https://www.um.edu.mt/library/oar/handle/123456789/2068 2026-11-06T00:33:59Z https://www.um.edu.mt/library/oar/handle/123456789/150784 Title: Dispersion-free steering beam based alignment at SwissFEL Authors: Ferrari, Eugenio; Calvi, Marco; Ganter, Romain; Kittel, Christoph; Prat, Eduard; Reiche, Sven; Schietinger, Thomas Abstract: Micron-level alignment of the undulator line is required for successful operation of linear accelerator based high gain free electron lasers to produce powerful radiation at X-rays’ wavelengths. Such precision in the straightness of the trajectory allows for an optimal transverse superposition between the electrons and the photon beam. This is extremely challenging and can only be achieved via beam-based techniques. In this paper we will report on the dispersionfree steering approach implemented at SwissFEL, that helped achieving improved performance for both the hard and soft X-ray beamlines. 2023-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/150534 Title: If at first you don’t succeed : extended monitorability through multiple executions Authors: Achilleos, Antonis; Francalanza, Adrian; Xuereb, Jasmine Abstract: This paper studies the extent to which branchingtime properties can be adequately verified using runtime monitors. We depart from the classical setup where monitoring is limited to a single system execution and investigate the enhanced observational capabilities when monitoring a system over multiple runs. To ensure generality, we focus on branchingtime properties expressed in the modal m-calculus, a well-studied foundational logic. Our results show that the proposed setup can systematically extend established monitorability limits for branching-time properties. We validate our results by instantiating them to verify actor-based systems. We also prove bounds that capture the correspondence between the syntactic structure of a property and the number of required system runs. 2025-06-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/150522 Title: Novel data analytics techniques for analysing satellite imagery data, traffic and environment pollution data to predict the hospital admission rates Abstract: Accurately forecasting hospital admissions is critical for optimizing healthcare resource allocation and emergency preparedness. This study presents a comprehensive, data-driven framework for predicting emergency department (ED) admissions related to respiratory, circulatory, and all-cause health conditions using local ground stations and satellite-based meteorological and air pollution data. Despite growing evidence linking environmental factors to hospital admissions, traditional hospital management systems lack integration with environmental intelligence and encounters problem in utilizing resources. To bridge this gap, our research employs advanced machine learning and deep learning models including tree-based, neural network, and regularized regression techniques to capture complex, nonlinear relationships between environmental exposure and health outcomes. A key contribution of this work is the introduction of three novel prediction schemes: Deep Explainable AI For Respiratory-Based Prediction (DXAIRP) for respiratory admissions, Deep Explainable AI For Circulatory-Based Prediction (DXAICP) for circulatory admissions, and Explainable Extreme Gradient Boosting Scheme for Environmental Impact on All-Cause Emergency Admissions (BGXAI) for all-cause admissions. These schemes were rigorously evaluated across multiple datasets, demonstrating high predictive performance and reliability. The analysis revealed that air pollution parameters (PM2.5, PM10, CO, NO2, and O3) are the most significant predictors of hospital admissions, while meteorological factors such as wind speed, temperature, and relative humidity also contribute, albeit to a lesser extent. The study further identified optimal lag times for different admission types, with a two-day lag providing the highest accuracy for respiratory and circulatory admissions, and a three-day lag optimizing all-cause admission predictions. Additionally, results indicate that ground-based environmental data yield superior predictive accuracy compared to satellite-derived measurements. To ensure interpretability and trust in AI-driven predictions, the framework integrates SHapley Additive exPlanations (SHAP), an Explainable AI (XAI) technique that elucidates the influence of each environmental factor on model outputs. The Extreme Gradient Boosting Regressor (XGBR) and Multi-Layer Perceptron Regressor (MLPR), and their ensembling with XAI establishes a robust, transparent, and adaptable predictive approach. Beyond its academic contributions, this research has significant practical implications. The predictive framework can be integrated into hospital management systems to support data-driven decision-making, enabling healthcare administrators to anticipate ED surges and optimize resource distribution. This framework includes ground station-based data to capture localized environmental conditions and also integrates satellite-based environmental data, making it particularly valuable for regions with limited ground monitoring infrastructure, thereby addressing spatial and temporal gaps in environmental health monitoring. Overall, this research advances climate-health analytics by offering a scalable, interpretable, and data-driven methodology for predicting hospital admissions based on environmental factors. The findings contribute to the growing field of AI-driven healthcare forecasting, with potential applications in national health systems, expansion to other disease categories, and deployment in resource-limited settings through cloud-based infrastructures. Description: Ph.D.(Melit.) 2025-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/150298 Title: Image denoising in low-light conditions Abstract: This project presents a concise comparison of classical and modern image denoising techniques, addressing the problem of restoring images degradated by noise while preserving perceptual and structural detail. The primary goal is to evaluate and contrast representative classical algorithms with contemporary deep-learning models across controlled synthetic noise levels and a curated real-world photograph dataset. Experiments measure restoration fidelity, inference time and storage footprint using standard image quality metrics (peak signal-to-noise ratio and structural similarity index), runtime benchmarks and recorded model weight sizes, and the evaluation scripts and per-image results are provided in the repository to ensure reproducibility. Results demonstrate that deep networks typically achieve higher fidelity on synthetically degradated images and superior qualitative restoration on many real-world examples, whereas classical methods remain competitive at lower noise intensities and require substantially less computational and storage resources. The report discusses practical trade-offs including latency, model size and quantisation considerations, and concludes with recommendations for selecting denoising approaches in research and deployment contexts. All generated tables and code used to produce the results are available in the project repository. Description: B.Sc. (Hons)(Melit.) 2026-01-01T00:00:00Z