Nonrigid medical image registration method based on self-adapting triangular meshes

Abstract

The invention discloses a nonrigid medical image registration method based on self-adapting triangular meshes in the field of image processing technology. In the method, the thought of graded registration is adopted to implement overall registration on a standard image to be registered and a reference image; then the angular points of the images are used to restrain the interested regions of the overall registration image and the reference image to generate irregular triangular meshes, and generate triangular units with smaller shapes and more quantity at places with larger deformation in theinterested region of the image; the triangular meshes generated by the triangular units with larger shapes and less quantity at places with smaller deformation in the interested region of the image are changed according to the changes of the image contents; and finally, the thought of near-rigid registration is used for realizing local accurate registration of the images. The nonrigid medical image registration method based on self-adapting triangular meshes can effectively improve the accuracy of nonrigid medical image registration and enhance the noise resisting capacity of registration.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to a non-rigid medical image registration method based on an adaptive triangular grid. Background technique [0002] Medical image registration is the basis of medical image fusion and an important technology of medical image analysis. It mainly searches for a spatial transformation to make the corresponding points on two medical images completely consistent in spatial position or anatomical structure. sex. In the process of medical diagnosis, there are practical problems such as the physical mechanism of different modes of images showing different properties, patient movement, changes in imaging parameters, and different resolutions of different imaging equipment. The spatial alignment of images in different modes is subject to many limitations, and often has a high degree of subjectivity, which inevitably leads to errors. Especially in application fields suc...

AI Technical Summary

Problems solved by technology

[0004] The technical defect of the medical image registration method based on image features is that it needs to segment the image to extract the features of the image. Because the structure of non-rigid tissue is very complex, some interfaces are not very obvious, and it usually requires manual pre-selection of features. It is time-consuming and labor-intensive, and the accuracy of registration is affected by the accuracy of segmentation. It is generally difficult to complete automatically, which makes the registration time process slow and the registration accuracy is not high.

[0005] The technical defect of the medical registration method based on image grayscale information is that it does not need to segment the image, and directly calculates the entire image, which will result in slow registration speed, long registration time, and poor robustness.

[0006] The traditional Harris method needs to artificially set a certain threshold when extracting the corner points of the image as a constraint condition for extracting the corner points, which will cause the number of corner points to be extracted when the threshold of the image is increased, and the corner points to be extracted when the threshold is lowered. The large number of points makes the distribution of extracted corner points uneven and prone to clustering and redundancy

This method cannot automatically select the corner points of the image according to the essential properties of the image

[0007] After searching the existing technologies, it was found that Wang Wei, Tang Yiping, Shi Bingchuan, etc. published an article "An Improved Harris Corner Extraction Method" in the Journal of Optical Precision Engineering in October 2008. The method described in this article is in the corner When extracting points, it is necessary to set a certain threshold according to the experimenter's previous experimental experience, which is time-consuming and laborious, and the distribution of extracted corner points is uneven, and it is easy to miss corner points or generate false corner points

Images