MEG/EEG Source Reconstruction, Statistical Evaluation, and Visualization with NUTMEG

Figure 5

Time-series source estimation GUI. The left subplot shows the averaged sensor time series for each sensor. The right subplot changes depending on user selection. The eigenvalues can be plotted (as shown) for assisting the eigenspace beamformer; alternatively, cleaned sensor data and factors can be plotted if SEFA denoising is selected. The condition number of the sensor data covariance is displayed to help prevent meaningless results computed with overly high condition numbers. Other options here include filtering, time window selection, averaging before/after data covariance computation, method of source inversion, other denoising methods, sensor covariance regularization, and the option to submit the computation to a grid cluster for site-specific setups.