Method, system and storage medium for image processing, analysis and segmentation of four-dimensional flow data

Abstract

The invention discloses a method, system and storage medium for image processing, analysis and segmentation of four-dimensional flow data. First read the four-dimensional flow data, collect and perform data reorganization processing; then the four-dimensional flow data data preprocessing is filtering, inversion, evaluation, linear operation, maximum density projection, and final operation processing, specifically including the three-dimensional image of the velocity component Filter processing, inverting the three-dimensional matrix, evaluating the velocity field image, performing linear operations on the inversion matrix and the amplitude result of the velocity field image, performing maximum density projection on the filter velocity merge matrix, and calculating the projection result matrix and The calculation and processing of the magnitude result of the velocity field image; finally, the image segmentation, and the data preprocessing results are input to the neural network to predict the segmentation. The method of the invention can greatly highlight the characteristics of the fluid flow area, and can more quickly and accurately perform neural network segmentation, avoiding manual operations and introducing errors, and greatly saving time.

Description

technical field [0001] The invention belongs to an image processing method, system and storage medium in the field of engineering technology, relates to post-processing analysis of digital images such as computed tomography (CT), magnetic resonance imaging (MRI), etc., and mainly relates to the flow area data analysis, and image segmentation of target regions therein. Background technique [0002] Among the four-dimensional flow data, the typical one is the four-dimensional flow magnetic resonance imaging technology (4DflowMRI). At present, the analysis of this technology is relatively slow, and it is mostly used in the field of medical analysis, such as cardiac MRI, aortic diagnosis, intracranial blood vessel diagnosis, etc. . In recent years, its research in aortic lesions is particularly prominent. In terms of its application, the similar software developed earlier on the market include Vitrea workstation of Siemens, CVI42 software of Circle Cardiovascular Imaging of Ca...

AI Technical Summary

Problems solved by technology

In terms of structure, due to the unclear positioning of the software, it uses a large number of classical morphological analysis in image processing, combined with a large number of manual operations, including but not limited to lasso, level set, region growing, etc. manual or semi-automatic Algorithms, so the extraction effect for complex 3D models is not good

[0005] In general, the above existing methods have the characteristics of cumbersome processes, relying on a large number of manual operations, and insufficient algorithm effects. In practical applications, the above points are currently widespread problems and relatively large challenges.

Images