Excited fluorescence tomography based on multilayer perceptual network

Abstract

The invention provides an excited fluorescence tomography method based on a multilayer perceptual network. The method comprises the following steps: S1, generating a training sample; S2, setting a meshless standardized model and mapping the training samples into the meshless standardized model; S3, constructing a multi-layer perceptual network of excited fluorescence tomography according to a meshless standardized model, wherein that multi-layer perceptual network comprises an input layer, a hidden lay and an output layer; S4, training the meshless standardized model according to the output result of the output layer; and S5, inputting the data of the organism into the trained meshless standardized model to obtain a reconstructed image of the organism.

Description

Technical field [0001] The invention relates to the field of biomedical molecular imaging, in particular to an excitation fluorescence tomography imaging method. Background technique [0002] Fluorescence Molecular Imaging (FMI) technology is an emerging molecular imaging technology in recent years. Use targeted excitation fluorescent molecular probes to label specific cells or molecules in biological tissues, and then excite the molecular probes by external light to generate red-shifted emission fluorescence, and collect the emitted fluorescence that propagates to the surface of the biological tissues, and use computer imaging Technology to obtain fluorescent images. Compared with other optical molecular imaging technologies, FMI technology has the characteristics of a wide variety of probes, high signal intensity, rich collection of information, and real-time in vivo imaging. [0003] Fluorescence Molecular Tomography (FMT) is a three-dimensional imaging of the distribution of ...

AI Technical Summary

Problems solved by technology

However, the traditional FMT forward problem model is complex to solve, and needs to go through multiple partial differential equation solving and spatial discretization processes, which reduces the positioning accuracy and imaging speed of FMT imaging

At the same time, since the measured fluorescence data is much smaller than the data volume of the imaging space, it is difficult to obtain an accurate solution.

Images