Multi-scale nasopharyngeal tumor segmentation based on CNN

Abstract

The invention relates to a multi-scale nasopharyngeal tumor segmentation method based on CNN. Includes collecting MRI image data of nasopharyngeal region of several cases with nasopharyngeal tumor; Performing artificial edge labeling on the lesion area of the MRI image data collected in the previous step as label data layer by layer; Performing standardized preprocessing on the label data obtainedin the previous step and converting the label data into a two-dimensional data set; A CNN-based multi-layer two-dimensional convolution neural network is constructed and trained by using the two-dimensional data set in the previous step. For the MRI image data of nasopharyngeal region to be segmented, medical images of the same region and the same mode are collected, and the collected images arestandardized. The MRI image data of nasopharyngeal region to be segmented is segmented automatically by the network model. The invention can realize automatic segmentation of nasopharyngeal tumor, andcan obtain higher precision compared with mainstream network.

Description

technical field [0001] The present invention relates to the technical field of nasopharyngeal tumor image segmentation in the field of image segmentation, in particular to a CNN-based multi-scale nasopharyngeal tumor segmentation method. Background technique [0002] Nasopharyngeal carcinoma refers to malignant tumors that occur on the roof and side walls of the nasopharyngeal cavity. It is one of the high-incidence malignant tumors in my country, and its incidence rate is the first among ear, nose and throat malignant tumors. Nasopharyngeal tumors have a more complex anatomy than other types of tumors. Nasopharyngeal tumors are spatially similar to several tissues (air, bone, muscle, and mucous membrane) that process similar image intensities, and the shape and size of nasopharyngeal carcinomas, as well as non-uniform tumor intensities, are quite different. Therefore, it is necessary to design specific segmentation method. Moreover, the traditional diagnosis of nasophary...

AI Technical Summary

Problems solved by technology

However, this approach will cause two problems: a large number of redundant calculations lead to low time efficiency, and the network cannot learn global features

However, this method does not extract enough global feature information, especially when the amount of data is small, it cannot fully learn global features.

In summary, there are two problems in the above method, on the one hand, lack of reuse of original feature information, on the other hand, insufficient learning of global feature information

Images