Dynamic mri image reconstruction method and device

Abstract

Embodiments of the present invention provide a dynamic MRI image reconstruction method and device. The method includes: establishing a first dynamic nuclear magnetic resonance image reconstruction model, the first dynamic nuclear magnetic resonance image reconstruction model including the original dynamic nuclear magnetic resonance image reconstruction model and a third-order tensor corresponding to the dynamic nuclear magnetic resonance image as an independent variable Low-rank constraints and sparse constraints; solve the first dynamic nuclear magnetic resonance image reconstruction model according to the first preset rule, and obtain the third-order tensor corresponding to the dynamic nuclear magnetic resonance image. The method uses the dynamic nuclear magnetic resonance image As a third-order tensor, combined with the two prior conditions of low rank and sparseness to regularize the image, it fully maintains the high-dimensional structural characteristics of dynamic MRI images, and at the same time makes full use of the image tensor structure. Intrinsic connection and sparsity between image sequences, thus improving the quality of image reconstruction.

Description

technical field [0001] The present invention relates to the field of image reconstruction, in particular to a dynamic MRI image reconstruction method and device. Background technique [0002] Magnetic resonance imaging (MRI) technology plays a very important role in modern medical treatment due to its high resolution, no radiation, and no harm to the human body. However, its disadvantage is the slow imaging speed, which seriously affects its performance in clinical applications. In order to improve the imaging speed of MRI, based on the compressed sensing theory, image reconstruction can be realized by undersampling the k-space data. Since it is an ill-conditioned inverse problem to reconstruct the original image using a small amount of under-sampled data, we need to add a priori condition of the ideal signal to regularize the problem. Similar to video files, dynamic MRI images are composed of background elements with low-rank characteristics and dynamic elements with spar...

AI Technical Summary

Problems solved by technology

Similar to video files, dynamic MRI images are composed of background elements with low-rank characteristics and dynamic elements with sparse characteristics. Some researchers have proposed to combine sparse and low-rank characteristics to reconstruct the model (so that the solution algorithm combining the two characteristics is also the same. can be applied to CT and ECT), such as k-t SLR, but most of these methods vectorize and arrange image data into matrix operations, which do not maintain the high-dimensional structural characteristics of dynamic MRI images, and do not make full use of image sequences The sparsity between, so the reconstruction quality of the image is limited

Images