Method for automatically segmenting chest anterior mediastinal focus based on CT image

Abstract

According to the method for automatically segmenting the chest anterior mediastinal focus based on the CT image, the segmentation result is good, quantitative feature analysis of subsequent deep learning can be met, and a methodological basis is provided for automatic segmentation research of the anterior mediastinal focus. The method for automatically segmenting the chest anterior mediastinal focus based on the CT image comprises the following steps: (1) CT image acquisition: performing chest CT contrast enhancement scanning on all patients to obtain a thin-layer image reconstructed by a mediastinal window; (2) designing a double-lung mask file on the original CT image by using the natural density difference between the lung and surrounding tissues and adopting a method for calculating the pixel value of lung tissues, and removing the area outside the mediastinal septum; (3) carrying out preliminary segmentation on the focus by adopting a V_Net network; (4) carrying out fine segmentation on the focus by adopting a Morphological Snakes algorithm; and (5) evaluating a segmentation result.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a method for automatically segmenting chest anterior mediastinal lesions based on CT images. Background technique [0002] The purpose of preoperative imaging examination of thoracic anterior mediastinal lesions is to initially evaluate the malignancy of the tumor, and further speculate on its histopathological changes and conduct risk assessment, so as to assist in the selection of preoperative treatment options and clinical prognosis. The evaluation of CT imaging signs is based on the lesion itself and its relationship with the surrounding tissue structure, and empirical and observational indicators are mostly used instead of quantitative indicators. Radiomics is a computerized quantitative imaging analysis method that can perform high-throughput feature extraction on medical imaging images. However, the traditional radiomics segmentation of lesions relies on m...

AI Technical Summary

Problems solved by technology

However, due to the relatively low incidence of anterior mediastinal lesions, there are few studies on the automatic segmentation of CT images, so how to accurately segment anterior mediastinal lesions still faces great challenges

Images