MEGqc - a standardized pipeline for MEG data quality control.
Magnetoencephalography (MEG) data can be easily corrupted by noise and artifacts, originating from environmental, biological sources, or device malfunction. Currently, the MEG community lacks the software to assess the quality of the data automatically, leading to individual quality control by researchers without a universal standard. To address this issue, we are developing MEGqc, a quality control tool for MEG data inspired by the already existing MRIqc tool for MRI data.
MEGqc performs an automated quality assessment on BIDS conform MEG data sets, relying on the MNE-python software package. The software detects specific noise patterns in the data and visualizes them in easily interpretable human-readable reports. Additionally, the calculated metrics are provided in machine-readable JSON files, allowing for better machine interoperability or integration into workflows. MEGqc identifies environmental (e.g.powerline), biological (heartbeat, blinks, muscle movement), and systematic noise sources (sensor malfunction).
The software strives to help researchers to standardize and speed up their quality control workflow and to select the data sets of the quality best suitable for reusing in further research. Furthermore, evaluating the quality of an empty-room recording prior to experimentation can identify machine malfunctions and environmental noise, conserving financial and human resources in MEG research.
2007-2012 B. Sc. Psychology, Ural Federal University, Russia
2020-now MSc. Neuroscience, University of Oldenburg