Aiming at the current high delay rate of the civil aviation transportation industry, and the fact that the high-precision delay prediction problem can hardly be solved by traditional algorithms, a randomly connected Clique Network (CliqueNet) based flight delay prediction model was proposed. Firstly, the flight data and related weather data were fused by the model. Then, making full use of the improved network model to extract features from the fused dataset. Finally, the softmax classifier was used to predict the flight departure delay of all levels with high precision. The main features of the model include random connection of clique feature layers and the introduction of Channel-wise and Spatial Attention Residual (CSAR) block to the transition layer. The former transmits the feature information in a more effective connection; and the latter double-calibrates the feature information on the channel and spatial dimensions to improve accuracy. Experimental results show that the prediction accuracy of the fused data is improved by 0.5% and 1.3% respectively with the introduction of random connection and CSAR block, and the final accuracy of the new model reaches 93.40%.

An automatic segmentation method based on image patch matching strategy was proposed to realize the automatic segmentation of glomerular basement membrane automatically. First of all, according to the characteristics of the glomerular basement membrane, the search range was extended from a reference image to multiple reference images, and an improved searching method was adopted to improve matching efficiency. Then,the optimal patches were searched out and the label image patches corresponding to the optimal patches were extracted, which were weighted by matching similarity. Finally, the weighted label patches were rearranged as the initial segmentation of glomerular basement membrane, from which the final segmentation could be obtained after morphological processing. On the glomerular Transmission Electron Microscopy (TEM) dataset, the Jaccard coefficient is between 83% and 95%. The experimental results show that the proposed method can achieve higher accuracy.

Due to the high radiation dose to the patient when acquiring lung four Dimensional Computed Tomography (4D-CT) data, this paper proposed a method for deriving the phase-binned 4D-CT image sets through deformable registration of the images acquired at some known phases. First, Active Demons registration algorithm was employed to estimate the motion field between inhale and exhale phases. Then, images at an intermediate phase were reconstructed by a linear interpolation of the deformation coefficients. The experiment results showed that the images at intermediate phases could be reconstructed efficiently. The quantitative analysis of landmark point displacements showed that 3 mm accuracy was achievable. The different maps of reconstructed and acquired images illustrated the similar level of success. The proposed method can accurately reconstruct images at intermediate phases of lung 4D-CT data.

A new method was proposed to accurately detect and quantitatively evaluate the lung nodule spiculation. First, the region growing method followed by level set method was used to accurately segment the main part of the lung nodule. Then, spiculated lines connected to the nodule boundary were extracted using a line detector in polar coordinates system. Finally, spiculation index was introduced as the quantitative measurement of spiculation features, which was then used as a criteria for distinguishing between spiculated and non-spiculated nodules. The consistency and correlation of spiculation index of the method and Lung Image Database Consortium (LIDC) were evaluated in detail. The experimental results show that the proposed method can effectively detect and quantitatively describe the lung nodule spiculation in CT images.

In response to the issue of mobile terminal heterogeneity in the existing Peer-to-Peer (P2P) data sharing network, an energy-aware P2P data sharing mechanism in heterogeneous mobile terminal named EADS was proposed, which enabled terminals to determine the types of end-user devices with introducing an energy-aware module to predict residual energy of these end-user devices. On this basis, the mechanism adjusted data sharing strategy dynamically in accordance with the changes of network environment. The simulation results demonstrate that EADS achieves significant performance improvement, in terms of energy utilization efficiency, balance load, as well as data sharing time, thus enhancing the success rate of data distribution. On the premise of maintaining high availability of files, average energy consumption has been reduced up to 15%.