Movement detection by brainwave : cloud computing

Abstract

This final year project aims to develop a low-cost wearable system that can detect the movement intention of the user using single-channel electroencephalography (EEG). The EEG sensor records EEG raw data in real-time from the user and communicates with the computer to analyze the changes in brain activities. The python program performs raw data extraction, signal processing, and data analysis in real-time. The classifier will identify the extracted feature via a pre-trained model utilizing a machine learning approach while another classification method is introduced to overcome the limitations of the machine learning approach in this project. The detection result of movement intention will be feedbacked to the user with a build-in alert feature to acknowledge the user when movement intention is detected. This design aims to provide a low-cost movement detection healthcare system for assisting the elderly/home-alone. In this study, repeated experiments with different sets of hardware were recorded to ensure the reliability of the data collected, to explore the correlation between EEG signal and lower-limb muscle activities, and to validate the feature extracted. The result of this prototype suggests that a single-channel EEG signal can be used to measure the movement intention of the user and provide feedback before the onset of the movement.