Source localization

Source localization in the context of functional brain imaging :

FAµST was used in an experience of source localization for brain image application.
Different FAµST approximation of a Magnetoencephalography (MEG) gain matrix “Mi” was computed using hierarchical_fact.
A SparseCoding algorithm was used to solve this source localization problem using the MEG gain matrix or its FAµST approximation.
The two following pictures illustrate the different trade-offs between speed-up and error of localization using dense or FAµST matrix.

Reproduce these figures! With matfaust: >> import matfaust.demo.bsl.* >> BSL() >> Fig_BSL() Seel also the API doc With pyfaust:: >>> from pyfaust.demo import bsl >>> bsl.run() >>> bsl.fig_convergence() >>> bsl.fig_time_cmp() Seel also the API doc