Source reconstruction and network construction
in this study, we conduct source level network analysis. the first step is to reconstruct source activation. source reconstruction analysis was performed with Brainstorm
\citep{Tadel_2011}, which is documented and freely available for download online under the GNU general public license (
http://neuroimage.usc.edu/brainstorm). Brainstorm integrates various distributed source model methods and can readily conduct source analysis. Here we used Brainstorm to estimate source activations of interictal MEG with or without spikes. We applied the method of Standardized Low Resolution Electromagnetic Tomography Analysis (sLORETA) (
Pascual-Marqui 2002) to identify and evaluate active sources. The sLORETA is based on Minimum Norm Estimation (MNE) and The activity (current density) is normalized by an individual estimate of the source standard deviation at each location. We applied the method of Minimum Norm Estimation (MNE) It has been identified as an efficient tool for functional mapping, since it is consistent with physiology and capable of correct localization. MEG were projected on individual cortical surfaces reconstructed by freesurfer. Cortical surface was divided into 15002 grid points (sources). in our study, when source current density was calculated, source orientation is constrained and is perpendicular to cortical surface. In a first step, MNE source estimation was calculated for whole brain at voxel level. Secondly, for computation convenience, source activation were spatially downsampled into 68 Desikan-Killiany atlas and within each atlas, the averaged source waveforms over each atlas were used to represent the atlas waveforms. Desikan-Killiany atlas relies on automatic parcellation using a surface-based alignment of the cortical folding (
\citep{Desikan_2006}). (矩阵重建部分,就按照你文章中写的就可以了)