Xidian University

Duofang Chen

Xueli Chen

Shouping Zhu

Jie Tian

Jimin Liang

Nan Wang

Pengfei Huang

Yonghua Zhan

Heng Zhao

Shenghan Ren

Honghao Cao

Hui Xie

Xiaochao Qu

Xu Cao

Yihan Wang

Fanzhen Meng

Jing Zhao

In a magnetic particle imaging reconstruction method based on a neural network constrained by a forward model, a system matrix is obtained through calibration, voltage data generated by a specimen is measured, collected data are transformed into a frequency-domain by Fourier transform, frequency features of data are screened by a signal-to-noise ratio threshold, a reconstruction network is built by a Pytorch to realize mapping from one-dimensional voltage data to a multi-dimensional magnetic particle concentration distribution, the system matrix is used as the forward model of magnetic particle imaging, and simulated voltage data is generated according to a reconstructed multi-dimensional magnetic particle concentration distribution, a difference between the simulated voltage data and input voltage data is calculated as a loss function for network parameter updating, a total variation regularization term is added to the loss function, and training parameters and regularization parameters are adjusted to achieve an optimal reconstruction effect.

Magnetic particle imaging reconstruction method based on neural network constrained by forward model

A breast diffusion optical tomography (DOT) device includes a light source, a detector, a mechanical driver and a contour acquisitor without using optical fiber and reduce the system complexity. The light source includes M1 number of multi-wavelength near-infrared light emitters for a continuous-wave mode and M2 number of laser diodes of different wavelengths for a frequency-domain mode. The detector includes N1 number of silicon photomultipliers for detecting near-infrared light in the continuous-wave mode and N2 number of silicon photomultipliers for detecting near-infrared light in the frequency-domain mode. The mechanical driver adjusts a distance between the light source and the detector. The contour acquisitor obtains the geometry of a patient's breast for reconstruction. The DOT device uses amplitude and phase information obtained in the frequency-domain mode to estimate initial optical parameters of tissues and then apply for reconstruction in the continuous-wave mode.

Non-fiber flat-panel breast diffusion optical tomography system

The invention discloses a two-mode Raman optical projection tomography system. Samples are irradiated by the laser beam after the beam being expanded by beam expander. Optical signal of each mode will be separated by the beam splitter. Sparse sampling method is used for signal collection. Optical transmission projection signal acquisition module collects transmitted light of samples to form optical projection image. Multi-spectral Raman scattering signal acquisition module collects Raman scattering light produced by samples. Background noise is removed from the collected data. Sparse sampling data are reconstructed by using algebraic reconstruction method (ART) based on TV minimization. The three-dimensional structure image obtained by reconstruction and the three-dimensional chemical compositions image are fused to obtain the three-dimensional volume image with multiple information.

Two-mode Raman optical projection tomography system

A system for the endoscopic X-ray luminescence computed tomographic imaging is provided. The system comprises an excitation source unit () configured to excite a probe inside a detecting object from an external excitation source; a signal collecting unit () configured to collect X-ray images of the detecting object and near-infrared light signal images emitting from the probe being excitated; a computing unit () configured to preprocess the collected X-ray images and the near-infrared light signal images, execute sparse reconstruction on the preprocessed X-ray images so as to obtain anatomical structure information of the detecting object and location information of an endoscopic detector in the signal collecting unit () inside the detecting object, and use the obtained anatomical structure information and the location information to execute three-dimensional optical reconstruction based on the preprocessed near-infrared light signal images so as to obtain localization and distribution of an imaging target inside the detecting object.

System and method for endoscopic X-ray luminescence computed tomographic imaging

Growing community of inventors

Xi'an, China

Duofang Chen