Interface multimodal com predição de movimentos para uso em reabilitação de membros inferiores

• Ano de defesa: 2014
• Autor(a) principal: Araújo, Douglas Ruy Soprani da Silveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Electroencephalography; Electromyography; Inertial sensors; Multimodal platform; Rehabilitation robotic; Interface Multimodal; Sensores Inerciais; Robótica – Reabilitação; Processamento de sinais; Eletroencefalografia; Eletromiografia; Eletrônica Industrial, Sistemas e Controles Eletrônicos; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/9644

his master thesis presents a multimodal platform for acquisition and signal processing. The proposed interface acquires, synchronizes and processes electroencephalographic (EEG) signals, electromiographic signals (EMG) and inertial sensors (IMUs) signals. The data acquisition is done in experiments with healthy subjects performing motor tasks of lower limbs. The objective is to analyze the movement intention, the muscle activation and the movement onset. To do so, an offline analysis was performed. In the analysis are shown EEG signal processing techniques, whose aim is to identify movement intention, and EMG signal techniques aiming at identifying the initial muscle activation. Techniques for processing signals from inertial sensors whose aim is to identify movement onset and measure the knee joint angles are also shown. An experimental protocol is proposed. The platform can be used in the development of interfaces for rehabilitation robotics devices aiming at adapting their control with respect to the patient’s intention. The results obtained showed that the system is capable to acquire, process and classify the signals synchronously. The movement intention was detected in 76, 0 ± 18, 2% of the movements. The movement antecipation achieved 716, 0 ± 546, 1 ms based on EEG signal and 88, 34 ± 67, 28 ms based on EMG signals. The results of the biological signal processing, the movement antecipation times, the accuracy of classifiers and joint angles measurements were in accordance with the currently related studies.