SPATIOTEMPORAL EEG SOURCE LOCALIZATION USING SIMULATED ANNEALING

The estimation of multiple dipole parameters in spatio-temporal source modeling (STSM) of electroencephalographic (EEG) data is a difficult nonlinear optimization problem due to multiple local minima in the cos t function. A straightforward iterative optimization approach to such a problem is very susceptible to being trapped in a local minimum, the reby resulting in incorrect estimates of the dipole parameters. In thi s paper, we present and evaluate a more robust optimization approach b ased on the simulated annealing algorithm. The complexity of this appr oach for the STSM problem was reduced by separating the dipole paramet ers into linear (moment) and nonlinear (location) components. The effe ctiveness of the proposed method and its superiority over the traditio nal nonlinear simplex technique in escaping local minima were tested a nd demonstrated through computer simulations. The annealing algorithm and its implementation for multidipole estimation are also discussed. We found the simulated annealing approach to be 7-31% more effective t han the simplex method at converging to the true global minimum for a number of different kinds of three-dipole problems simulated in this w ork. In addition, the computational cost of the proposed approach was only marginally higher than its simplex counterpart. The annealing met hod also yielded similar solutions irrespective of the initial guesses used. The proposed simulated annealing method is an attractive altern ative to the simplex method that is currently more common in dipole es timation applications.