PhD Position: Mechanistic and Deep-Learning Models for Liver-Heart Interaction in TIPS Procedures

This doctoral project sits at the intersection of computational hemodynamics and clinical hepatology, focusing on the liver-heart axis in portal hypertension. The successful candidate will design and validate hybrid mechanistic (3D CFD) and deep-learning models to characterize natural and artificial shunts, predict cardiac complications, and optimize TIPS interventions. Work will be conducted within the EU Artemis project in collaboration with clinical partners in France and Germany, leveraging patient-specific data and state-of-the-art simulation tools. The PhD student will join the SimbiotX team at Inria Saclay, contributing to both theoretical advances and translational outputs, including peer-reviewed publications and conference presentations.

• Develop and validate hybrid mechanistic (3D CFD) and deep-learning models for liver-heart hemodynamics in portal hypertension
• Implement and verify simulation code, ensuring numerical accuracy and reproducibility
• Collaborate with clinical partners to acquire and process patient-specific imaging and hemodynamic data
• Perform patient-specific simulations and validate model predictions against clinical outcomes
• Contribute to EU Artemis project deliverables, including progress reports and workshop presentations
• Engage actively in team seminars, interdisciplinary meetings, and social activities
• Write and publish research articles in peer-reviewed journals and present findings at international conferences
• Maintain reproducible research practices and share code via open repositories where appropriate