In this paper, we present a nonlinear interactive multiobjective opti-mization method for radiotherapy treatment planning using the Boltz-mann transport equation (BTE) in dose calculation. In radiotherapy,the goals are to destroy a tumor with radiation without causing damageto healthy tissue. These goals are conflicting, i.e. when one target is op-timized the other will suffer, and the solution is a compromise betweenthem. Our interactive approach is capable of handling multiple andstrongly conflicting objectives in a convenient way, and thus the weak-nesses of widely used optimization techniques in the field (e.g. definingweights and trial and error planning) can be avoided. In this paper,we use a parameterization technique to make the dose calculation faster in the BTE model. With our approach, the number of solutions to becalculated is rather small but still informative, planning times can beshortened and plan quality improved by finding only feasible solutionsand advantageous trade-offs. Importantly, we used the radiotherapyexpert’s knowledge to direct the solution process. To demonstrate theadvantages, we compare the results with those of the commonly usedpenalty function method.
QC 20120213