Low-dose CT reconstruction method based on double-domain interleaving network

Abstract

The invention discloses a low-dose CT reconstruction method based on a double-domain interleaving network, and belongs to the field of medical image processing and computer vision. According to the method, a low-dose CT image and a corresponding sinogram are used as input, a distortion branch, a perception branch and a fusion branch are trained respectively in two stages, and learnable parameters are used for balancing distortion and perception; the method is simple in procedure and easy to implement, and reconstruction results better meeting diagnosis requirements of doctors can be obtained in an end-to-end mode; according to the invention, an adjustable reconstruction result can be generated through one network, the corresponding reconstruction result can be freely selected for the CT image with non-uniform noise distribution, and the lesion texture and organ structure of the CT image are protected while the overall perception effect is ensured.

Description

technical field [0001] The invention belongs to the technical fields of medical image processing and computer vision, and in particular relates to a low-dose CT reconstruction method based on a dual-domain interleaving network to realize adjustable reconstruction results. Background technique [0002] At present, computed tomography (CT, Computed Tomography) has become a common imaging method for auxiliary diagnosis. However, in order to ensure that CT images are clear enough, the radiation risk of high-dose radiation to the human body cannot be underestimated. Most of the equipment is designed to reduce the risk of cancer or genetic damage that may be brought about by high-dose radiation, making low-dose CT emerge as the times require. How to solve the stronger noise and more artifacts brought about by reducing the radiation dose has become a challenging problem. [0003] Existing methods can be summarized into the following three categories: (1) Filtering and reconstruct...

AI Technical Summary

Problems solved by technology

[0005] Since the existing method does not have a good combination for the sinogram, which makes the noise generated in the process of reconstructing the image from the sinogram cannot be well processed, and the result is not adjustable, the present invention overcomes the shortcomings of the prior art, Provides a low-dose CT reconstruction method based on a dual-domain interleaving network to achieve adjustable reconstruction results, and can generate reconstruction results with different denoising levels through one network

Images