Method for constructing complete blood vessel tree structure by using continuous multi-frame images

Abstract

The invention discloses a method for constructing a complete blood vessel tree structure by using continuous multi-frame images, and relates to the technical field of coronary angiography image processing. The problems that the risk that the blood vessel form cannot be completely restored due to the fact that information is incomplete when an existing method is used for storing a single-frame image, and correction cannot be carried out during storage are solved. According to the invention, a complete blood vessel tree is automatically established according to an angiography image, and the method comprises the steps of coronary angiography blood vessel instance segmentation, segmented image post-processing, skeleton line extraction, continuous multi-frame image registration and blood vessel tree information storage. An image processing method based on the field of computer vision is used, and a unified judgment standard is used, so that misjudgment caused by human factors can be reduced. The blood vessel tree is established according to the proposed basis node information, so that the analysis of the blood vessel structure can be quickly helped. According to the invention, the stored skeleton line structure is furthest complete, the blood vessel information is furthest reserved, and a more reliable reference value is provided for establishing a complete blood vessel tree.

Description

technical field [0001] The invention relates to the technical field of coronary angiography image processing, in particular to a method for constructing a complete vascular tree structure using continuous multi-frame images. Background technique [0002] According to the reports of cardiovascular diseases in China in recent years, the mortality rate of cardiovascular diseases has been increasing year by year, even exceeding that of tumors and other major diseases. More patients and complicated diagnosis process make the doctor's workload suddenly increase. In addition, due to the complexity and variety of cardiovascular diseases, the doctor's professional knowledge and treatment and diagnosis standards are extremely important in the diagnosis of the disease. high demands. [0003] Existing patent name: a method for storing coronary angiography data sets based on skeleton images, patent number: 201911196224.3; this technical solution defines its coordinate offset according t...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the problem that the existing method is aimed at single-frame image storage, there is incomplete information, which leads to the risk that the shape of the blood vessel cannot be completely restored, and it cannot be corrected during storage. tree-structured approach

Images