Vascular Skeleton Line Reconstruction Method

Abstract

The invention provides a vascular skeleton line reconstruction and precise vessel diameter calculation method which includes the following steps: refining two-value vascular images in a CT (Computerized Tomography) image to form an initial skeleton line; dividing the vascular skeleton line into separate branches, and separating the skeleton line to form independent vascular sections; detecting the obtained branched vascular skeleton lines to remove surplus skeleton lines; smoothing all the obtained branched vascular skeleton lines to obtain precise center lines; outputting the smoothed vascular skeleton lines. The vascular skeleton line reconstruction and precise vessel diameter calculation method divides the initial skeleton line into separate branches to remove surplus skeleton lines and then smoothes the surplus skeleton lines, the obtained skeleton lines are located at the vascular center, and therefore a vascular model is made more accurate.

Description

technical field [0001] The invention relates to the technical fields of biomedical engineering and computer vision, in particular to a method for reconstructing a vascular skeleton line. Background technique [0002] In morphology, the centerline of an object is a description of the shape of an object after dimensionality reduction. It can not only combine the outline and area information of the object, but also reflect the important visual clues of the object; The linear connected structure is transformed into the abstract form of tree or graph, which facilitates the feature matching of objects. Centerlines represent objects with thin curved lines consistent with the connectivity and topology of the original object. [0003] At present, there is a method for obtaining the center line, which is the method adopted in K.Palágyi et al.'s "12-sub-iterative thinning algorithm for binary blood vessel images in CT images". This algorithm consists of several sub-iterations. A spec...

AI Technical Summary

Problems solved by technology

[0006] 2) The skeleton line is easy to deviate from the center line:

[0008] 3) The skeleton line is discontinuous and there are jagged edges:

[0009] The skeleton line obtained by this method is based on the CT sequence images, and the skeleton line formed must be based on the integer coordinates of the image layer, and the adjacent relationship can only be described by 8 directions, so the formed skeleton line Is not smooth and continuous, and prone to jagged

[0010] Therefore, the skeleton line obtained by this method can only be used as the initial skeleton line, and it is necessary to further improve the skeleton accuracy

Images