Generating and applying robust dose prediction models

A computer-implemented method, comprising:

• accessing a plurality of treatment plans that have associated therewith perturbations of nominal values of parameters of the treatment plans;
• for each treatment field of the treatment plans, determining a field-specific planning target volume (fsPTV) based on the perturbations;
• accessing information that delineates at least one clinical target volume (CTV) and information that delineates at least one organ-at-risk (OAR) volume, wherein each said OAR volume comprises at least one sub-volume that is delineated based on spatial relationships between each said OAR volume and the CTV and the fsPTV for said each treatment field;
• determining a plurality of dose distributions comprising a dose distribution based on the nominal values and dose distributions based on the perturbations;
• and
• generating a dose prediction model for each said sub-volume, the model trained using the plurality of dose distributions.

A computer-implemented method, comprising:

• accessing a radiation treatment plan;
• calculating dose-volume histograms for the radiation treatment plan using a plurality of dose prediction models, wherein the dose prediction models have associated therewith nominal values of parameters of the radiation treatment plan and perturbations of the nominal values, and wherein said calculating comprises:
• calculating a dose-volume histogram using the nominal values;
• and
• calculating dose-volume histograms using the perturbations;
• and
• generating objectives of the radiation treatment plan using the plurality of dose prediction models and based on the plurality of dose-volume histograms for the radiation treatment plan.