Analyzing Electroencephalograms Using Cloud Computing Techniques

摘要: Brain Computer Interfaces (BCIs) allow users to interact with a computer via electroencephalogram (EEG) signals generated by their brain. The BCI application that we consider allows user initiate actions such as keyboard input or control the motion of wheelchair. Our goal is be able train neural network and classify EEG from multiple infer intended in distributed environment. processing developed using Map-Reduce framework. We use our cloud runtime, Granules, these streams. One objectives process streams real-time. software has been R, which an interpreted language designed for fast computation matrix multiplications, making it effective development artificial networks. contrast approach Granules competing uses R package – Snowfall simplifies execution computations setting. have performed experiments evaluate costs introduced scheme training networks classifying signals. results demonstrate suitability