Preprocessing

Program

12:00–12:20	Quality assurance	How to do scalable quality assurance
12:20–12:40	Filtering	Filtering the data
12:40–13:10	Autoreject	Removing sensor artifacts in data
13:00–13:30	SSP	Signal Space Projection
13:30–14:00	ICA	Independent Component Analysis

Quality assurance

We will see how to use the MNE report to generate figures for quality assurance when analyzing tens of hundreds of subjects. We will also look at how to parallelize the analysis.

Filtering

A brief look at the filtering options available in MNE Python

Autoreject

Sometimes, sensors can be bad due to loose contact or flux jumps. Autoreject is an automated method to label and repair artifacts in the data.

Spatial filtering

Physiological artifacts that are not related to brain rhythms such as heart beats and eyeblinks have prototypical spatial signatures. They can be removed using Signal Space Projection (SSP) or Independant Component Analysis (ICA).