Reconstruction of Images from Neuromagnetic Fields

An Evaluation of Methods for Neuromagnetic Image Reconstruction

“…Neuromagnetic imaging on a single reconstruction plane in the brain was first demonstrated by Singh et al [7]. Their approach used discrete samples of the evoked magnetic field to reconstruct a current field modeled as discrete current dipole cells all lying in a single plane with either one-or two-constrained dipole orientations.…”

“…It therefore appears unlikely that high resolution volume images of brain activity, comparable to images obtained from an emission tomography system for example, can be reconstructed using SQUIDbased magnetic measurements. However, the use of restricted models, such as those used by Singh et al [7] or those presented in Section II-B, in conjunction with a minimum dipole approach should lead to clinically useful results. From the results obtained using the minimum norm and maximum entropy methods, we conclude that these approaches are not suitable for neuromagnetic imaging if the sources are actually distributed in three-dimensions.…”

“…In these cases, the single-dipole model is a poor match [5], [6] and we are motivated to treat the problem as one of image reconstruction. Our approach then emphasizes solutions to the inverse problem (current field from magnetic measurements) using image reconstruction algorithms and techniques which one might encounter in computed tomography or other medical imaging applications.Neuromagnetic imaging on a single reconstruction plane in the brain was first demonstrated by Singh et al [7]. Their approach used discrete samples of the evoked magnetic field to reconstruct a current field modeled as discrete current dipole cells all lying in a single plane with either one-or two-constrained dipole orientations.…”

Algorithms for biomagnetic source imaging with prior anatomical and physiological information

“…A more recent approach models the &own distribution as an array of dipoles with fixed positions but unknown magnitudes [45,79,92,105,1061. Then the magnetic field measurements b can be written as a linear function b = Fq + w of the unknown current distribution q and measurement noise w. The forward transfer matrix F is determined by solving the forward problem for unit sources.…”

Local multilayer analytic sensing for EEG source localization: Performance bounds and experimental results

“…Therefore, additional assumptions are required to make the solution unique. For this purpose, the various methods available are putting forward different hypotheses about the source model, which can be a single (equivalent) dipole [14,16], multi-dipole [8,9], or distributed dipoles [5,10,12]. In the latter case, the problem is underdetermined since the number of parameters of the model exceeds the number of measurements.…”