The need for sophisticated traceability systems has become essential in increasingly complex and globalized supply chains. The convergence of Blockchain (BC), Internet of Things (IoT), and Artificial Intelligence (AI) technologies offers promising solutions to enhance traceability systems across various sectors, particularly supply chain management (SCM). This paper presents a comprehensive bibliometric and systematic literature review to explore emerging trends, research patterns, and methodologies in integrating BC, IoT, and AI into traceability systems. In the study, 530 documents from the SCOPUS database for bibliometric analysis were examined, alongside a detailed review of 43 selected articles from multiple databases. The findings highlighted a significant increase in research output in recent years, with a dominant focus on agricultural supply chains and SCM. Notably, India and China lead the field in publications and citations. Furthermore, key authors and influential journals significantly contributed to advancing the research. The analysis also revealed a predominance of experimental and hybrid research methodologies, with Ethereum and Hyperledger Fabric emerging as the most widely adopted platforms for system implementations. The study identified critical research trends, including the growing role of AI-driven analytics, the importance of real-time IoT data collection, and the critical need for secure, tamper-proof data provided by BC. However, challenges, such as interoperability, scalability, and standardization, remain hindering the widespread adoption of these technologies. The paper proposes a four-layer conceptual framework for integrating BC, IoT, and AI into future traceability systems, emphasizing their potential to enhance transparency, security, and efficiency across various application areas. The paper concludes by offering directions for future research, highlighting the need for more empirical studies, industry-specific frameworks, and standardization to overcome existing limitations.

Hyperparameter optimization (HPO) is essential for deep learning in medical image classification, yet standard metaheuristics such as Manta Ray Foraging Optimization (MRFO) often suffer from premature convergence in high-dimensional search spaces. To address these limitations, an enhanced variant, MRFO-LF, was proposed by incorporating Lévy flight-based exploration, adaptive step-size decay, and a hybrid stochastic–deterministic search mechanism. This work details the first application of the proposed MRFO-LF to HPO in melanoma classification, a critical task within medical image analysis. The Lévy component enables long-range perturbations, while the adaptive decay mechanism gradually narrows the search scope, and the hybrid strategy balances global versus local exploration without relying on problem-specific heuristics. Experiments were conducted on the ISIC and PH $$ ^2 $$ 2 dermoscopic datasets using DenseNet121, InceptionV3, and VGG19. MRFO-LF attained peak validation accuracies of 99.49% (ISIC) and 100.00% (PH $$ ^2 $$ 2 ) for DenseNet121, with corresponding validation losses of 0.3580 and 0.0015. When compared to MRFO, PSO, and GA, the proposed method improved ISIC accuracy by 0.40%, reduced PH $$ ^2 $$ 2 loss by over 95%, and converged up to 30% faster. Statistical significance was confirmed through ANOVA and paired t-tests ( $$ p < 0.05 $$ p < 0.05 ). These results position MRFO-LF as a reliable and efficient optimizer for complex hyperparameter tuning in medical image classification.

Asthma exacerbations pose a significant global health concern, necessitating effective predictive models to anticipate and manage these events. This systematic literature review examined the optimization techniques employed in asthma exacerbation prediction models, spanning machine learning algorithms and computational optimization methods. The objective was to synthesize existing evidence, identify trends, and delineate future research directions in predictive modeling for asthma exacerbations to enhance predictive accuracy and clinical utility. A comprehensive search strategy was devised, yielding 27 eligible articles for analysis. The result revealed various optimization techniques, including feature selection, model optimization, and environmental factor integration. The result also revealed that machine learning algorithms’ effectiveness in predicting asthma exacerbations varied depending on various factors (such as dataset quality and model complexity), with various optimization techniques (such as feature selection and ensemble learning) used for improving predictive accuracy. Integrating environmental and spatial factors enhanced prediction models, enabling tailored interventions. In addition, personalized asthma management strategies informed by predictive models led to better control and reduced healthcare utilization. The review also highlighted the implications for personalized asthma management, as well as methodological limitations, and proposed future research directions to improve model reliability and advance personalized healthcare understanding, thereby contributing to the United Nations’ Sustainable Development Goals related to health, innovation, and sustainability. Thus, progress made in asthma exacerbation prediction and the identification of challenges and areas for improvement were covered, providing valuable insights for researchers, clinicians, and policymakers aiming to enhance asthma care through predictive modeling.

