Method and device for extracting centerline of tree-shaped lumen structure in three-dimensional tomography image

Abstract

The invention provides a method and device for extracting a centerline of a tree-shaped lumen structure in a three-dimensional tomography image, and relates to the field of image processing.The method comprises the steps of obtaining the three-dimensional tomography image to be processed, carrying out over-segmentation preprocessing and feature extraction preprocessing respectively, processing the image subjected to over-segmentation preprocessing through an over-segmentation network to obtain a first over-segmented image, and refining the first over-segmented image to obtain a first center line image; extracting a depth feature corresponding to each pixel point in the image after feature extraction preprocessing through a feature extraction network; and taking each pixel point on the first center line image as a node, constructing a first graph model according to the spatial feature and the depth feature of each node, obtaining an initial center line through a graph convolutional network, performing optimization through a minimum spanning tree algorithm, and outputting an extraction result of the center line of the tree-shaped lumen structure. According to the invention, the center line can be accurately extracted, and the stability and robustness of center line extraction are improved.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a method and a device for extracting a centerline of a tree-like lumen structure in a three-dimensional tomographic image. Background technique [0002] At present, in the field of medical image processing technology, the use of computerized tomography technology can provide a visual representation of the internal tissues of the human body according to the different degrees of absorption of X-rays by different tissues of the human body. Tomographic imaging can produce two-dimensional tomographic images of human tissues. A series of two-dimensional images are used to record three-dimensional information of human tissues. This imaging method has the advantages of high speed and high resolution, and is currently a mainstream medical imaging method. The tree-like lumen in the three-dimensional tomographic image includes arteries, veins, trachea, etc. The inspection a...

AI Technical Summary

Problems solved by technology

However, since the boundary between some lumens and the background may be highly blurred, resulting in a high feature similarity between the region of interest and the background, it is difficult to distinguish, and whether it is in the coronary artery or on the trachea, due to the presence of There are many branches, and there may be adhesion between the branches due to image quality problems, resulting in a complex lumen structure with large changes, which makes it difficult to extract the centerline. Therefore, the centerline of the tree-like lumen Extraction still faces great challenges

Images