Abstract

Beam angle optimization (BAO) largely determines the performance of fixed-field intensity modulated radiation therapy (IMRT), and it is usually considered as non-convex optimization and an NP hard problem. In this work, BAO is reformulated into a highly efficient framework of standard quadratic optimization. The maximum of beamlet intensities for each incident field as the surrogate variable indicating whether one radiation field has been selected. By converting the function of maximum value in the objective into a set of linear constraints, the problem is solved as standard quadratic optimization via reweighting l1-norm scheme. The performance of the proposed BAO has been verified on a digital phantom and two patients. And the conclusion is drawn: the proposed convex optimization framework is able to find an optimal set of beam angles, leading to improved dose sparing on OARs in fixed-field IMRT.