Localization of Brain Sources
Saeid Sanei, Jonathon A. Chambers · 2021
Localization of brain signal sources from solely electroencephalography (EEG) has been an active area of research during the last two decades. Brain source localization is probably the most challenging and difficult operation in dealing with EEG signals due to three main reasons: the human head is nonhomogeneous; and human heads are neither spherical nor have similarity to each other, and finally, the sources can be multiple, distributed, or correlated. One of the requirements for brain source localization, particularly for the forward model, is the information about the head model. The head model is the model for which the EEG forward solution is calculated. Poor spatial resolution of EEG/magnetoencepalogram motivates research into methods that can more accurately localize the sources from the recordings using these modalities. A popular strategy in brain source localization is by using the dipole source assumption. Multiple sparse priors and iterative regularization algorithms have been used to solve the inverse localization problem.