Deep brain source imaging with M/EEG and anatomical MRI

A method for imaging electromagnetic sources within deep brain structures based on M/EEG data identified while a subject is performing a functional task of interest, the method comprising:

• constructing source spaces in superficial (cortical) and deep (subcortical) regions of the brain from images of the brain, the source spaces including a first set of patches of the superficial and deep regions of a first preselected level of coarseness, and a second set of patches at a second preselected level of coarseness, each of the patches in each set being sized to homogenize current strengths, and wherein the first set of patches and the second set of patches span across the same surface;
• relating the source spaces with the M/EEG data;
• computing a minimum norm source current estimate (MNE) based on the relationship between the source spaces and the M/EEG data, the source spaces, and the M/EEG data,
• refining the MNE by iteratively reducing and expanding a subset of source spaces that are defined within the MNE
• to identify at least one superficial and at least one deep source whose activity underlies the M/EEG data.

A method for electromagnetic source imaging based on non-invasive M/EEG recordings and MM-based anatomic measures to employ information within M/EEG field patterns for estimating source currents across superficial and deep brain regions, the method comprising:

• constructing source spaces, the source spaces including a first set of cortical source spaces having patches with a first size, a second set of cortical source spaces having patches with a second size, and a third set of subcortical source spaces having a third size, wherein the second size being smaller than the first size, and wherein the first set of source spaces and the second set of source spaces span across the same surface;
• computing a minimum norm source current estimate (MNE) using the source spaces and the non-invasive M/EEG recordings;
• iteratively refining the MNE until the MNE includes
• a subset of source spaces having at least one superficial and at least one deep source whose activity underlies the non-invasive M/EEG recordings.

A subspace pursuit process for characterizing electrophysiological activity within deep brain regions based on non-invasive M/EEG measurements, the pursuit process being performed on a plurality of source spaces, the source spaces including a plurality of cortical surface patches and subcortical volume subdivisions for regions of the brain, the process comprising:

• acquiring measured data representative of brain activity of a subject, the measured data being indicative of electrophysiology activity within deep brain regions of the subject;
• computing a minimum norm source current estimate (MNE) using the cortical surface patches and the measured data;
• refining the MNE by performing a subspace pursuit to identify a first cortical surface patch that correlates to the measured data, the first cortical surface patch having a first size;
• further refining the MNE by performing a second subspace pursuit on a subset of cortical surface patches each having second size to determine a second cortical surface patch that correlates to the measured data, the second cortical surface patch being smaller than the first cortical surface patch and the subset of cortical surface patches, the second size being smaller than the first size, the subset of cortical surface patches at least partially overlapping with the first cortical surface patch;
• and
• performing a third subspace pursuit between the second cortical surface patch and the subcortical volume subdivisions to determine cortical and subcortical brain regions whose activity correlate best to the measured data.