Radiation therapy inverse treatment planning using a regularization of sparse segments

Abstract

A method of reducing a total number of beam segments in a dose distribution for a radiation therapy field is provided. The method includes providing a multiobjective radiation therapy treatment plan using a suitably programmed computer, where the multiobjective radiation therapy treatment plan includes a radiation beam dose performance objective and a fluence map sparsity objective in a given fluence function domain, and providing a Pareto frontier of tradeoff criteria between the beam dose performance and a total number of radiation segments (or sub-fields) of the multiobjective radiation therapy treatment plan using the suitably programmed computer, where an achieved set of radiation beam dose distributions associated with efficiency points of the Pareto frontier are evaluated using a clinical acceptance criteria, where a clinically acceptable radiation beam dose distribution having a smallest number of the multileaf collimator segments is a final solution for the multiobjective radiation therapy treatment plan.

Dependencies are inferred from the claim text.