A TV-Based 3D Pet Image Reconstruction Method

Abstract

The invention discloses a TV-based three-dimensional PET image reconstruction method, comprising: (1) obtaining 3D projection data by using the 3D acquisition mode of PET; (2) recombining the 3D projection data into a series of 2D projection data; Calculate the frequency domain distribution of the reconstructed image from the projection data; (4) Add the total variation operator to the direct Fourier transform model to establish the objective function; (5) Use the BOSVS algorithm to solve the above model to obtain the 2D reconstruction result; (6) ) to combine all 2D reconstruction results into a 3D reconstruction image. The present invention uses total variation as a regular term to establish a model on the basis of direct Fourier transform, uses BOSVS algorithm to iteratively solve, obtains a reconstructed image with higher resolution and clearer edges, and ensures fast reconstruction of direct Fourier transform The quality of the reconstructed image is significantly improved on the basis of , and at the same time, good results can still be obtained in the case of low count rate.

Description

technical field [0001] The invention belongs to the technical field of PET imaging, and in particular relates to a TV-based three-dimensional PET image reconstruction method. Background technique [0002] PET (Positron Emission Tomography) is positron emission computed tomography, as an advanced technology in the development of nuclear medicine, it has become a very important inspection method for cancer diagnosis, heart function detection and brain function imaging. It utilizes the advantages of many new technologies such as nuclear physics and medical imaging, and can observe the activities of cell metabolism at the molecular level. PET imaging technology is to inject a radioactive isotope that can directly or indirectly reflect the biological metabolic process in the living body, and receive the photon pairs generated by the annihilation through the detector ring in the PET device, and then calculate and determine the annihilation (emission) position of the positron. Fin...

AI Technical Summary

Problems solved by technology

The 3D acquisition mode removes the limitation of the metal partition between the detectors in the 2D mode, and considers the coincidence counts in all detector rings, which greatly improves the count rate of the projection data and the sensitivity of the detector, but in the 3D acquired data Contains more scattering, which requires a lot of calculation consumption and storage requirements for reconstruction and other processing

Due to the large amount of data collected and the slow reconstruction speed of 3D PET, how to improve the reconstruction speed of 3D PET while ensuring the accuracy of reconstructed images has become one of the current research hotspots.

Images