PhD on AI-driven multimodal data fusion models in rare diseases

KU Leuven is seeking a doctoral candidate to join the Department of Human Genetics for a four-year PhD project focused on AI-driven multimodal data fusion. The research aims to develop and benchmark zero-shot machine learning models for diagnostic prediction in rare diseases, specifically using melanoma as a primary use case. The successful candidate will integrate spatial and single-cell multi-omics data with histopathology and clinical information to create robust, explainable cross-modal representations. This position is part of the MSCA Doctoral Network SPACE-MEL and involves interdisciplinary collaboration with academic and industry partners in the United Kingdom. The fellow will have access to advanced computational infrastructure and state-of-the-art omics technologies through the Leuven Institute for Single-cell Omics.

• Develop and benchmark zero-shot multimodal fusion models for rare disease prediction.
• Integrate spatial and single-cell multi-omics data with histopathology and clinical information.
• Design generative and explainable AI strategies for complex biological dataset analysis.
• Extend existing multimodal frameworks with additional omics layers and evaluate performance across heterogeneous datasets.
• Collaborate with international academic and industry partners through scheduled secondments.
• Maintain reproducible research workflows using version control and containerization tools.