PhD position in AI and multimodal digital liver pathology

Doctorat / PhD Type de contrat : ["FULL_TIME","TEMPORARY"] Date limite de candidature : 2026-11-18T00:00:00.000Z

Company description A full-time PhD position is available for an enthusiastic and highly motivated researcher to develop artificial intelligence methods for the multimodal analysis of chronic liver disease and liver cancer. The position is based in the Govaere Research Group at the Department of Imaging & Pathology, Faculty of Medicine, KU Leuven, and is embedded in a close collaboration with Prof. Jeroen Dekervel, with co-supervision by Prof. Asier Antoranz. The Govaere Research Group is part of the Laboratory of Translational Cell and Tissue Research and is fully embedded in the pathology department at KU Leuven and University Hospitals Leuven. We study chronic liver disease using patient-derived tissue, digital pathology and molecular profiling, with strong links to hepatology, oncology, pathology, radiology and computational biology. The group is also connected to the KU Leuven Institute for Single Cell Omics (LISCO). https://govaerelab.com/ Job description The project aims to develop and validate AI models that combine information from digitised histopathology slides with molecular datasets to predict biologically and clinically relevant features of liver disease. Where suitable datasets are available, MRI-derived information may also be incorporated to extend the approach towards multimodal prediction.The research will span the inflammation-to-cancer continuum, including metabolic dysfunction-associated steatohepatitis (MASH), alcohol-associated steatohepatitis (ASH) and hepatocellular carcinoma (HCC). Rather than focusing on a single cell type or disease stage, the project will investigate how tissue morphology, molecular programmes and imaging phenotypes jointly capture disease mechanisms, progression and heterogeneity (reference Boesch M et al Nature Genetics 2026). Research will include: Curation, quality control and harmonisation of whole-slide images and linked clinical and molecular datasets. Development of computational pathology workflows using deep learning, self-supervised learning, foundation models and/or multiple-instance learning. Integration of image-derived features with transcriptomic, spatial-omics, genomic, proteomic or other molecular profiles. Exploration of multimodal models incorporating radiological data, particularly MRI, where appropriate. Biological interpretation and visualisation of model predictions in close interaction with pathologists, clinicians and molecular researchers. Robust evaluation of model performance, generalisability and interpretability using reproducible and well-documented computational workflows. Key tasks Complete a doctoral training programme and conduct original research leading to a PhD in AI and digital pathology for chronic liver disease and HCC. Develop, train and evaluate machine-learning and deep-learning models for whole-slide pathology and multimodal biomedical…