An Inverse EEG Problem Solving Environment One of the challenges of solving the inverse EEG source localization problem is choosing initial configurations for the downhill simplex solver. A good choice can result in rapid convergence, whereas a bad choice can cause the algorithm to search somewhat randomly for a very long time before closing in on the solution. Furthermore, because the solution space has many plateaus due to the linear finite element model, it is generally necessary to re-seed the algorithm multiple times in order to find the global minimum. We have brought the user into the loop by enabling seed-point selection within the model. The user can seed specifically within physiologically plausible regions. This focus enables the algorithm to converge much more quickly, rather than repeatedly wandering through non-interesting regions. To steer our algorithm, we utilized the SCIRun problem solving environment [40]. SCIRun is a scientific programming environment that allows the interactive construction, debugging, and steering of large-scale scientific computations. SCIRun can be envisioned as a ``computational workbench,'' in which a scientist can design and modify simulations interactively via a dataflow programming model. As opposed to the typical ``off-line'' simulation mode (in which the scientist manually sets input parameters, computes results, visualizes the results via a separate visualization package, and then starts again at the beginning), SCIRun ``closes the loop'' and allows interactive steering of the design, computation, and visualization phases of a simulation. The images of our algorithm running within the SCIRun environment are shown in Fig. 10 and Fig. 11.     Figure 11: The SCIRun problem solving environment. The user can select physiologically plausible regions of the model in which to seed the downhill simplex algorithm, thereby steering the algorithm to a more rapid convergence.     Fri Oct 8 13:55:47 MDT 1999

 
Revised: March , 2005