The prevalence of metabolic associated steatotic liver disease (MASLD) has increased significantly over the past years. As this condition progresses, inflammation and liver damage can occur, leading to liver cirrhosis – the scarring of the liver. A scarred liver increases the resistance blood needs to overcome to pass through the organ. As a consequence, blood pressure increases in the portal vein, one of the vessels bringing blood to the liver, leading to portal hypertension. The body adapts by generating collateral vessels deviating blood from the detoxifying liver, and by pumping more blood which eventually can damage the heart. The TIPS procedure aims at lowering this pressure by adding an artificial shunt that further deviates blood from the liver. However much remains to understand in the liver-heart interaction and in optimizing the TIPS procedure.
The PhD thesis will thus aim at better characterizing geometrically and hemodynamically in 3D these natural and artificial shunts, predicting heart problems and eventually optimizing the TIPS procedure. This will be achieved by developping appropriate deep-learning and mechanistic (3D fluid mechanics) models.
Pavlos Varsos, Friederike Schäfer, Cristina Ripoll, Nicolas Golse, Irene E Vignon-Clementel. Hemodynamic insights into TIPS intervention for portal hypertension management: a comprehensive computational study. Submitted for publication. 2026. hal-05630883
Francesco Songia, Raoul Sallé de Chou, Hugues Talbot, Irene Vignon-Clementel. Multi-fidelity graph-based neural networks architectures to learn Navier-Stokes solutions on non-parametrized 2D domains. Submitted for publication. 2026 hal-05426284v2
Raoul Sallé de Chou, Matthew Sinclair, Sabrina Lynch, Nan Xiao, Laurent Najman, et al.. Finite Volume Informed Graph Neural Network for Myocardial Perfusion Simulation. Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, 2024 pp.276-288. hal-04828473
Nicolas Golse, Florian Joly, Prisca Combari, Maïté Lewin, Quentin Nicolas, et al.. Predicting the risk of post-hepatectomy portal hypertension using a digital twin: A clinical proof of concept. Journal of Hepatology, 2021, 74 (3), pp.661-669. 10.1016/j.jhep.2020.10.036 . hal-03523641
Fall 2026 (October - December)
You will be located at Inria Saclay Ile-de-France in the SimbiotX team, supervised by Irene Vignon-Clementel, a deep-learning expert and clinicians. You will be working together with the postdoc Friederike Schäfer and PhD student Francesco Songia.
Would you like to get more information about the project or the team, please contact the responsible persons mentioned below.
Are you convinced this position suits you? Apply online with your CV, motivation letter and grades, or contact us for more information:
The position will be filled as soon as the right candidate is found.
