Plaque classification method for coronary CT images

Abstract

The invention discloses a plaque classification method for coronary CT images. The method comprises: segmenting a coronary CTA sequence original image to obtain a coronary extraction map; extracting an coronary tree data isosurface, generating grid model data, calculating a normal vector, locating the start and end points from all points on a grid model, calculating the shortest path between the start point and the end point, and equidistantly filtering an obtained curve to obtain a center line and a radius; locating a bright spot in the coronary extraction map; locating a stenosis position and a stenosis section according to the radius and the curve; obtaining a coronary binary image with calcified plaque removed, obtaining the center line of the coronary binary image with calcified plaque removed, and calculating the radius; locating the stenosis position andthe stenosis section according to the radius and the curve, and determining that the plaque is a non-calcified plaque if the stenosis position is the same as the stenosis position in the step S04, otherwise determining that the plaque is a calcified plaque. The method can detect calcified and non-calcified plaques in CTA images and automatically classifies the plaques.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a plaque classification method for coronary CT images, which can be applied to X-ray coronary angiography image analysis in clinical research. Background technique [0002] Finding a safe and reliable method for detecting coronary artery disease is the main goal of clinical development in the future, so it is of great clinical value and practical significance to be able to accurately extract plaque from CTA image sequences to judge coronary artery disease. In the past ten years, the proportion of deaths caused by coronary artery disease has been increasing year by year. Therefore, it is necessary to accurately extract the quantification of arterial vascular disease, especially the early detection and quantitative analysis of plaque. However, the early detection and quantification of plaque requires very experienced doctors to spend a long time on manual plaque se...

AI Technical Summary

Problems solved by technology

However, the early detection and quantification of plaques requires very experienced doctors to spend a long time on manual plaque segmentation and analysis. Therefore, it is necessary to propose an automatic and rapid detection method for cardiac coronary plaques to improve the work efficiency of doctors.

[0003] In the detection direction of coronary artery plaques, some methods have been proposed to improve the detection of plaques, but almost all of them are for the detection of calcified plaques, and the incompletely developed non-calcified plaques in CTA images are lost.

Images