Hydrogels are hydrated three-dimensional polymeric networks with biomedical potential because they can encapsulate therapeutic agents and provide localised, sustained, or stimulus-responsive drug delivery. Their performance is determined by interacting variables, including polymer composition, synthesis route, crosslinking chemistry, drug loading, swelling, degradation, and the biological microenvironment. This multidimensional design space often makes hydrogel development slow and dependent on trial-and-error experimentation. This review examines the data-driven development of biomedical hydrogels for controlled drug delivery, focusing on clinical applications and emerging machine-learning approaches that support material selection, formulation design, synthesis optimisation, and release prediction. The review first discusses natural and synthetic hydrogels, including alginate, chitosan, gelatin-based systems, hyaluronic acid, and polyethylene glycol, with emphasis on how their physicochemical properties influence biocompatibility, synthesis flexibility, and drug-release behaviour. Key applications are then considered, including wound healing, cancer therapy, glucose-responsive insulin delivery, and inflammatory disease management. Particular attention is given to injectable and stimuli-responsive hydrogels, where formulation conditions and synthesis parameters can be tuned to improve localisation, therapeutic exposure, and release control. The review evaluates machine-learning methods, including random forest, gradient boosting, artificial neural networks, Gaussian process regression, and active learning, for predicting hydrogel properties, modelling release profiles, optimizing synthesis and formulation variables, and prioritizing experimental candidates. Finally, translational challenges are addressed, including small non-standardised datasets, limited external validation, weak in vitro-clinical correlations, material safety, explainability, reproducibility, scalability, and regulatory requirements. By integrating clinical, materials, synthesis, and machine-learning perspectives, this review highlights opportunities for developing safer and clinically relevant hydrogel-based drug-delivery systems.

With the surge of e-commerce, understanding user behavior through online data has become paramount. This study harnesses web usage mining, focusing on refining user experience by analyzing interaction data from a bespoke e-commerce platform. Utilizing web log analysis, user behaviors and engagement across various website pages were meticulously explored, revealing insights that augment our understanding of user patterns and behavior on the site. Furthermore, through a meticulous review of existing literature and an exploratory analysis of the access log files of a medical e-commerce website, we not only underscore improvements for enhanced user navigation and interaction but also aspire to foster uplifts in sales and customer retention by creating a more personalized user journey. The implications of this study extend to the tailored development of e-commerce platforms in the medical domain, promoting a user-centric approach in interface design and functionality based on discernible user interaction trends.

Hantavirus remains an important zoonotic threat because of its association with severe human diseases, including hemorrhagic fever with renal syndrome and hantavirus pulmonary syndrome. Its transmission is strongly influenced by rodent reservoirs, environmental conditions, human exposure patterns, and regional ecological variability. Recent advances in artificial intelligence (AI) and deep learning have created new opportunities for improving Hantavirus detection, outbreak prediction, ecological risk mapping, diagnostic support, and public health surveillance. This review examines the role of AI-driven methods in Hantavirus research, with emphasis on how machine learning, deep learning, image-based analysis, epidemiological modeling, and data-driven surveillance can support earlier detection and more informed decision-making. The review also discusses the potential of AI to integrate heterogeneous data sources, including clinical records, environmental variables, remote sensing indicators, genomic information, and epidemiological reports. Despite these advances, several challenges remain, including limited datasets, geographic bias, model generalization, lack of clinical validation, data imbalance, interpretability concerns, and the need for real-time deployment. Overall, AI and deep learning offer promising tools for strengthening Hantavirus surveillance and response, but their practical value depends on transparent models, high-quality data, interdisciplinary validation, and integration into public health systems.

Lung cancer arises from mutations in lung cells, disrupting their normal growth cycle and leading to uncontrolled cell division. These rapidly dividing cells lose function and fail to form healthy lung tissue. Several factors contribute to the difficulty of diagnosing and classifying lung nodules, including the high degree of morphological heterogeneity and overlapping characteristics between benign and malignant nodules. Recently, deep learning models have been used in computed tomography (CT)-based lung nodule diagnosis and have demonstrated diagnostic efficiency comparable to that of radiologists. This study introduces LungCNET, a high-performance multi-layer deep convolutional neural network trained on chest CT images to improve lung lesion classification efficiency and accuracy significantly. The data for the Lung Cancer convolutional neural network (LungCNET) were derived from the IQ-OTH/NCCD CT scan dataset (1097 images from 110 cases), split into training (767 images), validation (109) and a held-out test partition (221) that played no role in training or model selection. This dataset encompasses three diagnostic categories: benign, malignant, and normal lung tissues. LungCNET was evaluated against fine-tuned benchmark models that are both established and widely used, spanning architectures introduced between 2014 and 2024, including VGG16, ResNet50, InceptionV3, MobileNetV2, and YOLOv11. On the held-out test partition, LungCNET reached a macro-averaged F1-score of 95.19%, with VGG16 at 94.09% and InceptionV3 at 92.28%; these three models performed comparably, and the separation between them is small relative to the resolution of a test set of this size. LungCNET was, however, the only model to exceed 90% F1-score across all three diagnostic classes simultaneously, and recorded the highest F1-score on the benign class (91.0%), the smallest and most frequently misclassified category, where two of the six models failed entirely. These results support LungCNET as a candidate tool for lung cancer diagnosis, subject to validation on larger and independently sourced datasets.