The exponential growth of wearable medical devices (WMDs) and the increasing demand for real-time health data sharing necessitate secure and fine-grained access control mechanisms. However, existing ciphertext-policy attribute-based encryption (CP-ABE) schemes suffer from computational and storage overheads that grow linearly with policy complexity. To address this challenge, we propose fog-assisted composite attribute-based encryption (FA-CABE), a novel scheme that integrates composite attributes with fog computing to enhance efficiency. FA-CABE leverages the subset sum problem (SSP) to map conjunctive policy clauses to composite attributes, substantially reducing both encryption and decryption overhead. A dualfog-node architecture offloads cryptographic computations from WMDs, enabling lightweight local processing. Rigorous security analysis under the Decisional Bilinear Diffie-Hellman (DBDH) assumption demonstrates that FA-CABE achieves replayable chosen ciphertext attack (RCCA) security. Experimental results show that FA-CABE achieves encryption speeds that are 22.13×–145.02× faster and decryption speeds that are 6.71×–161.83× faster than existing schemes, while requiring only a constant number of operations for decryption. Additionally, experimental validation on the Raspberry Pi 4B shows that the energy consumption is as low as 0.72 W per core, with data processing speed reaching 23.38 MB/s.

Federated Large Language Model (FedLLM) shows excellent potential in collaboratively training large language models (LLM) under the federated learning (FL) framework, which is benefiting from its privacy protection advantage. However, FedLLM faces the significant challenge of the non-IID problem. In the real world, there are often cross-source or even cross-domain language set data between IoT devices. To address the issue, we propose a new FedLLM framework FedALoRA via personalized and efficient parameter fine-tuning (PEFT). Specifically, the proposed scheme combines the personalized aggregation method and the LoRA method, which can adaptively aggregate the downloaded global model and local model to the local target on each client while ensuring low training costs. This adaptation initializes the local model before each iterative training, enabling clients to learn general knowledge while enhancing their understanding of their own domain knowledge. Extensive experiments and analysis on cross-domain non-IID settings and the financial datasets on Dirichlet non-IID settings demonstrate the effectiveness and superiority of FedALoRA.

Protein-protein interactions (PPIs) are fundamental to understanding cellular mechanisms, signaling networks, disease pathways, and drug development. Over the years, numerous computational models with artificial intelligence (AI) have been developed to predict PPIs. However, these models mostly face significant challenges, such as fragmented feature extraction pipelines, inability to capture complex global relationships among proteins, and reliance on handcrafted features. These challenges often limit their prediction accuracy. To address these issues, the Knowledge Graph Fused Graph Neural Network (KGF-GNN) was proposed, offering an end-to-end learning approach that integrates Protein Associated Network (PAN) with observed PPI data. While KGF-GNN achieves notable performance improvements, it focuses primarily on local topological features extracted by Graph Neural Networks (GNNs), potentially overlooking critical global patterns. Moreover, its feature fusion process lacks the flexibility to effectively combine diverse biological information. To overcome these shortcomings, this paper introduces a Hybrid Ensemble End-to-End Neural Network (HEENN), which incorporates three key innovations: (1) Local Feature Extraction via Graph Attention Network (GAT): HEENN employs GAT to enable more precise extraction of local topological and semantic features, allowing the model to focus on the most relevant interactions and relationships within the data. (2) Global Feature Extraction via AutoEncoder: By leveraging an AutoEncoder framework, HEENN captures comprehensive global features from PANs and PPI datasets, complementing the GAT's local features to produce richer protein representations. (3) Attention-Enhanced Feature Fusion: An attention mechanism is employed during feature fusion to ensure an adaptive and effective integration of local and global features. Extensive experiments on real-world PPI datasets demonstrate that HEENN significantly outperforms KGF-GNN and other state-of-the-art models, achieving superior accuracy in PPI prediction. These advancements underscore the potential of HEENN in AI-driven bioinformatics research, which offers new opportunities for biological discovery and therapeutic innovation.

Protein-protein interactions (PPIs) are the foundation for numerous biological processes within cells, which are crucial for understanding cellular signaling networks, disease mechanisms, and drug development. Recently, numerous artificial intelligence (AI)-based approaches have emerged for predicting PPIs. Nevertheless, existing AI-based approaches either partially or loosely consider these relationships and mechanisms by a non-end-to-end learning framework, resulting in sub-optimal feature extractions and fusions for prediction. To address this issue, this paper proposes an end-to-end graph neural network model for protein-protein interaction prediction, termed the Knowledge Graph Fused Graph Neural Network (KGF-GNN). First, protein associated network (PAN) is constructed by comprehensively exploiting protein-associated relationships and mechanisms among drugs, diseases, ribonucleic acid, protein structures, etc. Then, a graph neural network (GNN) is built to extract both the topological and semantic features from PAN. Secondly, the observed interactions between proteins are constructed into a PPI network, and another GNN is built to extract the hidden topological features within the PPI network. Third, a multi-layer perceptron is designed to fuse the extracted various features by end-to-end learning. With such designs, the feature extractions and fusions of PPIs are guaranteed to be comprehensive and optimal for prediction. Finally, by conducting extensive experiments on real PPI datasets, we demonstrate that our KGF-GNN can accurately predict PPIs and significantly outperform state-of-the-art models.

