Threshold optimization method and device based on K edge imaging, equipment and medium

Abstract

The embodiment of the invention discloses a threshold optimization method and device based on K edge imaging, equipment and a medium. The method comprises the following steps of: pre-scanning a to-be-detected object by a photon counting detector according to a preset threshold group to obtain first detection data; according to the first detection data, determining the line integral value of the decomposition coefficient of the basis function combination in each ray, wherein each ray is a fixed basis function combination; in the linear integral values of the decomposition coefficients of the basis function combination in each ray, screening out the rays of which the linear integral values of the contrast agent material basis functions of the basis function combination are greater than a preset linear integral threshold value; inputting the screened detection data corresponding to each ray into a pre-trained neural network model to obtain a screened local optimal threshold group of eachray; and determining the overall optimal threshold group of K edge imaging of the photon counting detector according to the local optimal threshold group.

Description

technical field [0001] The invention relates to the field of spectral electron computer tomography (Computed Tomography, CT), in particular to a K-edge imaging-based optimization threshold method, device, equipment and medium. Background technique [0002] For spectral K-edge (K-edge) imaging, photon-counting detectors are required. However, once a photon-counting detector is used, it is necessary to determine the threshold of the photon-counting detector in order to acquire a higher quality image. [0003] For the threshold selection method, there are currently two empirical methods: one is to perform energy equalization in the entire energy spectrum range, that is, to make the width of each energy window roughly the same. The other is to equally divide the number of photons, that is, to make the number of photons in each energy window roughly the same, but this method requires a known energy spectrum. [0004] Neither of the above two methods can select the optimal thres...

AI Technical Summary

Problems solved by technology

[0004] Neither of the above two methods can select the optimal threshold of the photon counting detector when the composition of the object to be detected is unknown.

Images