Paloma Rabaey

About Me

I am a Ph.D. student at the T2K research group at IDLab Ghent. My research focuses on finding ways to include expert knowledge in deep learning models to make them more reliable for critical settings. I am especially interested in applying these ideas to practical use-cases in healthcare, to improve trustworthiness and interpretability of clinical decision support systems. My research is funded by FWO Flanders. In my free time I love to crochet, travel, read and go to the cinema, among many other things.

In February 2026, I defended my PhD on the topic of augmenting Bayesian networks with textual evidence for expert-based, uncertainty-aware clinical decision support. You can read the full dissertation here!

Research Interests

• AI for healthcare: clinical decision support systems, injection of expert knowledge into AI models
• Bayesian networks: integrate text into Bayesian networks, augment them with neural representations
• Natural Language Processing: information extraction from unstructured clinical notes, quantification of uncertainty in clinical reports
• Causal Machine Learning: causal neural networks, partial causal discovery, causal inference with Bayesian networks

Talks

• Augmenting Bayesian Networks with Textual Evidence for Expert-Based, Uncertainty-Aware Clinical Decision Support. Public defense of my PhD dissertation. [Slides]
• Clinical Reasoning over Tabular Data and Text with Bayesian Networks. Oral presentation at the 2024 Conference of Artificial Intelligence in Medicine (AIME), in Salt Lake City, Utah. [Slides]
• Why we need Neuro-Symbolic Methods for Clinical Information Extraction. Oral presentation at the Workshop “Primary Care: Electronic Scribes and More” at the 2024 Conference of Artificial Intelligence in Medicine (AIME), in Salt Lake City, Utah. [Slides]

Science Communication

Dokter Chatbot. Blog post in Dutch, written for popular science magazine EOS. [Blog post]

Publications

Powered by Jekyll and Minimal Light theme.