Search Results (57)

Background/Objectives: Federated learning (FL) offers a way for healthcare institutions to collaboratively train machine learning models without sharing sensitive patient data. This systematic review aims to comprehensively synthesize the ethical dimensions of FL in healthcare, integrating privacy preservation, algorithmic fairness, governance, and equitable access into a unified analytical framework. The application of FL in healthcare between January 2020 and December 2024 is examined, with a focus on ethical issues such as algorithmic fairness, privacy preservation, governance, and equitable access. Methods: Following PRISMA guidelines, six databases (PubMed, IEEE Xplore, Web of Science, Scopus, ACM Digital Library, and arXiv) were searched. The PROSPERO registration is CRD420251274110. Studies were selected if they described FL implementations in healthcare settings and explicitly discussed ethical considerations. Key data extracted included FL architectures, privacy-preserving mechanisms, such as differential privacy, secure multiparty computation, and encryption, as well as fairness metrics, governance models, and clinical application domains. Results: Out of 3047 records, 38 met the inclusion criteria. The most popular applications were found in medical imaging and electronic health records, especially in radiology and oncology. Through thematic analysis, four key ethical themes emerged: algorithmic fairness, which addresses differences between clients and attributes; privacy protection through formal guarantees and cryptographic techniques; governance models, which emphasize accountability, transparency, and stakeholder engagement; and equitable distribution of computing resources for institutions with limited resources. Considerable variation was observed in how fairness and privacy trade-offs were evaluated, and only a few studies reported real-world clinical deployment. Conclusions: FL has significant potential to promote ethical AI in healthcare, but advancement will require the development of common fairness standards, workable governance plans, and systems to guarantee fair benefit sharing. Future studies should develop standardized fairness metrics, implement multi-stakeholder governance frameworks, and prioritize real-world clinical validation beyond proof-of-concept implementations. Full article

Artificial intelligence (AI) has become a strategic technology for global health, with increasing relevance amid the climate emergency and persistent digital inequalities. This study examines how AI has been applied in Brazilian healthcare through a scoping review with an in-depth methodological synthesis, focusing on Transfer Learning (TL) and Federated Learning (FL) as approaches to address data scarcity, privacy, and technological dependence. We searched PubMed, SciELO, and the CNPq Theses and Dissertations Repository for peer-reviewed studies on AI applications in Brazil, screened titles using AI-assisted tools with manual validation, and analyzed thematic patterns across methodological and infrastructural dimensions. Among 349 studies retrieved, six explicitly used TL or FL. These techniques were frequently implemented through multi-country research consortia, demonstrating scalability and feasibility for collaborative model training under privacy constraints. However, they remain marginal in mainstream practice despite their ability to deploy AI solutions with limited computational resources while preserving data sovereignty. The findings indicate an emerging yet uneven integration of resource-aware AI in Brazil, underscoring its potential to advance equitable innovation and digital autonomy in health systems of the Global South. Full article

In the context of privacy-preserving face recognition systems, entropy plays a crucial role in determining the efficiency and security of computational processes. However, existing schemes often encounter challenges such as inefficiency and high entropy in their computational models. To address these issues, we propose a privacy-preserving face recognition method based on the Face Feature Coding Method (FFCM) and symmetric homomorphic encryption, which reduces computational entropy while enhancing system efficiency and ensuring facial privacy protection. Specifically, to accelerate the matching speed during the authentication phase, we construct an N-ary feature tree using a neural network-based FFCM, significantly improving ciphertext search efficiency. Additionally, during authentication, the server computes the cosine similarity of the matched facial features in ciphertext form using lightweight symmetric homomorphic encryption, minimizing entropy in the computation process and reducing overall system complexity. Security analysis indicates that critical template information remains secure and resilient against both passive and active attacks. Experimental results demonstrate that the facial authentication efficiency with FFCM classification is 4% to 6% higher than recent state-of-the-art solutions. This method provides an efficient, secure, and entropy-aware approach for privacy-preserving face recognition, offering substantial improvements in large-scale applications. Full article

