CT image target lung segment identification method based on pulmonary artery tree grading

Abstract

The invention relates to a CT image target lung segment identification method based on pulmonary artery tree grading, and the method comprises the following steps: S1, obtaining CT image data, and reconstructing lung structures, including pulmonary nodules, bronchus, pulmonary artery trees and lung lobes; s2, extracting an initial center line from the pulmonary artery tree, and constructing a complete and continuous topological tree through an endpoint detector and a trajectory extractor; s3, according to the topological tree, combining lung lobe and bronchus prior knowledge for constraint, and determining 18 artery segment subtrees; s4, determining a targeted artery branch; s5, calculating the nearest arterial segment subtree for each voxel point on the lung lobe, and attributing the nearest arterial segment subtree to the corresponding lung segment; similarly, according to the distance relationship, calculating the dominated area of the target artery branch in the associated lung segment, and determining the target lung segment. According to the method, the relevance between the lung anatomical structures is fully considered, grading labeling of arteries is achieved, targeted artery branches are accurately determined, and then the lung segment is divided and the target lung segment is determined.

Description

technical field [0001] The invention relates to the field of image recognition, and relates to a method for recognizing a target lung segment of a CT image based on pulmonary artery tree classification. Background technique [0002] The confirmation of the target lung segment is based on extracting the lung parenchyma, dividing the corresponding lung segment according to the classification of the pulmonary bronchi, and then determining the minimum resection range according to the resection margin of the lesion. The target lung segment is not only a single lung segment, but also the combined lung segment, combined sub-segment, and combined lung segment and sub-segment. However, due to the partial volume effect of CT imaging, bronchial resolution is limited, and it is difficult to obtain complete subsegmental bronchi. Given that subsegmental bronchi are accompanied by pulmonary arteries at corresponding levels, and arteries have better visibility and characteristics on CT ima...

AI Technical Summary

Problems solved by technology

However, the current methods still have the following limitations: (1) they are easily affected by the CT imaging quality and the patient's disease, resulting in insufficient bronchial segmentation and poor lung segment segmentation; (2) at the same time, it is difficult to extract the bronchi to the subsegment level Carry out sub-segment division; (3) The anatomical structure of the lung is complex and changeable, often considering the local anatomical features, ignoring the overall structure of the lung, and not fully combining the corresponding relationship with other anatomical structures

Images