Deep Learning Approach for Automatic Classification of Ocular and Cardiac Artifacts in MEG Data

Figure 2

Architecture of the neural network model with an example of a typical ocular artifact component. The model is trained using the temporal (a) and spatial (b) information simultaneously, passing it through the different input branches (c) and (d). Features from both domains are extracted (e) and passed into a 2-node layer (f) for classification.