A Comparative Study of Minimum Norm Inverse Methods for MEG Imaging

摘要: The majority of MEG imaging techniques currently in use fall into the general class (weighted) minimum norm methods. minimization a is used as basis for choosing one from generally infinite set solutions that provide an equally good fit to data. This ambiguity Solution arises inherent non-unique-ness continuous inverse problem and compounded by imbalance between relatively small number measurements large source voxels. Here we present unified view methods describe how can Tikhonov regularization avoid instabilities due noise. We then compare Performance regularized versions three well known linear [51 [7] with non-linear iteratively reweighted method [11 Bayesian approach described our companion paper (“MEG-based Imaging Focal Neuronal Current Sources,” Phillips J.W., Leahy R.M., Mosher J.C.).