Hierarchical Hybrid Networks for Automatic Pulmonary Blood Vessel Segmentation in Computed Tomography Images.

Pulmonary arterial hypertension (PAH) is considered the third most common cardiovascular disease after coronary heart disease and hypertension. The diagnosis of PAH is mainly based on the comprehensive judgment of computed tomography and other medical image examinations. Medical image processing based on deep learning has achieved significant success. However, the data belongs to the patient's privacy; therefore, the medical institutions as data custodians have the responsibility to protect the security of their data privacy. This situation makes medical institutions face a dilemma when building data-driven deep learning-assisted medical diagnosis methods. On the one hand, they need to pursue more high-quality data based on Big Data architecture for deep learning; on the other hand, they need to protect patient privacy to avoid data leakage. In response to the above challenges, we propose a hierarchical hybrid automatic segmentation model for pulmonary blood vessels based on local learning and federated learning approaches for segmenting the pulmonary blood vessels. The experiments prove the proposal could automatically segment the vessels from the original CT. It also indicates that the model based on a federated learning approach can achieve impressive performance under the premise of protecting data privacy for Big Data.

Forestry Digital Twin With Machine Learning in Landsat 7 Data.

Forest succession analysis can predict forest change trends in the study area, which provides an important basis for other studies. Remote sensing is a recognized and effective tool in forestry succession analysis. Many forest modeling studies use statistic values, but only a few uses remote sensing images. In this study, we propose a machine learning-based digital twin approach for forestry. A data processing algorithm was designed to process Landsat 7 remote sensing data as model's input. An LSTM-based model was constructed to fit historical image data of the study area. The experimental results show that this study's digital twin method can effectively forecast the study area's future image.

Intelligent malware detection based on graph convolutional network.

Malware has seriously threatened the safety of computer systems for a long time. Due to the rapid development of anti-detection technology, traditional detection methods based on static analysis and dynamic analysis have limited effects. With its better predictive performance, AI-based malware detection has been increasingly used to deal with malware in recent years. However, due to the diversity of malware, it is difficult to extract feature from malware, which make malware detection not conductive to the application of AI technology. To solve the problem, a malware classifier based on graph convolutional network is designed to adapt to the difference of malware characteristics. The specific method is to firstly extract the API call sequence from the malware code and generate a directed cycle graph, then use the Markov chain and principal component analysis method to extract the feature map of the graph, and design a classifier based on graph convolutional network, and finally analyze and compare the performance of the method. The results show that the method has better performance in most detection, and the highest accuracy is 98.32 % , compared with existing methods, our model is superior to other methods in terms of FPR and accuracy. It is also stable to deal with the development and growth of malware.

Regulation mechanism of three key parameters on catalytic characterization of molybdenum modified bimetallic micro-mesoporous catalysts during catalytic fast pyrolysis of enzymatic hydrolysis lignin.

Novel preparation of molybdenum modified bimetallic micro-mesoporous catalyst was proposed innovatively to conduct catalytic fast pyrolysis of enzymatic hydrolysis lignin. The optimal catalytic characterization of the prepared catalyst was attributed to appropriate porous structure, the interaction between zeolite support and metal species, and the synergetic and stable mechanism of bimetallic active sites. With the incorporation of metal species into micro-mesoporous catalyst, the distribution of active sites experienced a regulation, which contributed to MAHs production and cracking of oxygen-containing substances. NiMo/AZM catalyst exhibited the most obvious coke inhibition effect (8.47 wt% of mass yield) and converted more high-ordered graphite carbon to low-ordered one, so as to make it easier to remove and prolong the catalyst lifetime, and obtained the highest mass yield of MAHs (13.15 wt%) as well as the minimum selectivity of bulky oxygenates (3.82%), which was the joint contribution of three key parameters.

An Edge Based Multi-Agent Auto Communication Method for Traffic Light Control.

With smart city infrastructures growing, the Internet of Things (IoT) has been widely used in the intelligent transportation systems (ITS). The traditional adaptive traffic signal control method based on reinforcement learning (RL) has expanded from one intersection to multiple intersections. In this paper, we propose a multi-agent auto communication (MAAC) algorithm, which is an innovative adaptive global traffic light control method based on multi-agent reinforcement learning (MARL) and an auto communication protocol in edge computing architecture. The MAAC algorithm combines multi-agent auto communication protocol with MARL, allowing an agent to communicate the learned strategies with others for achieving global optimization in traffic signal control. In addition, we present a practicable edge computing architecture for industrial deployment on IoT, considering the limitations of the capabilities of network transmission bandwidth. We demonstrate that our algorithm outperforms other methods over 17% in experiments in a real traffic simulation environment.

Relationship between arterial atheromatous plaque morphology and platelet-associated miR-126 and miR-223 expressions.

OBJECTIVE: To study the expression of miR-126 and miR-223 in platelet of rabbit arterial plaque models, and explore its correlation with plaque morphology. METHODS: Rabbit arterial plaque models were established, peripheral blood of models and control animals was collected. Plaque morphologies were divided into type I, type II and type III based on angiography plaque morphology and Ambrose method. Platelet isolation kit was applied to isolate and purify peripheral blood platelets, CD45 immunomagnetic beads were used to remove the residual white blood cells. The miRNAs of platelets was extracted by miRNA Isolation Kit, and expressions of miR-126 and miR-223 of the platelets samples were detected by Real-time PCR. The correlation between plaque morphology and platelet-associated miR-126 and miR-223 expressions were analyzed. Expressions of target gene VCAM-1 and P2Y12 receptors of miR-126 and miR-223 in the atherosclerosis plaque of rabbit model were detected by Western blot. RESULTS: Relative expression levels of miR-126 and miR-223 in the model group were 0.27±0.10 and 0.71±0.14, respectively. Plaque morphology was divided into types I, II and III; and miR-126 and miR-223 expression levels were detected in each type. Expression levels of miR-126 in each type were 0.42±0.07, 0.17±0.11 and 0.22±0.15, respectively; and expression levels of miR-223 in each type are 0.68±0.02, 0.57±0.06 and 0.88±0.10, respectively. Relative to the control group, miR-126 and miR-223 known target genes in VCAM-1 and P2Y12 receptors increased platelets in rabbit atherosclerotic plaque models (P<0.05). CONCLUSIONS: Relative to normal control animals, miR-126 and miR-223 platelets were reduced in the rabbit atherosclerotic plaque model group (P<0.05). In the type II plaque morphology group, miR-126 was greatly reduced; and there is no significant correlation between miR-223 and plaque morphology.

Correlation between eigenvalue spectra and dynamics of neural networks.

This letter presents a study of the correlation between the eigenvalue spectra of synaptic matrices and the dynamical properties of asymmetric neural networks with associative memories. For this type of neural network, it was found that there are essentially two different dynamical phases: the chaos phase, with almost all trajectories converging to a single chaotic attractor, and the memory phase, with almost all trajectories being attracted toward fixed-point attractors acting as memories. We found that if a neural network is designed in the chaos phase, the eigenvalue spectrum of its synaptic matrix behaves like that of a random matrix (i.e., all eigenvalues lie uniformly distributed within a circle in the complex plan), and if it is designed in the memory phase, the eigenvalue spectrum will split into two parts: one part corresponds to a random background, the other part equal in number to the memory attractors. The mechanism for these phenomena is discussed in this letter.