X-ray CT local high-resolution imaging method and device based on deep learning

Abstract

The invention discloses an X-ray CT local high-resolution imaging method and device based on deep learning, and the method comprises the steps: carrying out the enhancement of an initial estimation image after the projection data of a to-be-imaged region is pre-reconstructed, and constructing an image basic data set of the to-be-imaged region; pre-reconstructing projection data outside the to-be-imaged area to construct a basic background image data set; image data are randomly selected from the two data sets to be combined into an object to be imaged, and the object to be imaged is scanned and simulated to obtain simulated local scanning data; and matching the simulation local scanning data with corresponding data in the to-be-imaged area image basic data set to form training data, training the local area deep learning reconstruction network by using the training data, and performing high-resolution local area reconstruction by using the trained local area deep learning reconstruction network. Therefore, brand new local region high-resolution reconstruction combining deep learning and an imaging mechanism improves the quality and efficiency of the reconstructed image.

Description

technical field [0001] The present application relates to the technical field of radiation imaging, in particular to a deep learning-based X-ray CT local high-resolution imaging method, device, electronic equipment and storage medium. Background technique [0002] X-ray CT imaging systems are widely used in medical, security inspection, industrial non-destructive testing and other fields. Theoretically, from the line integral value of the physical characteristic parameters in each direction of a certain section of the object, the distribution of the physical characteristic parameters of the section can be calculated. In the CT system, the ray source and the detector surround the object, and collect projection data in various directions according to a certain orbit, and the distribution of the linear attenuation coefficient of the object in three-dimensional space can be calculated through the image reconstruction algorithm. [0003] Traditional CT reconstruction methods mai...

AI Technical Summary

Problems solved by technology

However, this type of method has a large amount of computation, and it is difficult to restore the high-frequency components in the reconstruction area well, so that the reconstruction result cannot reach the theoretical spatial resolution.

On the other hand, in a practical system, it is difficult to simultaneously obtain the paired data of the ROI and the high-resolution scan outside the ROI, which brings certain challenges to the deep learning reconstruction method.

Images