The integration of Blockchain (BC) and the Internet of Things (IoT) has emerged as a transformative solution for addressing traceability challenges in logistics, offering enhanced transparency, data security, and operational efficiency. This study presents a systematic review of the current state of BC-IoT integration for logistics traceability, focusing on its motivations, deployment strategies, technical implementations, and evaluation approaches. A total of 1,619 records were initially retrieved from IEEE Xplore, MDPI, ScienceDirect, Scopus, and Web of Science, from which 61 peer-reviewed studies published between 2015 and 2024 were selected and analyzed. The selection process adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a rigorous, transparent, and high-quality synthesis. Only studies explicitly addressing the convergence of BC and IoT within logistics traceability contexts were included. Key findings reveal that BC-IoT systems significantly enhance traceability and transparency across logistics networks but face persistent challenges such as scalability, latency, energy efficiency, and security. The review categorizes various deployment architectures including cloud, edge, fog, and hybrid models, and examines their implications for data responsiveness and system reliability. Additionally, it evaluates popular BC platforms (e.g., Hyperledger Fabric, Ethereum, Solana) and consensus mechanisms (e.g., RAFT, PBFT, PoS) based on their suitability for logistics applications. Emerging research directions emphasize the need for cross-chain interoperability, domain-specific frameworks, and decentralized traceability models, particularly in sectors such as humanitarian logistics and regulatory compliance. This review consolidates fragmented knowledge and provides actionable insights for developing scalable, secure, and transparent BC-IoT systems to support next-generation logistics operations.

Accurate detection of oil spills in dynamic port environments remains critical for timely response and effective environmental protection. Recent studies have explored the application of deep learning models to high-resolution RGB imagery; however, challenges persist due to the complex visual characteristics of ports, including dynamic water surfaces, infrastructural occlusions, and heterogeneous backgrounds. Traditional detection approaches have demonstrated limited robustness under such conditions, often resulting in elevated rates of false positives and missed detections. In this study, a semantic segmentation model based on the DeepLabV3 architecture was developed and optimized for oil spill detection using drone-acquired RGB imagery. Model performance was enhanced through the integration of convolutional neural networks, transfer learning, and systematic hyperparameter tuning involving batch size, learning rate, and training epochs. Evaluation was conducted through five-fold cross-validation to ensure generalization capability, and the results were benchmarked against established segmentation models. The optimized model achieved a validation accuracy of 96.53%, with an F1-score of 0.9557 and an Intersection over Union (IoU) of 0.8643. Cross-validation yielded a mean accuracy of 89.35% (±0.0103), precision of 88.52% (±0.0081), recall of 85.26% (±0.0138), F1-score of 86.45% (±0.0122), and IoU of 76.97% (±0.0177), indicating consistent performance across all partitions. Per-class performance further supported model reliability, with F1-scores of 0.92 for oil, 0.91 for other objects, and 0.97 for water, and corresponding IoU values of 0.8569, 0.8273, and 0.9469, respectively. When compared to baseline models such as Fully Convolutional Networks (FCN, IoU: 83.48%), U-Net (IoU: 78.80%), and MFSCNet (IoU: 86.24%), the proposed DeepLabV3-based architecture demonstrated superior segmentation accuracy. These findings indicate the model’s potential for real-time deployment in high-risk maritime zones, offering significant contributions to environmental monitoring, rapid incident response, and operational decision-making.