Directory of Experts
Back to search results

Research project title

GAI-ORKG : Generative AI for Oncology Research with Knowledge Graphs / Intelligence Artificielle Générative pour la Recherche en Oncologie avec Graphe de Connaissances (Post-doctoral fellowship)

Education level

Post-doctoral fellowship


Director: Amal Zouaq

End of display

July 19, 2025

Areas of expertise

Artificial intelligence

Medical sciences

Unit(s) and department(s)

Department of Computer Engineering and Software Engineering

LAMA-WeST (Web, Semantics and Text)


The LAMA-WeST laboratory ( is in the process a post-doctoral student for a project in artificial intelligence and digital health.

The student will collaborate with the LAMA-WeST team at Polytechnique, with the LAMA-WeST lab at MILA  and with our McGill team.

The post-doctoral candidate must hold a doctorate in natural language processing / machine learning and Semantic Web (creation of ontologies, knowledge bases, etc.). He or she should have an in-depth knowledge of Python programming. Experience in the field of AI and health is a plus.

The candidate must be passionate about AI research and will contribute to the methodology of the project, the implementation of certain models, the supervision of doctoral and master's students, and the writing of articles. Leadership, as well as oral and written communication skills, are also required.Please send to  a CV, a transcript, as well as a letter motivating how your past experience can contribute to this project. Please indicate in the subject of the message: Postdoc - GAI-ORKG: Generative AI for oncology research with Knowledge Graph.

Detailed description

Over the last two decades, healthcare has moved from a paper-based reality to a digital one and a trove of digital health data now exists. Simultaneously, an era of AI has dawned with benefits to many areas of society. But, the unstructured and siloed nature of a lot of health data mean these parallel developments have barely converged and the benefits of AI in healthcare remain, as yet, unrealized. This is particularly true in cancer care. For many cancer patients, important information is buried in clinical notes in disparate parts of their electronic health record. Likewise, useful information, that could otherwise contribute to AI-powered cancer research, lies trapped and inaccessible to researchers. A solution to combine, consolidate, and exploit unstructured health data is needed.

To achieve this objective, the research team will leverage modern standards for health data to build/learn a cancer patient knowledge base (i.e. a fully-structured record for each patient) from both structured and unstructured data in electronic health records. We will investigate how neural architectures, pretrained language models, and knowledge graphs can be used to extract such a knowledge base and provide relevant information to specialists through natural language generation approaches.

Professors involved in the project:

Polytechnique Montréal :  Amal Zouaq
McGill :  John Kildea

Financing possibility

Funding available.