Financial time series are heterogeneous, nonstationary, and dispersed across institutions that cannot share raw data. While federated learning enables collaborative modeling under privacy constraints, fixed architectures struggle to accommodate cross-market drift and device-resource diversity; conversely, existing neural architecture search techniques presume centralized data and typically ignore communication, latency, and privacy budgets. This paper introduces FedRegNAS, a regime-aware federated NAS framework that jointly optimizes forecasting accuracy, communication cost, and on-device latency under user-level $(\epsilon ,\delta )$-differential privacy. FedRegNAS trains a shared temporal supernet composed of candidate operators (dilated temporal convolutions, gated recurrent units, and attention blocks) with regime-conditioned gating and lightweight market-aware personalization. Clients perform differentiable architecture updates locally via Gumbel-Softmax and mirror descent; the server aggregates architecture distributions through Dirichlet barycenters with participation-weighted trust, while model weights are combined by adaptive, staleness-robust federated averaging. A risk-sensitive objective emphasizes downside errors and integrates transaction-cost-aware profit terms. We further inject calibrated noise into architecture gradients to decouple privacy leakage from weight updates and schedule search-to-train phases to reduce communication. Across three real-world equity datasets, FedRegNAS improves directional accuracy by 3–7 percentage points and Sharpe ratio by 18–32%. Ablations highlight the importance of regime gating and barycentric aggregation, and analyses outline convergence of the architecture mirror-descent under standard smoothness assumptions. FedRegNAS yields adaptive, privacy-aware architectures that translate into materially better trading-relevant forecasts without centralizing data. Full article

With the widespread application of multi-view data across various domains, multi-view unsupervised feature selection (MUFS) has achieved remarkable progress in both feature selection (FS) and missing-view completion. However, existing MUFS methods typically rely on centralized servers, which not only fail to meet privacy requirements in distributed settings but also suffer from suboptimal FS quality and poor convergence. To overcome these challenges, we propose a novel federated incomplete MUFS method (Fed-IMUFS), which integrates a fractional Sparsity-Guided Whale Optimization Algorithm (SGWOA) and Tensor Alternating Learning (TAL). Within this federated learning framework, each client performs local optimization in two stages: in the first stage, SGWOA introduces an $L_{2,1}$ proximal projection to enforce row-sparsity in the FS weight matrix, while fractional-order dynamics and fractal-inspired elite kernel injection mechanisms enhance global search ability, yielding a discriminative and stable weight matrix; in the second stage, based on the obtained weight matrix, an alternating optimization framework with tensor decomposition is employed to iteratively complete missing views while simultaneously optimizing low-dimensional representations to preserve cross-view consistency, with the objective function gradually minimized until convergence. During federated training, the server employs an aggregation and distribution strategy driven by normalized mutual information, where clients upload only their local weight matrices and quality indicators, and the server adaptively fuses them into a global FS matrix before distributing it back to clients. This process achieves consistent FS across clients while safeguarding data privacy. Comprehensive evaluations on CEC2022 and several incomplete multi-view datasets confirm that Fed-IMUFS outperforms state-of-the-art methods, delivering stronger global optimization capability, higher-quality feature selection, faster convergence, and more effective handling of missing views. Full article

Safeguarding data confidentiality and enforcing precise access regulation in cloud platforms continue to be major research concerns. Attribute-based encryption (ABE) offers a versatile framework for policy-driven control, whereas public key encryption with keyword search (PEKS) supports efficient querying of encrypted datasets. However, ABE lacks keyword search support, and PEKS offers limited control over access policies. To overcome these limitations, attribute-based keyword search (ABKS) schemes have been proposed, with recent advances such as ciphertext-policy decryptable ABKS (CP-DABKS) enabling secure channel-free keyword search. Nevertheless, the existing CP-DABKS schemes still face important challenges: the master public key grows linearly with the attribute universe, secure channels are often required to deliver trapdoors, and many designs remain vulnerable to keyword guessing attacks. This work introduces an efficient CP-DABKS scheme built upon a Type-3 pairing framework to directly overcome these limitations. The proposed design employs a commit-to-point mechanism that prevents linear key growth, eliminates the need for secure trapdoor transmission, and resists keyword guessing attacks. We implement and evaluate the proposed scheme using real-world data from the Enron Email dataset and demonstrate its practicality for secure and searchable cloud-based storage. We also discuss implementation considerations and outline directions for future enhancement of privacy-preserving searchable encryption systems. Full article

