https://www.um.edu.mt/library/oar/handle/123456789/2068 Wed, 24 Jun 2026 13:34:09 GMT 2026-06-24T13:34:09Z https://www.um.edu.mt/library/oar/handle/123456789/147489 Title: A LiFi-based innovative 6G solution for hospitals using green wavelength, directly modulated laser Authors: Sharma, Ajay; Xuereb, Peter A.; Garg, Lalit Abstract: This paper proposes an innovative light-fidelity (Li-Fi) system for high-speed communication in hospital environments that operates at a green wavelength of 500 nm with Directly Modulated Laser (DML). The proposed system shows an excellent performance and achieves a Q factor of 18.84, a bit error rate (BER) of 1.6e-79, and a signal-to noise ratio (SNR) of 74.94 dB, which is significantly better than the previous research. It also has a range of up to 25 m line-of-sight (LOS) and can transfer data at speeds in excess of 1 Gbps, making it significantly faster than previous work conducted with much lower LOS ranges while being robust against interference. New applications of DML combined with optical splitters contribute to providing signal stability and system scalability, overcoming problems such as low range. This design ensures safe, reliable, and non-intrusive communication, ideal for applications that require high data reliability, such as real-time imaging and telemedicine in hospitals. This new Li-Fi system is found to be compatible with modern hospital power requirements, and it also provides a solid foundation for future 6G communication networks. Thu, 01 Jan 2026 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/147489 2026-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/147402 Title: Enhancing hospital security and patient monitoring through WhoFi-inspired LiFi channel sensing with privacy preservation Authors: Sharma, Ajay; Garg, Lalit; Xuereb, Peter Albert Abstract: Ensuring both secure connectivity and patient safety has become a growing concern in modern hospitals. Although LiFi (Light Fidelity) provides high-speed and interference-free communication, the possibility of using it as a sensing platform has not been investigated to the full extent. This paper presents a LiFi-based system that integrates WHOFi for hospital security and patient monitoring. With simulations based on MATLAB, we simulate the change in LiFi channels due to human presence, movement, and falls and extract statistical and spectral characteristics of the machine learning classifier. The system has a high accuracy of around 94% in activity recognition (empty, movement, fall) and the Equal Error Rate (EER) of 5% in staff authentication. Such a solution is privacy-sensitive, non-invasive and inherently limited to the room boundaries, unlike camera-based or wearable systems, which increase the level of security and patient monitoring in healthcare settings. The findings point to the two-fold nature of LiFi as a communication and sensing technology, which opens the potential for smart hospital infrastructures. This numerical evaluation study will be expanded to hardware testbeds and deep learning models to be applicable in the real world in the future. The proposed system enhances hospital data security and patient tracking efficiency using optical wireless communication. Sun, 01 Feb 2026 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/147402 2026-02-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/146931 Title: Leveraging invariant prediction for mitigating specificity constraints in affect modelling Abstract: Affect modelling aims to predict human emotional states from multimodal signals, yet current approaches often struggle to generalise beyond the specific datasets or contexts in which they are trained. This dissertation investigates the use of in variant features, predictors whose relationship with affective states remains stable across distinct environments, as a strategy to improve generalisability. To this end, two publicly available corpora, AGAIN and RECOLA, were systematically parti tioned into environments defined by user, task, and annotator triplets. An envi ronment refers to the conditions under which data is collected, and data gathered within the same environment is assumed to come from the same underlying distri bution. The Invariant Causal Prediction (ICP) framework was employed to identify stable features across these environments, which were then compared against full feature sets and principal components derived through PCA. Three supervised learning models—Logistic Regression, a feed-forward Neural Network,and a Long Short-Term Memory (LSTM)network — were trained under all three feature conditions, using group-based cross-validation to avoid information leakage. Results demonstrate that invariant features can deliver measurable benefits for feed-forward models, particularly in enhancing accuracy and correlation while substantially reducing feature dimensionality. However, their advantages were less consistent for sequence models like LSTMs, where temporal dependencies were not fully captured by invariants alone. Statistical significance tests further showed that invariant features improved balanced classification (F1) more strongly in AGAIN than in RECOLA,underscoring the dataset-specific nature of their effectiveness. Overall, the findings highlight both the promise and the limitations of invariance in affect modelling. While not a universal solution, invariant features represent a principled means of isolating robust predictors across heterogeneous contexts, contributing to the broader goal of developing affective systems that are reliable, interpretable, and adaptable across diverse real-world settings. Description: M.Sc.(Melit.) Wed, 01 Jan 2025 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/146931 2025-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/146930 Title: Siamese network‐based vector embeddings of MRI scans for twin identification Abstract: Monozygotic twins are identical twins that develop from a single fertilised egg that spontaneously splits, resulting in two individuals sharing 100% genetic material. Identifying monozygotic twins from brain MRI scans represents a frontier challenge in computational medical imaging with significant implications for understanding genetic influences on neuroanatomical structure through direct pattern recognition. While classical twin studies using ACE models decompose statistical variance to establish independent regional heritability estimates (60‐80%), this study introduces a fundamentally different computational framework that learns directly from MRI data to rank neuroanatomical regions by their collective discriminative capacity for genetic similarity detection, complementing traditional statistical approaches through data‐driven analysis. Adeep learning methodology employing Siamese networks with 3D CNN backbones is developed for automated twin identification using 138 genetically verified monozygotic twin pairs (276 subjects) from the Human Connectome Project S1200 dataset. Modified U‐Net, ResNet, and DenseNet architectures generate 128‐dimensional embeddings optimised via triplet loss with hard negative mining, forcing models to learn subtle genetic signatures by focusing on challenging discriminative examples that distinguish twins from their most similar morphological matches. U‐Net achieved superior computational performance with 92.0% F1‐score (σ = 2.5%), 95.2% AUC‐ROC, and 91.4% accuracy, while ResNet demonstrated competitive results (89.6% F1‐score) and DenseNet showed greater variability (88.5% F1‐score). Embedding analysis reveals clear bimodal separation between genetically related and unrelated individuals through learned morphological patterns. Layer‐Wise Relevance Propagation analysis provides the first data‐driven ranking of neuroanatomical regions by discriminative importance for genetic relatedness detection. Statistical analysis reveals pronounced subcortical dominance with large effect size (Cohen’s d = 2.80, p = 3.89e‐6), with six subcortical structures occupying top positions, including the thalamus (0.955), brainstem (0.875), and hypothalamus (0.707). This computational hierarchy contrasts with traditional ACE studies reporting highest heritability in cortical areas (frontal 78‐95%, temporal 77‐89%), demonstrating that direct pattern recognition from MRI data identifies different neuroanatomical signatures than statistical variance decomposition. Notably, models utilise practically all brain regions (most importance scores > 0.2), indicating distributed multivariate processing rather than selective regional dependence. Ablation studies confirm data augmentation’s critical role, with substantial i performance improvements across CNN architectures. Clinical integration through standard neuroimaging formats in Connectome Workbench demonstrates immediate practical utility, positioning this computational approach for adoption in research and clinical environments requiring direct analysis of genetic influences in brain structure. The framework advances precision neuroimaging by providing automated, quantitative genetic similarity detection through direct pattern recognition, revealing spatial insights that complement traditional heritability studies while offering methodological advances applicable to diverse medical imaging classification tasks requiring regional discriminative analysis Description: M.Sc.(Melit.) Wed, 01 Jan 2025 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/146930 2025-01-01T00:00:00Z