PROJECT TITLE :

Brain-Source Imaging: From sparse to tensor models

ABSTRACT:

A variety of application areas like biomedical engineering need solving an underdetermined linear inverse drawback. In such a case, it's necessary to make assumptions on the sources to restore identifiability. This drawback is encountered in brain-supply imaging when identifying the source signals from noisy electroencephalographic or magnetoencephalographic measurements. This inverse drawback has been widely studied throughout recent decades, giving rise to an impressive number of strategies using completely different priors. Nevertheless, a thorough study of the latter, as well as especially sparse and tensor-based approaches, continues to be missing. In this text, we tend to propose 1) a taxonomy of the algorithms based on methodological concerns; a pair of) a discussion of the identifiability and convergence properties, advantages, drawbacks, and application domains of numerous techniques; and 3) an illustration of the performance of seven selected ways on identical data sets. Directions for future research in the area of biomedical imaging are eventually provided.