معرفی
Dr. Islem Rekik is an Associate Professor in Computing at the Department of Computing, Faculty of Engineering, Imperial College London. His email is i.rekik@imperial.ac.uk, and he is affiliated with the Artificial Intelligence Network and based at the Translation & Innovation Hub Building, White City Campus, United Kingdom.
His research focuses on Biomedical Engineering and Clinical Sciences, with a strong emphasis on developing advanced machine learning techniques for medical imaging and brain connectivity analysis. Key areas include the application of Graph Neural Networks (GNNs), federated learning, and diffusion models to address challenges in data scarcity and heterogeneity.
Analysis of his recent publications reveals a trend toward trustworthy AI in healthcare (e.g., FUTURE-AI guidelines), topology-aware graph modeling (e.g., Strongly Topology-Preserving GNNs), and ethical considerations in medical imaging. Techniques like contrastive learning, hyperedge sampling, and feature-label constraints are central to his work on memory-efficient and reproducible AI systems.
He contributes to the development of innovative AI frameworks for clinical applications, including autism detection, Alzheimer's differentiation, and multiple sclerosis analysis. His work spans theoretical advancements in GNNs and practical deployments in resource-constrained healthcare settings.
He operates within the Translation & Innovation Hub Building at Imperial College London's White City Campus, which serves as a nexus for interdisciplinary research in AI and biomedical innovation. This environment supports his focus on bridging computational methods with clinical neuroscience challenges.
