X-space magnetic particle imaging deconvolution method

Abstract

The invention discloses an X-space magnetic particle imaging deconvolution method. The method comprises the following steps: collecting an X-space original image; a neural network structure is set, convolution and deconvolution layers are set, and a jump connection layer is arranged between the corresponding convolution layer and deconvolution layer; neural network training: network input is a simulated original image, a corresponding original clear image is used as a label to train a neural network, and a mean square error is adopted as a loss function; neural network detection: selecting three indexes of a root-mean-square error, a peak signal-to-noise ratio and a structural similarity index measure to carry out quantitative evaluation on image quality, and modifying network training parameters according to an evaluation result, so that a reconstructed image is closer to an original image; and X-space deconvolution: inputting a to-be-deconvolved original image into the trained and detected neural network model for prediction to obtain a deconvolution result. According to the method, the influence of system noise on the reconstruction process is greatly reduced, and the resolution of the magnetic particle imaging system for X-space reconstruction is improved.

Description

technical field [0001] The invention belongs to the field of magnetic particle imaging, and in particular relates to an X-space magnetic particle imaging deconvolution method. Background technique [0002] Magnetic nanoparticle imaging (MPI) is a tracer-based, functional, tomographic modality that can directly detect the spatial distribution of magnetic nanoparticles and shows great potential as a safe alternative to iodine or gadolinium contrast . Magnetic nanoparticles (MNPs) -- also known as superparamagnetic iron oxide nanoparticles (SPIOs) -- are commonly used as tracers. MPI can detect nanoparticle contrast agents without any background and without depth attenuation. In addition, MPI does not apply ionizing radiation, therefore, MPI is safe for patients and medical personnel if the applied magnetic field does not exceed safe limits. [0003] In the field of MPI image reconstruction, two reconstruction methods, system matrix and X-space, are mostly adopted at present...

AI Technical Summary

Problems solved by technology

[0003] In the field of MPI image reconstruction, two reconstruction methods, system matrix and X-space, are mostly used now. The system matrix method has a huge amount of calculation. Although the X-space method has a higher reconstruction speed, the resolution of the reconstructed image needs to be improved.

The original MPI image generated in X space is actually the convolution of magnetic particle concentration and point spread function (PSF). The existing X space reconstruction method directly uses the traditional deconvolution method, but the traditional deconvolution method requires Accurate estimation of the point spread function, which is difficult to measure in the experiment, the resulting error will lead to deconvolution of the image quality

And due to the ill-conditioned nature of the problem, the traditional deconvolution algorithm often introduces unpredictable noise, which has a great impact on the reconstruction results

Traditional non-blind deconvolution methods, such as Wiener filtering, rely too much on the convolution kernel, and the effect of deconvolution is not ideal

Images