Neural network prediction method for VMAT radiotherapy plan

Abstract

The invention relates to the technical field of prediction of radiotherapy plans, in particular to a neural network prediction method for a VMAT radiotherapy plan, and is used for solving the problems of single control object, poor performance and large error of a prediction method for the radiotherapy plan in the prior art. The method comprises the following steps: step 1: data preparation, step 2: model design: establishing a multi-branch neural network model, processing the input of the multi-branch neural network model to obtain multi-dimensional features, and then outputting a predicted gamma passing rate; step 3, performing model training, firstly performing feature extraction, performing regression calculation on a gamma passing rate, then performing back propagation, updating parameters of the model, and performing continuous iteration for multiple times until good model parameters are obtained; step 4, performing model testing, and outputting a predicted gamma passing rate through forward calculation; and step 5, performing clinical treatment and verification. According to the method, radiotherapy plan prediction of more body parts can be accepted through the steps, the prediction performance error is smaller, and the effect is better.

Description

technical field [0001] The invention relates to the technical field of quality control of radiotherapy plans, and more particularly relates to a neural network prediction method for VMAT radiotherapy plans. Background technique [0002] In clinical tumor treatment, surgery, radiotherapy, and chemotherapy are the three most important treatment methods. Due to the wide indications and high selectivity of radiotherapy, more than 70% of malignant tumor patients need to be treated at a certain stage. Radiation therapy, modern radiotherapy technologies mainly include intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). Radiation therapy is an important cancer treatment method, which uses high-energy rays to irradiate target cells , destroy its DNA to stop growth, and make it lose the ability to divide and replicate. Since the accelerator used in its treatment is very precise and sensitive to the environment and the complexity of the treatment p...

AI Technical Summary

Problems solved by technology

[0008] Based on the above problems, the present invention provides a neural network prediction method for VMAT radiotherapy planning, which is used to solve the problems of single control object, poor performance and large error in the quality control method of radiotherapy plan in the prior art

Images