Excitation fluorescence tomography method based on GCN residual connection network

Abstract

The invention belongs to the field of biomedical molecular imaging, particularly relates to an excitation fluorescence tomography method, system and equipment based on a GCN residual connection network, and aims to solve the problems of model precision reduction, reconstruction precision reduction and slow reconstruction speed in the traditional FMT reconstruction based on a photon propagation model. The method comprises the following steps: meshing segmented CT image data of an organism, and carrying out graph structure modeling; simulating a photon propagation process of an in-vivo light source in the organism to obtain fluorescence distribution on the surface and in the organism, and expanding the fluorescence distribution as a light source sample; constructing a first node set; inputting the expanded light source sample and each node in the first node set into a deep learning network model, and training the model; and utilizing the trained deep learning network model to carry out excitation fluorescence tomography reconstruction on the organism. According to the invention, excitation fluorescence tomography with high reconstruction quality and high reconstruction speed is realized.

Description

technical field [0001] The invention belongs to the field of biomedical molecular imaging, and in particular relates to an excited fluorescence tomography method, system and equipment based on a GCN residual connection network. Background technique [0002] Fluorescence Molecular Imaging (FMI) is an important optical molecular imaging technology, and it is also an international research hotspot. Compared with other medical imaging technologies, FMI imaging technology has millimeter-level spatial resolution, and has the advantages of high detection sensitivity, high signal intensity, and easy clinical transformation. FMI imaging technology is used to detect specific optical signals on the surface of biological tissues. Detection can infer the approximate distribution area of ​​fluorescent probes in biological tissues. However, limited by the absorption and scattering of photons propagating in biological tissues, it is difficult for FMI imaging technology to locate the depth ...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problems of model accuracy decrease, reconstruction accuracy decrease, and reconstruction speed slow when traditionally based on the photon propagation model for FMT reconstruction, the first aspect of the present invention proposes a GCN-based A method for excited fluorescence tomography of a residual connectivity network comprising:

Images