With the widespread application of Industrial Internet of Things (IIoT) technology in the industry, the security threats are also increasing. To ensure the safe sharing of resources in IIoT, this paper proposes a data security sharing model based on privacy protection (DSS‐PP) for blockchain‐enabled IIoT. Compared with previous works, DSS‐PP has obvious advantages in several important aspects: (1) In the process of identity authentication, it protects users' personal information by using authentication technology with hidden attributes; (2) the encrypted shared resources are stored in off‐chain database of the blockchain, while only the ciphertext index information is stored in the block. It reduces the storage load of the blockchain; (3) it uses blockchain logging technology to trace and account for illegal access. Under the hardness assumption of Inverse Computational Diffe–Hellman (ICDH) problem, this model is proven to be correct and safe. Through the analysis of performance, DSS‐PP has better performance than the referred works.

Background: Premature ovarian insufficiency (POI) is one of the most common causes of female infertility and the etiology is highly heterogeneous. Most cases are idiopathic and the pathogenesis remains unclear. Previous studies proved that the immune system plays a crucial role in POI. However, the precise role of immune system remains unclear. This study aimed to analyze the characteristics of peripheral blood mononuclear cells (PBMC) from patients with POI by single-cell RNA sequencing (scRNA-seq) and to explore the potential involvement of immune response in idiopathic POI.Methods: PBMC was collected from three normal subjects and three patients with POI. PBMC was subjected to scRNA-seq to identify cell clusters and differently expressed genes (DEGs). Enrichment analysis and cell-cell communication analysis were performed to explore the most active biological function in the immune cells of patients with POI.Results: In total, 22 cell clusters and 10 cell types were identified in the two groups. Compared with normal subjects, the percentage of classical monocytes and NK cells was decreased, the abundance of plasma B cells was increased, and CD4/CD8 ratio was significantly higher in POI. Furthermore, upregulation of IGKC, IFITM1, CD69, JUND and downregulation of LYZ, GNLY, VCAN, and S100A9 were identified, which were enriched in NK cell-mediated cytotoxicity, antigen processing and presentation, and IL-17 signaling pathway. Among them, IGHM and LYZ were respectively the most significantly upregulated and downregulated genes among all cell clusters of POI. The strength of cell-cell communication differed between the healthy subjects and patients with POI, and multiple signaling pathways were assessed. The TNF pathway was found to be unique in POI with classical monocytes being the major target and source of TNF signaling.Conclusions: Dysfunction of cellular immunity is related to idiopathic POI. Monocytes, NK cells, and B cells, and their enriched differential genes may play a role in the development of idiopathic POI. These findings provide novel mechanistic insight for understanding the pathogenesis of POI.

MultiModal Remote Sensing Object Detection (MM-RSOD) holds great promise for around-the-clock application. However, it faces challenges in effectively extracting complementary features due to the modality inconsistency and redundancy. Inconsistency can lead to semantic-spatial misalignment, while redundancy introduces uncertainty that is specific to each modality. To overcome these challenges and enhance complementarity exploration and exploitation, this paper proposes Dual Dynamic Cross-modal Interaction Network (DDCINet), a novel framework comprising two key modules: Dual Dynamic Cross-modal Interaction (DDCI) module and Dynamic Feature Fusion (DFF) module. The DDCI module simultaneously addresses both modality inconsistency and redundancy by employing a collaborative design of the Channel-gated Spatial Cross Attention (CSCA) and Cross-Modal Dynamic Filters (CMDF) on evenly segmented multimodal features. The CSCA component enhances the semantic-spatial correlation between modalities by identifying the most relevant channel-spatial features through cross attention, addressing modality inconsistency. In parallel, the CMDF component achieves cross-modal context interaction through static convolution and further generates dynamic spatial-variant kernels to filter out irrelevant information between modalities, addressing modality redundancy. Following the improved feature extraction, DFF module dynamically adjusts interchannel dependencies guided by modal-specific global context to fuse features, achieving better complementarity exploitation. Extensive experiments conducted on three MM-RSOD datasets confirm the superiority and generalizability of the DDCINet framework. Notably, our DDCINet, based on the RoI Transformer benchmark and ResNet50 backbone, achieves 78.4% mAP50 on the DroneVehicle test set and outperforms state-of-the-art methods by large margins.

The support vector machine (SVM) is verified to be effective for predicting cyber security situations, however, the long training time of the prediction model is a drawback to its use. To address this, a cyber security situation prediction model based on MapReduce and the SVM is proposed. The base classifier for this model uses an SVM, and parameter optimization is performed by the Cuckoo Search (CS) to determine the optimal parameters of the SVM. Considering the problem of time cost when a data set is large, we choose to use MapReduce to perform distributed training on SVMs to improve training speed. “Map" is used to map distributed training network security situation data, and “Reduce" merges and sorts the prediction results. Experimental results show that the proposed prediction model has improved the accuracy and decreased the training time cost compared to the traditional model.

While deep learning techniques have been extensively employed in malware detection, there is a notable challenge in effectively embedding malware features. Current neural network methods primarily capture superficial characteristics, lacking in-depth semantic exploration of functions and failing to preserve structural information at the file level. Motivated by the aforementioned challenges, this paper introduces MalHAPGNN, a novel framework for malware detection that leverages a hierarchical attention pooling graph neural network based on enhanced call graphs. Firstly, to ensure semantic richness, a Bidirectional Encoder Representations from Transformers-based (BERT) attribute-enhanced function embedding method is proposed for the extraction of node attributes in the function call graph. Subsequently, this work designs a hierarchical graph neural network that integrates attention mechanisms and pooling operations, complemented by function node sampling and structural learning strategies. This framework delivers a comprehensive profile of malicious code across semantic, syntactic, and structural dimensions. Extensive experiments conducted on the Kaggle and VirusShare datasets have demonstrated that the proposed framework outperforms other graph neural network (GNN)-based malware detection methods.