Accurate lumbar spine segmentation is essential for diagnosing and treating spinal disorders. Existing methods often struggle to balance accuracy and computational efficiency, particularly with multimodal images such as MRI and CT. To address this, we propose a deep learning framework using a three-step approach: the "Balanced E-Net" detects lumbar regions, SpatialConfigurationNet 3D annotates vertebrae centroids, and HyperDenseNet performs precise segmentation. A novel gradient-sensitivity-based pruning method optimizes HyperDenseNet, enhancing efficiency by removing redundant filters. Tested on MRI and CT datasets, the framework achieves dice, IoU, and Hausdorff scores of 95.42%, 95.68%, and 4.51mm, outperforming current methods and streamlining clinical workflows.

Spine CT image reconstruction and lesion classification are crucial in diagnosing spine disorders, supporting treatment through automated lesion detection. Leveraging advancements in machine learning, this paper introduces a novel approach using transfer learning and a Quantum Convolutional Neural Network (QCNN) to improve accuracy in spine image analysis. First, the study applies MobileNetv2 and ResNet50-based transfer learning models for spine image reconstruction, comparing their results to select the best-performing model. For lesion classification, a QCNN model is proposed, with its outputs passed into a traditional neural network comprising convolutional, max-pooling, and residual layers. The proposed methods outperform prior models and other pre-trained machine learning classifiers, demonstrating enhanced accuracy and efficiency in spine lesion classification. This approach offers valuable support for radiologists, suggesting an efficient automated solution for identifying and classifying spinal lesions in CT images.

Driver behavior detection is a crucial aspect of enhancing road safety and preventing accidents by identifying and classifying driving actions in real-time. This paper presents "Efficient-MobileNet," a deep learning-based model for automated driver behavior detection. The proposed model employs a feature fusion approach, leveraging EfficientNetB7 for feature extraction from input data, combined with Atrous Spatial Pyramid Pooling (ASPP), attention blocks, and residual blocks to enhance the model's performance. ASPP captures multi-scale contextual information, critical for detecting driver actions of varying sizes and positions within an image or video frame. Attention blocks improve the model's focus by emphasizing relevant features, such as hand movements or facial expressions, while ignoring irrelevant background details. Residual blocks address the vanishing gradient problem, enabling the model to learn complex behavior patterns by maintaining information flow through identity mappings. The fused features are passed through MobileNetV3 for final classification, resulting in a robust and accurate system capable of reliably classifying both safe and risky driving behaviors. Extensive experiments on a real-world dataset demonstrate the effectiveness of the proposed model, outperforming existing methods in terms of accuracy, precision, recall, f1 score, and AUC of 99.62%, 99.91%, 99.96%, 99.93%, and 1.00% respectively.

The deep coupling within the interdependent networks may cause a series of serious network security problems, and even lead to the collapse of the networks completely. In order to study the influence of the coupling on the interdependent networks, this paper constructed the model of interdependent networks and analyzed its robustness, which is used to evaluate the security of interdependent networks. Firstly, inspired by the logistics-warehouse, we proposed three kinds of interdependent networks models: positive correlation model, negative correlation model and random correlation model, which are based on closeness centrality of the nodes. Secondly, we analyzed the impact of different attack strategies, dependent ways, dependent probability on the robustness. Experimental results show that the three models are more sensitive to deliberate attacks; the robustness of the negative correlation model is the strongest; and the robustness can be enhanced by appropriately increasing the dependent probability. Finally, according to the experimental results, we put forward some defense strategies on constructing interdependent networks and maintaining its security.

Summary Desktop virtualization, which is to make the desktop virtual so that users can access any application through the network with any devices at any time and any place, is being widely used now as an emerging trend. However, it is important to further improve this approach while maintaining good user experience. Though the simple protocol for independent computing environments as a virtual desktop solution can achieve a user experience similar to an interaction with a local machine, there are still many deficiencies in it. For instance, it cannot apply to environment of high controllability of user, and the quality of graphic interactive experience is to be improved. In this paper, to meet QoE requirements of user, we build a feasible file transfer and sharing mechanism and propose a graphics subsystem optimization strategy based on simple protocol for independent computing environments (SPICE), namely Transparent Desktop. We also verify that the Transparent Desktop can provide users with ubiquitous desktop services of higher efficiency, stronger user‐controllability and better QoE through the experimental. Copyright © 2016 John Wiley & Sons, Ltd.

Radio Frequency Identification (RFID) technology that is used in pervasive computing environments is widely concerned in recent years. At present, RFID technology is mainly applied in logistics, manufacturing, public service industry and so on. Based on the characteristics of pervasive computing environment, in this paper, a RFID context-aware service model is proposed. The model combines Ontology Language OWL with CC/PP and FOAF specifications to build a context model. Context ontology includes domain ontology and RFID tags ontology. On this basis, the context reasoning and interpretation mechanism based on ontology and rule is introduced. Through the rewritable RFID tag memory, dynamic management of the context in RFID system is realized. Finally, examples are used to demonstrate the feasibility of the model.

This paper presents a plug-in based integrated development framework for industrial control system named P-IDP4ICS which adopts the role-based access control technique, named RBAC. And the implementation mechanisms of P-IDP4ICS are explained in this paper. The P-IDP4ICS framework consists of some components, such as Extension Protocol, Kernel, Plug-in Registry and Host Application. Among those components, Extension Protocol component can keep the industrial control software customizable and scalable, and the reusing plug-ins which have been well developed like Host Application plug-in can improve reusing capability of industrial control software. The plug-ins in P-IDP4ICS are presented in the form of DLLs, which protect the source code. Furthermore, P-IDP4ICS which implements the RBAC model reduces the complexity of authorization management and the costs of system maintenance. Finally, this paper reconstructs a Sinter Integrated Control Expert System through P-IDP4ICS framework and has proved its effectiveness.