Background and Objectives: Public health needs collaborative, privacy-preserving analytics, but centralized AI is constrained by data sharing and governance. Federated learning (FL) enables training without moving sensitive data. This review assessed how FL is used for disease prevention in population and public health, and mapped benefits, challenges, and policy implications. Methods: Following PRISMA 2020, we searched PubMed, Scopus, Web of Science, IEEE Xplore, and Google Scholar for peer reviewed English-language studies from January 2020–30 June 2025, applying FL to surveillance, outbreak detection, risk prediction, or policy support. Two reviewers screened and extracted data with third-reviewer arbitration. Quality was appraised with a tool adapted from MMAT and AI reporting frameworks. No meta-analysis was performed. Results: Of 5230 records identified (4720 after deduplication), 200 full texts were assessed and 19 were included. Most used horizontal FL across multiple institutions for communicable diseases, COVID-19, tuberculosis and some chronic conditions. Reported gains included privacy preservation across sites, better generalizability from diverse data, near real-time intelligence, localized risk stratification, and support for resource planning. Common barriers were non-IID data, interoperability gaps, compute and network limits in low-resource settings, unclear legal pathways, and concerns about fairness and transparency. Few studies linked directly to formal public-health policy or low-resource deployments. Conclusions: FL is promising for equitable, secure, and scalable disease-prevention analytics that respect data sovereignty. Priorities include robust methods for heterogeneity, interoperable standards, secure aggregation, routine fairness auditing, clearer legal and regulatory guidance, and capacity building in underrepresented regions. Full article

Large Language Models (LLMs) are increasingly proposed to personalize healthcare delivery, yet their real-world readiness remains uncertain. We conducted a systematic literature review to assess how LLM-based systems are designed and used to enhance patient engagement and personalization, while identifying open challenges these tools pose. Four digital libraries (Scopus, IEEE Xplore, ACM, and Nature) were searched, yielding 3787 studies; 16 met the inclusion criteria. Most studies, published in 2024, span different types of motivations, architectures, limitations and privacy-preserving approaches. While LLMs show potential in automating patient data collection, recommendation/therapy generation, and continuous conversational support, their clinical reliability is limited. Most evaluations use synthetic or retrospective data, with only a few employing user studies or scalable simulation environments. This review highlights the tension between innovation and clinical applicability, emphasizing the need for robust evaluation protocols and human-in-the-loop systems to guide the safe and equitable deployment of LLMs in healthcare. Full article

The Attribute-Based Access Control (ABAC) model provides access control decisions based on subject, object (resource), and contextual attributes. However, the use of sensitive attributes in access control decisions poses many security and privacy challenges, particularly in cloud environment where third parties are involved. To address this shortcoming, we present a novel privacy-preserving Dummy-ABAC model that obfuscates real attributes with dummy attributes before transmission to the cloud server. In the proposed model, only dummy attributes are stored in the cloud database, whereas real attributes and mapping tokens are stored in a local machine database. Only dummy attributes are used for the access request evaluation in the cloud, and real data are retrieved in the post-decision mechanism using secure tokens. The security of the proposed model was assessed using a simulated threat scenario, including attribute inference, policy injection, and reverse mapping attacks. Experimental evaluation using machine learning classifiers (“DecisionTree” DT, “RandomForest” RF), demonstrated that inference accuracy dropped from ~0.65 on real attributes to ~0.25 on dummy attributes confirming improved resistance to inference attacks. Furthermore, the model rejects malformed and unauthorized policies. Performance analysis of dummy generation, token generation, encoding, and nearest-neighbor search, demonstrated minimal latency in both local and cloud environments. Overall, the proposed model ensures an efficient, secure, and privacy-preserving access control in cloud environments. Full article

