Automatic Delineation Method of Target Area Based on Three-Dimensional Convolutional Neural Network Model

With the rapid development of image processing technology in recent years, the use of 3D vision, DSP imaging and artificial neural networks and other related theoretical knowledge is of great significance for identifying the lesions of genotype cancer patients. This paper introduces a three-dimensional convolutional BP neural network, and on this basis, completes the delineation of the nasopharyngeal carcinoma target area through training and classification. The purpose of this study is to provide medical technical support through the analysis of automatic delineation methods. This article mainly uses the comparative method to test and compare the volume of the three groups of experiments, and obtain the experimental objects through the survey method to analyze the data. Through experiments, it can be obtained that the P value of the volume difference between the MR-PWI group and the other two groups is less than 0.05, which is statistically significant. Applying three-dimensional convolutional neural network to target area delineation can improve the accuracy.