Type de contrat : CDD

Niveau de diplôme exigé : Bac + 5 ou équivalent

Autre diplôme apprécié : Master level

Fonction : Doctorant

Created in 2008, the Inria Saclay Center is located at the heart of the Paris-Saclay scientific and technological excellence cluster, which alone accounts for 15% of French research. Serving the development of the Université Paris-Saclay and the Institut Polytechnique de Paris, the Inria Saclay center employs 80 people in research support services and 500 scientists of 54 nationalities.

Benefiting from continuous growth, the center now has a total of 42 project-teams and two in the process of being created, including 21 jointly with the Institut Polytechnique de Paris, 16 with the Université Paris-Saclay, as well as 7 Inria EPs, including one in collaboration with Onera and one with the Pôle Universitaire Centre Val de Loire. These research teams are spread over more than ten sites.

This project is part of the European Artemis project https://artemis-euproject.eu/, where the SimbiotX team of Inria-Saclay is mainly involved in the work packages on mathematical modelling and model coupling for specific clinical use cases. Our work is carried out in collaboration with many hospitals, such as AP-HP in France and Universitätsklinikum Jena in Germany.

Topic

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.

Bibliography

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

Starting date

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.

Contact and application

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:

• Irene Vignon-Clementel: vignon-clementel@inria.fr
• Friederike Schäfer: friederike.schaefer@inria.fr
• Francesco Songia: francesco.songia@inria.fr

The position will be filled as soon as the right candidate is found.

Main activities:

• Take initiatives to propose the relevant deep-learning and mechanistic models to meet the clinical needs
• Implement and verify code, run simulations as needed by the project
• Learn about the clinical context, understand the collected patient-data, perform patient-specific simulations and validate them with clinical data
• Be an active member of the EU project Artemis (online progress meetings, workshops)
• Actively participate in activities of the team (seminars, meetings, social activities)
• Write reports, generate several journals and present the results to the research group/conferences

The ideal candidate has

• scientific computing, mechanical (CFD)/electrical/computational engineering or applied mathematics background
• deep-learning and computational fluid mechanics/PDE experience
• experience in programming (Python, pyTorch or TensorFlow)
• strong analytical skills
• a taste for challenge and excellence
• good communication skills in English
• want to work in an international, multidisciplinary team
• motivated by mathematical modelling to solve clinical challenges

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training

Monthly gross salary : 2.300 Euros

Attention: Les candidatures doivent être déposées en ligne sur le site Inria. Le traitement des candidatures adressées par d'autres canaux n'est pas garanti.