Evaluación de electrodos EEG secos, húmedos y con gel para la implementación de sistemas BCI basados potenciales evocados P300

Over the years, there has been a search for the integration of intelligent systems capable of supporting rehabilitation and communication processes for patients in severe conditions, whether due to neurodegenerative diseases, congenital issues, or accidents. Implementing Brain-Computer Interface (BCI) systems in the daily lives of these users brings the opportunity to improve their quality of life once they leave a clinical environment. This technology enables communication and control devices through real-time analysis of brain activity. The most used technique to measure brain activity in BCI systems is electroencephalography (EEG). This technique records the electrical activity generated by neurons by placing electrodes on the scalp, where a conductive gel is applied to improve signal quality. However, using an electrolytic material each time the equipment is used presents a limitation in the goal of deploying BCI systems outside of laboratories. Considering this, various investigations have sought alternative methods for this application. Dry electrodes have been presented as a solution. However, this approach leads to a partial loss of EEG signal quality due to the removal of the conductive material, which reduces impedance and sensitivity to external artifacts. Therefore, this work aims to characterize dry, wet, and gel electrodes to assess their viability for use in BCI systems based on P300-evoked potentials. EEG recordings were obtained from 15 subjects in 2 sessions. In each session (per condition) was performed: 1 resting-state recording (with eyes closed) and 5 recordings using the RCP speller. In total, 450 recordings were obtained to evaluate the electrodes in BCI systems. Once the data was collected, the signals were preprocessed, features were extracted, and the intentions were classified to control the BCI system. As a result, it became possible to compare the different electrode responses for BCI systems. The results showed significant differences in EEG characteristics across the 3 conditions. Moreover, the analyzed BCI system achieved command decoding accuracy rates of 36%, 32%, and 87% using dry, wet, and gel electrodes, respectively. It was concluded that, under the experimental conditions, dry and wet electrodes do not allow for satisfactory control of the BCI system

• Autores: Sosa Rojo, Maria Valentina
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2024
• Institución: Escuela Colombiana de Ingeniería Julio Garavito
• Repositorio: Repositorio Institucional ECI
• Idioma: spa
• OAI Identifier: oai:repositorio.escuelaing.edu.co:001/3350
• Acceso en línea: https://repositorio.escuelaing.edu.co/handle/001/3350; https://catalogo-intra.escuelaing.edu.co/cgi-bin/koha/catalogue/detail.pl?biblionumber=23900
• Palabra clave: Electroencefalografía; Daño cerebral - Pacientes - Rehabilitación; Electrodos; Interacción hombre-computador; Brain computer interface (BCI); Evoked related–potentials (ERPs); P300; Row-Column Paradigm
• Rights: openAccess
• License: https://creativecommons.org/licenses/by-nc/4.0/