μ-STAR: A novel framework for spatio-temporal M/EEG source imaging optimized by microstates

• DOI: 10.1016/j.neuroimage.2023.120372
• Publication Date: 2023-09-24T12:06:00Z
• Authors: Zhao Feng; Sujie Wang; Linze Qian; Mengru Xu; Kuijun Wu; Ioannis Kakkos; Cuntai Guan; Yu Sun
• Keywords: Magnetoencephalography; Benchmark (surveying); Temporal resolution

Source imaging of Electroencephalography (EEG) and Magnetoencephalography (MEG) provides a noninvasive way monitoring brain activities with high spatial temporal resolution. In order to address this highly ill-posed problem, conventional source models adopted spatio-temporal constraints that assume stability the activities, neglecting transient characteristics M/EEG. work, novel method μ-STAR includes microstate analysis Bayesian model was introduced problem. Specifically, applied achieve automatic determination time window length quasi-stable activity pattern for optimal reconstruction dynamics. Then user-specific prior data-driven basis functions were utilized characterize information sources within each state. The solution obtained through computationally efficient algorithm based upon variational convex analysis. performance first assessed numerical simulations, where we found inclusion in significantly improved reconstruction. More importantly, achieved robust under various settings (i.e., numbers/areas, SNR levels, depth) fast convergence speed compared five widely-used benchmark (including wMNE, STV, SBL, BESTIES, & SI-STBF). Additional validations on real data then performed two publicly-available datasets block-design face-processing ERP continuous resting-state EEG). reconstructed exhibited neurophysiologically plausible results consistent previously-revealed neural substrates, thereby further proving feasibility applications.