Recognition of the upper limbâ s motion intention using electroencephalogram and machine-learning techniques
Grant
Overview
Affiliation
View All
Overview
abstract
The EEG â electroencephalogram signals are a measurable electrical activity of the brain. EEG signals are generated 200-600 ms before the actual voluntary action, for example, as moving a limb to walk or drink water. This research helps to expand the growing biomedical research field in Qatar that relates to Brain-Computer Interface (BCI) technology. The study aims to help people with minimal arm movement, muscular dystrophy in a less invasive method. This research will help implement Qatarâ s aspiration of a healthy population, which is in adherence to Qatar National Vision 2030 of human development. The proposed study will utilize experimental and theoretical investigation to predict the userâ s movement through the prediction of EEG signals using machine learning algorithms. The study is divided into 3 phases: (1) Acquire EEG signals from healthy subjects using high tech ENOBIO EEG device. (2) Develop a unique machine-learning algorithm for predicting the motion intention. (3) Evaluate and improve the performance of the machine learning algorithm. The research team will determine the best machine learning algorithm through evaluating five different machine-learning algorithms including (1) k Nearest Neighbor (KNN), (2) Linear Discriminant Analysis (LDA), (3) Artificial Neural Network (ANN), (4) Support Vector Machine (SVM), (5) and Random Forest (RF). A rigorous statistical analysis will be conducted to evaluate the performance of the machine-learning algorithms. The evaluation stage will include analysis of variance (ANOVA) to compare the accuracy values due to using different window sizes, feature subsets, and algorithms. The research project will acquire EEG signals from ten healthy subjects while performing different activities. The research team has extensive experience in securing Institutional Review Board (IRB) approval for human studies and implementing them. Undergraduate students will be exposed to state-of-the-art technologies in the biomedical field; this includes using ENIBIO device, developing programming skills, and understanding machine learning algorithms. The exposure of future engineers to the research process is expected to improve their critical thinking, writing, and presentation skills.
