P300-Based Brain-Computer Interface to Improve Usability for People With Severe Motor Disability: Electromyographic Signals for Error Correction During a Spelling Task

Abstract

ObjectiveTo evaluate the impact of a hybrid control on usability of a P300-based brain-computer interface (BCI) system that was designed to control an assistive technology software and was integrated with an electromyographic channel for error correction. DesignProof-of-principle study with a convenience sample. SettingNeurologic rehabilitation hospital. ParticipantsParticipants (N=11) in this pilot study included healthy (n=8) and severely motor impaired (n=3) persons. The 3 people with severe motor disability were identified as potential candidates to benefit from the proposed hybrid BCI system for communication and environmental interaction. InterventionsTo eventually investigate the improvement in usability, we compared 2 modalities of BCI system control: a P300-based and a hybrid P300 electromyographic-based mode of control. Main Outcome MeasuresSystem usability was evaluated according to the following outcome measures within 3 domains: (1) effectiveness (overall system accuracy and P300-based BCI accuracy); (2) efficiency (throughput time and users' workload); and (3) satisfaction (users' satisfaction). We also considered the information transfer rate and time for selection. ResultsFindings obtained in healthy participants were in favor of a higher usability of the hybrid control as compared with the nonhybrid. A similar trend was indicated by the observational results gathered from each of the 3 potential end-users. ConclusionsThe proposed hybrid BCI control modality could provide end-users with severe motor disability with an option to exploit some residual muscular activity, which could not be fully reliable for properly controlling an assistive technology device. The findings reported in this pilot study encourage the implementation of a clinical trial involving a large cohort of end-users.