2026-05-30T23:13:44Zhttps://riuma.uma.es/rest/oai/request

oai:riuma.uma.es:10630/75802026-02-03T11:08:10Zcom_10630_2254col_10630_37953

A Brain–Computer Interface Speller with a Reduced Matrix: A Case Study in a Patient with Amyotrophic Lateral Sclerosis Ron-Angevin, Ricardo Varona-Moya, Sergio Da Silva-Sauer, Leandro Carrión-Robles, Trinidad Lingüística computacional Visual P300-based Brain–Computer Interface (BCI) paradigms for spelling are aimed at offering a non-muscular communication channel for those people with severe motor impairment, such as locked-in patients. To be as effective as other assistive technologies, these systems have to achieve a greater communication rate. One way to do so is to develop better interfaces. In this regard, we thought of using a 4 x 3 symbol matrix based on the T9 interface developed for mobile phones. Due to presenting a reduced matrix and relying on an adaptation of the T9 predictive text system, we expected that this speller would provide a higher communication rate than usual 6 x 6 matrix spellers that are based on Farwell and Donchin’s classic proposal. As a proof of concept, a locked-in patient with amyotrophic lateral sclerosis tested our T9-like visual BCI speller along with two different 7 x 6 conventional matrix spellers. The comparison of her performance results with those of a sample of three healthy participants suggested that it was possible for this locked-in patient to control the T9-like speller as well as they did, and thus, write a target sentence considerably faster than when she used the alternative spellers 2014-06-03T10:21:24Z 2014-06-03T10:21:24Z 2014-06-03T10:21:24Z 2014-06-03 journal article http://hdl.handle.net/10630/7580 eng COGNITIVE 2014 : The Sixth International Conference on Advanced Cognitive Technologies and Applications Venecia Mayo 2014 open access