No abstract

Positron emission tomography (PET) has a wide range of applications in the treatment and prevention of major diseases owing to its high sensitivity and excellent resolution. However, there is still much room for optimization in the readout circuit and fast pulse sampling to further improve the performance of the PET scanner. In this work, a LIGHTENING® PET detector using a 13 × 13 lutetium-yttrium oxyorthosilicate (LYSO) crystal array read out by a 6 × 6 silicon photomultiplier (SiPM) array was developed. A novel sampling method, referred to as the dual time interval (DTI) method, is therefore proposed to realize digital acquisition of fast scintillation pulse. A semi-cut light guide was designed, which greatly improves the resolution of the edge region of the crystal array. The obtained flood histogram shown that all the 13 × 13 crystal pixels can be clearly discriminated. The optimum operating conditions for the detector were obtained by comparing the flood histogram quality under different experimental conditions. An average energy resolution (FWHM) of 14.3% and coincidence timing resolution (FWHM) of 972 ps were measured. The experimental results demonstrated that the LIGHTENING® PET detector achieves extremely high resolution which is suitable for the development of a high performance time-of-flight PET scanner.

This paper obtains the numerical solutions of the elliptic solitons in a (1+2)dimensional anisotropic nonlocal nonlinear fractional Schrödinger equation, and verifies their stabilities by the direct propagation method. The results show that the properties of such solitons relatively depend on the Lévy index. Such as the soliton shape varies with the change of Lévy index. When the Lévy index decreases, the ellipticity will increase, while the critical power will decrease. Furthermore, we demonstrate the physical features exhibited by the higher order elliptic solitons for a different Lévy index.

In fault detection and the diagnosis of large industrial systems, whose chemical processes usually exhibit complex, high-dimensional, time-varying and non-Gaussian characteristics, the classification accuracy of traditional methods is low. In this paper, a kernel limit learning machine (KELM) based on an adaptive variation sparrow search algorithm (AVSSA) is proposed. Firstly, the dataset is optimized by removing redundant features using the eXtreme Gradient Boosting (XGBOOST) model. Secondly, a new optimization algorithm, AVSSA, is proposed to automatically adjust the network hyperparameters of KELM to improve the performance of the fault classifier. Finally, the optimized feature sequences are fed into the proposed classifier to obtain the final diagnosis results. The Tennessee Eastman (TE) chemical process is used to verify the effectiveness of the proposed method through multidimensional diagnostic metrics. The results show that our proposed diagnosis method can significantly improve the accuracy of TE process fault diagnosis compared with traditional optimization algorithms. The average diagnosis rate for 21 faults was 91.00%.

No abstract