With the growing adoption of circular economy principles in manufacturing, efficient disassembly and reassembly of end-of-life (EOL) products has become a key challenge in smart factories. This paper addresses the Disassembly and Assembly Line Balancing Problem (DALBP), which involves scheduling robotic tasks across workstations while minimizing total operation time and accounting for directional switching time between disassembly and assembly phases. To solve this problem, we propose an improved reinforcement learning algorithm, IQ($\lambda$), which extends the classical Q($\lambda$) method by incorporating eligibility trace decay, a dynamic Action Table mechanism to handle non-conflicting parallel tasks, and switching-aware reward shaping to penalize inefficient task transitions. Compared with standard Q($\lambda$), these modifications enhance the algorithm’s global search capability, accelerate convergence, and improve solution quality in complex DALBP scenarios. While the current implementation does not deploy live IoT infrastructure, the architecture is modular and designed to support future extensions involving edge-cloud coordination, trust-aware optimization, and privacy-preserving learning in Industrial Internet of Things (IIoT) environments. Four real-world disassembly-assembly cases (flashlight, copier, battery, and hammer drill) are used to evaluate the algorithm’s effectiveness. Experimental results show that IQ($\lambda$) consistently outperforms traditional Q-learning, Q($\lambda$), and Sarsa in terms of solution quality, convergence speed, and robustness. Furthermore, ablation studies and sensitivity analysis confirm the importance of the algorithm’s core design components. This work provides a scalable and extensible framework for intelligent scheduling in cyber-physical manufacturing systems and lays a foundation for future integration with secure, IoT-connected environments. Full article

This paper explores the integration of Large Language Models (LLMs) and secure Gen-AI technologies within engineering design and manufacturing, with a focus on improving inventory management, component selection, and recommendation workflows. The system is intended for deployment and evaluation in a real-world industrial environment. It utilizes vector embeddings, vector databases, and Approximate Nearest Neighbor (ANN) search algorithms to implement Retrieval-Augmented Generation (RAG), enabling context-aware searches for inventory items and addressing the limitations of traditional text-based methods. Built on an LLM framework enhanced by RAG, the system performs similarity-based retrieval and part recommendations while preserving data privacy through selective obfuscation using the ROT13 algorithm. In collaboration with an industry sponsor, real-world testing demonstrated strong results: 88.4% for Answer Relevance, 92.1% for Faithfulness, 80.2% for Context Recall, and 83.1% for Context Precision. These results demonstrate the system’s ability to deliver accurate and relevant responses while retrieving meaningful context and minimizing irrelevant information. Overall, the approach presents a practical and privacy-aware solution for manufacturing, bridging the gap between traditional inventory tools and modern AI capabilities and enabling more intelligent workflows in design and production processes. Full article

Among searchable encryption techniques, multi-user dynamic searchable encryption (MUDSE) schemes are an important research direction. After the data owner transfers data to the cloud, it may be necessary to authorize different users to access some or all of the data while allowing for dynamic updates. Enabling dynamic data sharing in cloud storage while preserving users’ ability to search the data is crucial for promoting data flow and maximizing its value. This approach is particularly significant in addressing the data silo problem. However, existing security mechanisms remain imperfect, and most current scenarios assume that cloud servers are merely “curious but honest”. In reality, cloud servers may exhibit malicious behavior, such as returning incorrect or incomplete search results. Similarly, malicious users might falsify search results—for example, to avoid payment—or collude with cloud servers to steal other users’ search privacy. To address these challenges, this paper proposes an updatable multi-user dynamic searchable encryption scheme with bidirectional verification. The scheme enables secure dynamic data sharing in multi-user scenarios by constructing an index structure using homomorphic message authentication codes and bitmaps. This ensures secure updates to encrypted data without revealing the relationship between files and keyword search keys while providing forward and backward security. Regarding privilege management, the scheme employs updatable keys, ensuring that users can only generate valid search commands if they possess the latest encryption key. Additionally, blockchain technology is introduced to assist in verifying user honesty. Through actual testing and security analysis, the proposed solution demonstrates improved search speed over traditional methods while maintaining security. It also exhibits high adaptability for handling frequently changing cloud data. Full article

