Electroencephalogram Based Exoskeleton For Finger Rehabilitation
Published in Undergraduate Thesis, 2019
Abstract: More than 200 million people worldwide live with some type of disability, with a large number of people having motor disabilities. This hinders their ability to perform daily activities of life. Rehabilitation is a technique through which this lost motor functioning can be regained. In rehabilitation the affected part is subjected to continuous passive motions, this exploits the plasticity trait of human brain in achieving back the lost motion. Exoskeleton have potential capabilities to provide rehabilitation. Several studies in the past have utilised exoskeletons to provide the same. In the proposed study, we have designed an exoskeleton consisting of the following parts Circular part, Linkage structure, Base. For calculating the dimensions of linkages we used anthropometric data of human finger and dynamically simulated the model to testify that the designed part is mimicking the natural trajectory of human hand. Static simulation of the designed model is also carried out to assess the structural strength. In the simulation results, all the design parameters comes well under stipulated limits. For actuation and control, we proposed the use of microcontrollers coupled with Brain Machine Interface (BMI). The different stages of BMI include Signal acquisition, feature extraction, feature selection and classification. The EEG dataset used in this work was created by the University of Tubingen, Germany. EEG signals associated with the imagined movement of right hand and relaxation state were processed using wavelet transform analysis for feature extraction. The optimum classification performance of 82.911% was achieved with a random forest classifier. As a Future prospect this classified data can be utilised to programme a microcontroller linear actuator named Firgelli for the execution of different tasks.