Dosimetry assessment method and system for organ at risk in esophageal radiotherapy plan

A dosimetry assessment method for an organ at risk (OAR) in an esophageal radiotherapy plan, comprising:

• step 1:
• acquiring volumetric-modulated arc therapy (VMAT) plan data of a patient with esophageal cancer, comprising a computed tomography (CT) image and a structural contour image that are extracted;
• step 2:
• calculating the VMAT plan data for the esophageal cancer to obtain a distance-target histogram (DTH);
• step 3:
• subjecting a geometric feature vector of the DTH in step 2 to dimension reduction through an encoding layer in an autoencoder structure to obtain a dimension-reduced geometric feature vector;
• step 4:
• establishing a deep belief network (DBN) model, iterating until convergence, and completing training the DBN model;
• step 5:
• non-linearly fitting a correlation between a dose feature vector and the dimension-reduced geometric feature vector by the dimension-reduced geometric feature vector through a trained DBN to obtain a dose feature vector with a dimension being identical to a dimension of the dimension-reduced geometric feature vector;
• step 6:
• reconstructing the dose feature vector of step 5 through a decoding layer of the autoencoder structure to obtain a dose feature vector with a dimension being identical to a dimension of the geometric feature vector before the dimension reduction, and finally obtaining a dose-volume histogram (DVH) for predicting the OAR.