Coastal wetlands are ecologically significant ecosystems that provide essential services such as flood control, carbon sequestration, and biodiversity support. However, they are increasingly threatened by climate change, urbanization, and human activities. Effective monitoring and classification are crucial for conservation and sustainable management. Traditional methods often struggle with accuracy and scalability, but remote sensing and machine learning have emerged as powerful tools for large-scale, high-precision wetland mapping. The random forest algorithm is used to classify Level-2 water bodies in coastal wetlands, specifically distinguishing features such as rivers and aquaculture ponds within the broader wetland landscape and assess accuracy using key performance metrics. This study utilizes high-resolution satellite imagery from Sentinel-2 and Landsat 8 for feature extraction and wetland classification using machine learning techniques. The classification results demonstrate strong model performance, particularly for river features, which achieved an F1 score of 0.9362 based on a precision of 93.33
Plant diseases represent a major challenge to global food security, often resulting in severe yield reductions if not detected and controlled promptly. This work introduces an AI-based framework that integrates deep learning, drone-assisted imaging, and IoT-enabled real-time monitoring for effective disease detection and management. A convolutional neural network (CNN) is trained on crop leaf image datasets to accurately classify different disease types. The system is further equipped with a mobile application and cloud-based alert service to support timely farmer interventions. Experimental evaluation demonstrates a classification accuracy exceeding 95% and reliable alert generation, underscoring the system's potential as a scalable solution for precision agriculture and smart farming.
Healthcare data is emerging as the soul of the cyberattack due to information system security in the health system. AI and ML are new and promising technologies that can improve the protection of digital systems in healthcare organizations. This paper becomes a literature review of how AI and ML applications can reinforce cybersecurity in health systems. They go to how these technologies are being deployed in areas including anomaly detection, intrusion prevention, and threat intelligence and accentuate how they provide capabilities to recognize and disable risk in real-time. These concerns include data privacy, training, and high false-positive rates. This paper also examines some exemplification by giving examples of how AI and ML solutions can work in practice. Further, it suggests avenues of work, such as how the advancements in AI can be adopted in the healthcare sector and what sort of cooperation is required to create enhanced cybersecurity developments. The study points to AI and ML's impossibility of implementing sensitive health data protection and maintaining system security.
This work presents a modelling methodology, to predict sand erosion rate in deepwaters rigid jumpers, employed for the production of gas in deepwater fields, which connects a subsea tree with a PLET. Computational fluid dynamics modelling (CFD) was performed to integrate a discrete phase model, and incorporate sand produced in the gas well, in the Navier Stokes equations employed to model gas flow. The sand was incorporated in the models with a distribution of sand particle size, using a Rosin-Rammler distribution function, in the continuous gas phase flow with flow movement effects. Different volumes of gas, produced in a gas well, were incorporated in the modelling. Hence parametric modelling analyses were performed to study the effect of sand distribution sizes, sand amounts and gas volumes in order to identify the critical operational conditions that accelerate sand erosion in rigid jumpers. The results showed gas and sand velocity increment along the jumper, which is higher on the last elbow, consequently because sand impact velocity is one of the main parameters controlling sand erosion rate, the erosion was more severe on the last elbow. Theoretical sand erosion analyses were also performed, and it was observed that CFD modelling produced more realistic sand erosion prediction in the jumper. This occurred because theoretical erosion models do not incorporate flow velocity increment through the jumper, drag forces, particle flow interaction effect along the jumper, etc. Theoretical models generally consider constant particle size, which over predict sand erosion rate, as clearly shown by CFD modelling.
