Public defence of the PhD thesis of Ghada Ben Othman
On 6 May 2026, Ghada Ben Othman will publicly defend the PhD thesis Sparse Data Integration for Safe and Reliable AI-Driven Decision Making in Resource-Limited Clinical Settings.
The research addresses a key challenge in modern healthcare: clinical measurements are often incomplete, irregular, or short in duration, which limits the practical use of data-driven decision support tools.
This thesis presents a sparsity-aware AI framework that transforms limited measurements into reliable and actionable information across a range of real-world scenarios. The work combines time-series forecasting, patient clustering for more data-efficient personalisation, artefact detection in wearable and clinical signals, and windowed learning to connect sparse subjective reports with continuous physiological data.
In addition, the thesis introduces digital-twin generation using conditional generative models to simulate realistic perioperative trajectories when real data are unavailable, enabling safe what-if testing and further method development without the need for new large-scale trials.
Where: Online or Campus Zwijnaarde, lecture room galileo ferraris (building131 volta)
Wanneer: Wednesday 6 May at16h
Participation is free but registration is mandatory. Register here.