The rapid adoption of cloud storage for healthcare data brings enhanced accessibility and streamlined data management, but it also introduces significant security and privacy challenges. This paper presents a multi-layered framework designed to securely manage and protect sensitive healthcare data in cloud environments. The proposed approach leverages encryption for securing data, combined with Deterministic Additive Noise (DAN) to enhance privacy during data storage and retrieval. Additionally, differential privacy mechanisms are employed to safeguard patient data during query processing, preventing unauthorized access while maintaining data utility. Key management is handled via a Hardware Security Module (HSM) to ensure the secure generation, storage, and usage of cryptographic keys. The framework also integrates secure authentication methods, including multi-factor authentication for both healthcare providers and patients, enabling secure data access and retrieval. This comprehensive approach strengthens the Confidentiality, Integrity, and Availability (CIA) of healthcare data in cloud environments, ensuring compliance with stringent privacy regulations such as HIPAA.

A Convolutional Neural Network (CNN) relies on intricate pixel correlations to predict objects within images, demonstrating improved performance when identifying objects alongside commonly co-occurring objects. For instance, a helicopter can be more accurately predicted when it appears with an aircraft carrier compared to when it stands alone. Influence on detecting an object due to the presence or absence of another object can be regarded as a form of contextual bias. This phenomenon poses a significant challenge in object detection. Addressing this bias is crucial. Our study explores correlation-aware prediction techniques to disentangle and leverage connections between objects. We propose CART-OD, a novel object detection approach that enhances model performance by adjusting the confidence scores of correlated objects during inference. By strategically intervening during prediction, CART-OD targets instances where commonly co-occurring objects appear separately, effectively mitigating the influence of object correlation. The adjustment in real-time makes the approach computationally efficient and scalable to handle large-scale datasets. Results indicate that CART-OD significantly enhances precision, recall, and mean average precision (mAP) in detecting objects that appear without their most correlated counterparts, underscoring the potential of correlation-aware strategies in object detection tasks.

The attention mechanism is a core module in today’s well-established models like Transformers and Graph Attention Networks (GAT). The attention coefficients needed typically requires heavy computational resources for stability and better predictions. This requirement limits the size of the sequence or graph a model can learn from, leading to separating connected graphs or sequences to ease training. In this paper, we introduce an attention mechanism that leverages Bilinear forms to handle sequences of arbitrary length, offering greater flexibility and scalability. This improvement not only optimizes computational resources but also offers better model performance. Furthermore, our Bilinear Attention Mechanism (BAM) can capture more complex relationships between nodes for less parameters count.

This paper explores the potential of Dynamic Model Selection (DMS) combined with asynchronous Federated Learning (FL) to enhance military devices within the Internet of Battlefield Things (IoBT). While previous studies have primarily focused on synchronous FL, we argue this approach has limitations. Hence, this study addresses this research gap by investigating asynchronous FL and its advantages over synchronous methods. Our main contribution lies in developing a dynamic client selection algorithm that optimizes the utility function, incorporating communication delay or latency, available computational resources, and model accuracy. This algorithm ensures clients are selected based on their potential contribution to the global model’s performance, thereby improving the efficiency and effectiveness of the training process. Through experimentation with standard datasets (MNIST, CIFAR-10) and our custom dataset (Common Objects in Battlefield (COBA)) tailored for battlefield scenarios, we demonstrate the effectiveness of asynchronous FL and dynamic client selection. By addressing this critical research gap and providing empirical evidence, our study advances the use of FL for battlefield decision-making, significantly contributes to developing IoBT, and offers practical insights for military applications.

Deep reinforcement learning (DRL) has shown remarkable success in tackling complex tasks by learning representations directly from raw data. However, as DRL agents become increasingly sophisticated, they often face challenges in adapting to new tasks or environments due to their monolithic structure. In this paper, we prove that by decomposing the agent into modular components, we enable more efficient adaptation as well as more efficient than that of the singular DRL agent on tackling complex tasks. Overall, our work contributes to advancing the field of DRL by introducing a flexible and scalable framework for agent decomposition. By enabling the creation of modular agents, we empower DRL systems to efficiently adapt to changing environments, learn new tasks with minimal intervention, and achieve higher levels of performance and versatility.