Automated processing of M/EEG
Ever since I began to work on electrophysiology and cognitive research questions, it has been my foremost conviction that reproducible research needs scalable automated data processing methods. Only when MEG and EEG data can be processed at scale and automatically, large data sets can be assembled at ease to support fore-front research using data-intense machine learning methods needed for optimizing prediction of clinical end-points in translational clinical neuroscience. Moreover, automated methods reduce researcher degrees of freedom, and, hence, may mitigate the ubiquitous replication crises encountered in several empirical sciences.
The goal of this axis of reasearch is to develop tools that simplify analysis of EEG and MEG data. Please consider the related publications listed below.
- Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states
- MNE: Software for Acquiring, Processing,and Visualizing MEG/EEG Data
- Encoding and Decoding Neuronal Dynamics: Methodological Framework to Uncover the Algorithms of Cognition
- A reproducible MEG/EEG group study with the MNE software: recommendations, quality assessments, and good practices
- Automated model selection in covariance estimation and spatial whitening of MEG and EEG signals