Abstract:Source estimation in magnetoencephalography (MEG) involves solving a highly ill-posed problem without a unique solution. Accurate estimation of the time course and spatial extent of the source is important for studying the mechanisms of brain activity and preoperative functional localization. Traditional methods tend to yield small-amplitude diffuse or large-amplitude focused source estimates. Recently, the structured sparsity-based source imaging algorithm has emerged as one of the most promising algorithms for improving source extent estimation. However, it suffers from a notable amplitude bias. To improve the spatiotemporal resolution of reconstructed sources, we propose a novel method called the source imaging method based on spatial smoothing and edge sparsity (SISSES). In this method, the temporal dynamics of sources are modeled using a set of temporal basis functions, and the spatial characteristics of the source are represented by a first-order Markov random field (MRF) model. In particular, sparse constraints are imposed on the MRF model residuals in the original and variation domains. Numerical simulations were conducted to validate the SISSES. The results demonstrate that SISSES outperforms benchmark methods for estimating the time course, location, and extent of patch sources. Additionally, auditory and median nerve stimulation experiments were performed using a 31-channel optically pumped magnetometer MEG system, and the SISSES was applied to the source imaging of these data. The results demonstrate that SISSES correctly identified the source regions in which brain responses occurred at different times, demonstrating its feasibility for various practical applications.

Stability and Super-resolution of MUSIC and ESPRIT for Multi-snapshot Spectral Estimation

Nonasymptotic Performance Analysis of ESPRIT and Spatial-Smoothing ESPRIT

Mathematical Foundation of Sparsity-based Multi-snapshot Spectral Estimation

A robust super-resolution approach with sparsity constraint for near-field wideband acoustic imaging

A Robust Super-Resolution Approach with Sparsity Constraint in Acoustic Imaging

Robust Bayesian Super-Resolution Approach Via Sparsity Enforcing a Priori for Near-Field Aeroacoustic Source Imaging

The ESPRIT algorithm under high noise: Optimal error scaling and noisy super-resolution

Stable estimation of pulses of unknown shape from multiple snapshots via ESPRIT

Nonasymptotic Analysis of Direct-Augmentation ESPRIT for Localization of More Sources Than Sensors Using Sparse Arrays.

SCAN-MUSIC: An Efficient Super-resolution Algorithm for Single Snapshot Wide-band Line Spectral Estimation

Distributed Radar Super-Resolution Imaging Based on MUSIC

Spectral-Spatial Joint Noise Estimation for Hyperspectral Images

A measurement decoupling based fast algorithm for super-resolving point sources with multi-cluster structure

Source Imaging Method based on Spatial Smoothing and Edge Sparsity (SISSES) and Its Application to OPM-MEG

Sparsity-Inducing Super-Resolution Passive Radar Imaging with Illuminators of Opportunity.

Large information plus noise random matrix models and consistent subspace estimation in large sensor networks

A New High-Resolution Esprit Algorithm Based on MSVD for Estimating Coherent Signals

A Super-Resolution Sparse Aperture ISAR Sensors Imaging Algorithm via the MUSIC Technique

The high resolution sampling methods for acoustic sources from multi-frequency far field patterns at sparse observation directions