Constructing head models by computation

A new method that considers the head as an inhomogeneous object consisting of many small homogeneous elements is presented. The authors consider the feasibility of including the inhomogeneity of head tissue conductivity in developing a general framework for constructing a head model for forward and backward computation. The new method proposed here takes the inherent inhomogeneity of the head tissues into consideration in developing a head model. The advantages of numerical FEM, which can handle the inhomogeneity, are used to their full effect. One possible way of constructing a head model by computation was explored. The simulation results confirm that it is a promising methodology, although there is a lot of work yet to be done. The current problem is to determine how to include more location-related conductivity data into the authors' model and to reduce the number of model data sets. Another task is to explore the efficiency of the annealing algorithm in the infinite cases.