The emergence of Unmanned Aerial Vehicles (UAVs), commonly known as drones, has presented numerous transformative opportunities across sectors such as agriculture, commerce, and security surveillance systems. However, the proliferation of these technologies raises significant concerns regarding security and privacy, as they could potentially be exploited for unauthorized surveillance or even targeted attacks. Various research endeavors have proposed drone detection models for security purposes. Yet, deploying these models on edge devices proves challenging due to resource constraints, which limit the feasibility of complex deep learning models. The need for lightweight models capable of efficient deployment on edge devices becomes evident, particularly for the anonymous detection of drones in various disguises to prevent potential intrusions. This study introduces a lightweight deep learning-based drone detection model (LDDm-CNN) by fusing knowledge distillation with Bayesian optimization. Knowledge distillation (KD) is utilized to transfer knowledge from a complex model (teacher) to a simpler one (student), preserving performance while reducing computational complexity, thereby achieving a lightweight model. However, selecting optimal hyper-parameters for knowledge distillation is challenging due to a large number of search space and complexity requirements. Therefore, through the integration of Bayesian optimization with knowledge distillation, we present an enhanced CNN-KD model. This novel approach employs an optimization algorithm to determine the most suitable hyper-parameters, enhancing the efficiency and effectiveness of the drone detection model. Validation on a dedicated drone detection dataset illustrates the model’s efficacy, achieving a remarkable accuracy of 96% while significantly reducing computational and memory requirements. With just 102,000 parameters, the proposed model is five times smaller than the teacher model, underscoring its potential for practical deployment in real-world scenarios. Full article

Synthetic medical text generation has emerged as a solution to data scarcity and privacy constraints in clinical NLP. This review systematically evaluates the use of Large Language Models (LLMs) for structured medical text generation, examining techniques such as retrieval-augmented generation (RAG), structured fine-tuning, and domain-specific adaptation. Four search queries were applied following the PRISMA methodology to identify and extract data from 153 studies. Key benchmarking metrics, such as performance measures, and qualitative insights, including methodological trends and challenges, were documented. The results show that while LLM-generated text improves fluency, hallucinations and factual inconsistencies persist. Structured consultation models, such as SOAP and Calgary–Cambridge, enhance coherence but do not fully prevent errors. Hybrid techniques that combine retrieval-based grounding with domain-specific fine-tuning improve factual accuracy and task performance. Conventional evaluation metrics (e.g., ROUGE, BLEU) are insufficient for medical validation, highlighting the need for domain-specific benchmarks. Privacy-preserving strategies, including differential privacy and PHI de-identification, support regulatory compliance but may reduce linguistic quality. These findings are relevant for clinical NLP applications, such as AI-powered scribe systems, where structured synthetic datasets can improve transcription accuracy and documentation reliability. The conclusions highlight the need for balanced approaches that integrate medical structure, factual control, and privacy to enhance the usability of synthetic medical text. Full article

Background: Healthcare information systems are hindered by delayed data sharing, privacy breaches, and lack of patient control over data. The growing need for secure, privacy-preserved access control interoperable in health informatics technology (HIT) systems appeals to solutions such as Blockchain (BC), which offers a decentralized, transparent, and immutable ledger architecture. However, its current adoption remains limited to conceptual or proofs-of-concept (PoCs), often relying on simulated datasets rather than validated real-world data or scenarios, necessitating further research into its pragmatic applications and their benchmarking. Objective: This systematic literature review (SLR) aims to analyze BC-based healthcare implementations by benchmarking peer-reviewed studies and turning PoCs or production insights into real-world applications and their evaluation metrics. Unlike prior SLRs focusing on proposed or conceptual models, simulations, or limited-scale deployments, this review focuses on validating practical BC real-world applications in healthcare settings beyond conceptual studies and PoCs. Methods: Adhering to PRISMA-2020 guidelines, we systematically searched five major databases (Scopus, Web of Science, PubMed, IEEE Xplore, and ScienceDirect) for high-precision relevant studies using MeSH terms related to BC in healthcare. The designed review protocol was registered with OSF, ensuring transparency in the review process, including study screening by independent reviewers, eligibility, quality assessment, and data extraction and synthesis. Results: In total, 82 original studies fully met the eligibility criteria and narratively reported BC-based healthcare implementations with validated evaluation outcomes. These studies highlight the current challenges addressed by BC in healthcare settings, providing both qualitative and quantitative data synthesis on its effectiveness. Conclusions: BC-based healthcare implementations show both qualitative and quantitative effectiveness, with advancements in areas such as drug traceability (up to 100%) and fraud prevention (95% reduction). We also discussed the recent challenges of focusing more attention in this area, along with a discussion on the mythological consideration of our own work. Our future research should focus on addressing scalability, privacy-preservation, security, integration, and ethical frameworks for widespread BC adoption for data-driven healthcare. Full article