DSA Vascular Image Segmentation Method Based on Sift Feature Point Clustering and Boolean Difference Operation

Abstract

The invention belongs to the field of medical image processing, and particularly relates to a DSA vascular image segmentation method based on SIFT feature point clustering and a Boolean different operation. The method comprises the following steps: corresponding DSA mask images and live images are inputted; under the condition of a same threshold value, geometric feature points of the mask images and the live images are extracted by utilizing an SIFT algorithm; adjustment of local position of the geometric feature points is performed by adopting a method based on gray gradient values; the geometric feature points clustering and the Boolean different operation are performed based on Euclidean distance; and vascular image segmentation is performed by adopting the method based on an Otsu threshold value. The process of obtaining vascular images of patients in DSA subtraction has advantages of being high in robustness and convenient in calculation process so that accurate vascular image data can be provided to clinical operations based on DSA interventional therapy.

Description

technical field [0001] The invention belongs to the field of medical image processing, in particular to a DSA blood vessel image segmentation method based on SIFT feature point clustering and Boolean difference operation. Background technique [0002] Digital Subtraction Angiography (DSA) technology is a powerful technology for visualizing blood vessels in X-ray sequence pictures. It is the product of the combination of conventional angiography and computer image processing technology. It has been clinically used for more than 20 years , is an important basis for non-invasive diagnosis of vascular diseases and surgical navigation for interventional therapy, and is widely used in the visualization system of blood vessels in X-ray sequential imaging. The basic principle is simply: X-rays pass through various anatomical structures of the human body to form fluorescent images, which are enhanced by an image intensifier and collected by a TV camera tube to form video images; then...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is just to overcome the deficiencies in the prior art, solve the problem of how to accurately use the patient mask image and live image to extract the patient's blood vessel image from the perspective of segmentation, and propose a method based on SIFT feature point clustering and Boolean DSA Vascular Image Segmentation Method Based on Difference Operation

Images