Dose determination method and device

Abstract

The invention discloses a dose determination method and device. The method comprises the following steps: calculating a first dose of a plan file through a first dose calculation algorithm; acquiringfeature information in the plan file, wherein the feature information comprises a CT image corresponding to the plan file; and inputting the first dose and the feature information into a deep learningnetwork model, outputting a corresponding second dose by the deep learning network model, the deep learning network model being formed by training a plurality of groups of training samples, wherein each group of training samples comprising the first dose and the feature information of the input plan file and the corresponding second dose. According to the invention, the technical problem of low calculation efficiency of a clinical dose determination method in related technologies is solved.

Description

technical field

[0001] The present invention relates to the field of dose control, in particular to a dose determination method and device. Background technique

[0002] At present, there are problems in the field of dose calculation. The high-accuracy dose calculation accuracy is high and the speed is slow, while the low-accuracy dose calculation accuracy is low and the speed is fast. The trade-off between accuracy and speed is an unsolved problem.

[0003] Some of the current Eclipse radiotherapy systems use the low-accuracy AAA (Analyticalanisotropic algorithm) algorithm for optimization, and use the high-accuracy AXB algorithm (Acuros XB) or Monte Carlo algorithm for final dose calculation to determine whether the plan is passed. The AAA algorithm is fast but its accuracy is lower than that of Monte Carlo and AXB. It may be that the AAA optimized plan deviates greatly from the prescribed dose after calculation using AXB, and a new plan is required.

[0004] For the abo...

Images