Abstract: Described embodiments relate to determining an objective function to be used for mapping radiation beams to a patient body volume. The objective function has a first term related to at least one target volume and a second term related to at least one non-target volume. The second term is zero only when a product of the weight of a beamlet mapped to pass through a non-target volume portion and the dose deposited by said beamlet is equal to a first predetermined average dose constraint value for the respective non-target volume portion, for all beamlets mapped to pass through the at least one non-target volume. This limit aims to reduce the occurrence of negative beam weights. In another embodiment, the objective function has a smoothing term for biasing the weight of beamlets towards a uniform distribution within the respective beam. Radiotherapy is delivered based on the objective function.

Abstract: The invention relates to methods and systems for computationally efficient optimization of radiation dose delivery that involve determining an objective function to be used for mapping radiotherapy beams to a patient body volume having at least one target volume and at least one non-target volume. The objective function has a first term related to the at least one target volume and a second term related to the at least one non-target volume. The second term is zero only if the weights of beamlets mapped so as to pass through the non-target volume(s) are zero. The second term helps to avoid the occurrence of negative beam weights, thereby facilitating computationally efficient optimization using matrix inversion. An optimal set of weights of beamlets is determined using the objective function. Radiotherapy is delivered based on the determined optimal set of weights.