PET-CT multimodality nasopharyngeal carcinoma image segmentation method based on hypergraph model

Abstract

The invention relates to a PET-CT multimodality nasopharyngeal carcinoma image segmentation method based on a hypergraph model. The PET-CT multimodality nasopharyngeal carcinoma image segmentation method comprises the steps of: extracting gray scale information and location information of pixel points in a nasopharyngeal tumor image to construct a data set, constructing a sparse representation model according to the data set, solving the sparse representation model to obtain a reconstruction coefficient matrix and constructing hyperedges, calculating a similarity degree of data samples by utilizing a Gaussian kernel model, regarding the similarity degree as a weight value of the hyperedges, and solving a hyperedge order and a vertex order to construct a hypergraph Laplacian matrix; and marking the nasopharyngeal tumor image to obtain mark vectors, constructing a semi-supervised learning model based on the mark vectors, solving a least square problem to obtain an optimal tangent vector, and finally returning a classification result to the pixel level, namely, completing the segmentation of the tumor image. The PET-CT multimodality nasopharyngeal carcinoma image segmentation method is higher in segmentation precision when compared with a single-modality method, and the hypergraph model based on the combination of sparse representation and Gaussian kernel has higher segmentation precision for nasopharyngeal tumor image data when compared with other simple graph model or hypergraph model.

Description

technical field [0001] The invention belongs to the field of tumor image segmentation, and in particular relates to a PET-CT multimodal nasopharyngeal carcinoma image segmentation method based on a hypergraph model. Background technique [0002] With the development of medical imaging technology, medical image segmentation has gradually become an important research topic and has great significance in clinical applications. Especially in the processing of tumor images, accurate segmentation of tumor images can help doctors more accurately outline tumor contours to determine radiotherapy targets. However, most of the traditional tumor image segmentation methods are aimed at single-modal image data, which contains less information, so it is difficult to achieve high segmentation accuracy. [0003] Tumor is a malignant disease that endangers human health. Head and neck cancer is the fifth most common tumor in the world. Nasopharyngeal carcinoma, as the most common tumor in the ...

AI Technical Summary

Problems solved by technology

For example: use the K nearest neighbor algorithm to determine the connection of sample points to construct a hyperedge, and then obtain the weight value of the hyperedge based on the hyperedge. However, due to the fixed K value of this method, it is difficult to obtain ideal classificatio

Images