EEG Multipurpose Eye Blink Detector using convolutional neural network

The electrical signal emitted by the eyes movement produces a very strong artifact on EEG due to its close proximity sensors and abundance of occurrence. In context detecting eye blink artifacts in waveforms for further removal purification, multiple strategies where proposed literature. Most commonly applied methods require use large number electrodes, complex equipment sampling processing data. goal this work is create reliable user independent algorithm removing signals using CNN (convolutional neural network). For training validation, three sets public data were used. All contain samples obtained while recruited subjects performed assigned tasks that included voluntarily specific moments, watch video read an article. model used study was able have embracing understanding all features distinguish trivial from contaminated with without being overfitted only occurred situations when registered.