Abstract Anemia (hemoglobin (Hb) < 12.0 g/dL) is significantly correlated with many diseases. An invasive technique is the peripheral blood Hb detection method, which is used to examine red and white blood cells and platelets in clinical laboratory settings. However, non‐invasive methods for measuring Hb mainly include low‐precision prediction based on eye images and complex operation prediction based on fundus images. Moreover, these types of anemia testing techniques are time‐consuming, tedious, or prone to errors. Thus, developing a convenient and high‐precision method is vital for predicting Hb concentration. This study proposes self‐supervised causal features using actor‐critical reinforcement learning to improve the model prediction performance. Two networks are proposed: Actor Predictor and Hemoglobin Predictor to predict Hb concentration. Moreover, the model performance is evaluated using different techniques, namely, Mean Absolute Error (MAE) and Mean Square Error (MSE), via real eye image data and a smartphone. This model achieved 1.19(1.01,1.38) on the MAE and 2.25(1.59,2.90) on the MSE, which outperformed previous eye images' Hb prediction methods and was nearly similar to the fundus images' Hb prediction methods. The inference time was less than 0.05 s, making it efficient and accurate for predicting Hb. This model can be used for mobile deployment and health self‐screening.

In hyperspectral images (HSIs), mixed noise (e.g., Gaussian noise, impulse noise, stripe noise, and deadlines) contamination is a common phenomenon that greatly reduces the visual quality of the image. In recent years, methods combining global and non-local low-rankness have been widely used in the field of HSI denoising. However, most methods apply original space-based denoising strategies (low-rank tensor decomposition, total variation, and tensor sparse representation, etc.) directly to the modeling of non-local low-rank tensors in subspace, without fully exploiting the intrinsic and latent properties of the non-local similar tensors. In this paper, we propose a hybrid prior denoising method based on global tensor low-rankness and non-local SVD-aided group sparsity (GTL_NSGS). This method introduces a novel plug-and-play NSGS denoiser that uses singular value decomposition as assistance to successively explore self-similarity of spatial dimension, low-rankness of spectral dimension, and group sparsity of difference domain in subspace non-local similar tensors. Globally, we utilize the existing three-way log-based tensor nuclear norm (3DLogTNN) to approximate the HSI tensor fibered rank and introduce a difference continuity regularization to obtain a continuous smooth spectral basis. Finally, we combine the Alternating Direction Method of Multipliers (ADMM) with the Augmented Lagrangian Multiplication (ALM) algorithm to solve the proposed model effectively. Extensive experiments on simulated and real data sets demonstrate that the proposed method has superior performance in removing mixed noise compared to state-of-the-art denoisers.

Despite the remarkable progress made by the salient object detection of natural sensing images (NSI-SOD), the complex background and scale diversity issues of remote sensing images (RSIs) still pose a substantial obstacle. In this study, we build an end-to-end channel-enhanced remodeling-based network (CRNet) for optical RSIs (ORSIs) to highlight salient objects through feature augmentation. First, the backbone convolutional block is used to suggest the fundamental characteristics. Then, we use the channel enhance module (CEM) to enhance the shallow features. CEM primarily relies on the channel attention mechanism and employs a no-downscaling strategy to produce local cross-channel interaction, which lowers model complexity while enhancing extraction performance. Meanwhile, we use the redefined feature module (RFM) to reconstruct the deep features and generate global attention features by dimensional transformation and feature relationship aggregation to achieve the role of locating salient targets. Finally, the cascade combines the multi-scale features to provide the final saliency map. To further enhance the representational power of the network, we use a hybrid loss function to improve performance. The proposed approach outperforms current state-of-the-art methods, as shown by several experiments on three available datasets. The source code of the proposed CRNet is available publicly at https://github.com/hilitteq/CRNet.git.

Hyperspectral image (HSI) classification refers to accurately corresponding each pixel in an HSI to a land-cover label. Recently, the successful application of multiscale and multifeature methods has greatly improved the performance of HSI classification due to their enhanced utilization of the available spectral–spatial information. However, as the number of scales and the number of features increases, it becomes more difficult to achieve an optimal degree of fusion for multiple classifiers [e.g., kernel extreme learning machine (KELM)]. On the other hand, a limited sample size of the HSI may cause overfitting problems, which seriously affects the classification accuracy. Therefore, in this article, a novel multi-structure KELM with attention fusion strategy (MSAF-KELM) is proposed to achieve accurate fusion of multiple classifiers for effective HSI classification with ultrasmall sample rates. First, a multi-structure network is built, which combines multiple scales and multiple features to extract abundant spectral–spatial information. Second, a fast and efficient KELM is employed to enable rapid classification. Finally, a weighted self-attention fusion strategy (WSAFS) is introduced, which combines the output weights of each KELM subbranch and the self-attention mechanism to achieve an efficient fusion result on multi-structure networks. We conducted experiments on four types of HSI datasets with different evaluation methods and compared them with several classical and state-of-the-art methods, which demonstrate the excellent performance of our method on ultrasmall sample rates. The code is available athttps://github.com/Fang666666/MSAF-KELMfor reproducibility.