진학프로석·박사 채용 접수 플랫폼

Mission

The successful candidate will contribute to the development of a digital twin of the heart and blood circulation and will be responsible to design, train and optimise artificial intelligence models capable of predicting the effects of different treatments for mitral insufficiency and ventricular failure with significantly reduced computing time.

Activities

• • Collect and process patient-specific data (e.g., imaging, haemodynamics) to build a substantial database
• • Become proficient in the use of digital twins
• • Design and integrate artificial intelligence methods (neural networks) for predicting and optimising treatments
• • Create interactive graphical user interfaces (GUIs) for monitoring and visualising simulated data
• • Set up a complete pipeline (simulation, visualisation, interpretation) to facilitate translational research
• • Handle contacts with clinicians and manufacturers
• • Participate in meetings and promote the results obtained (reports, publications, presentations)
• • Disseminate the results obtained through publications in peer-reviewed journals and by participating in international conferences.

Background

The ‘Biological Fluid-Structure Interactions’ (IFSB) team is developing a digital twin of the heart and blood circulation with the objective to improve the management of heart disease, and in particular the treatment of mitral insufficiency and ventricular failure. The digital twin is based on complex equations (equations of continuum mechanics and fluid mechanics) and requires the integration of patient-specific data (heart geometry, haemodynamic measurements, etc.).

While the digital twin represents a high-fidelity model of the patient's reality, it is associated with significant costs in terms of resources and computing time. It is therefore not suitable for routine clinical use.

One strategy to provide clinicians with a tool that is compatible with clinical use and that reduces computational costs is artificial intelligence (AI). It will enable the design of an enriched model while ensuring reasonable computing time.

The research project is funded by the PostGenAI@Paris cluster, which brings together academic partners with diverse expertise, ranging from mathematics and computer science to health, engineering, law, and political science. The consortium benefits from extensive support from industrial and social partners—comprising over 60 institutions, including startups and tech giants. The PostGenAI@Paris cluster addresses fundamental challenges posed by the most recent advances in AI, wherein the interface between computing technology and human intelligence is increasingly blurred. This era of “post-generative AI” offers remarkable benefits to society, but brings with it a host of new uncertainties related to reliability, regulation, and ethics. The PostGenAI@Paris cluster is led by the Sorbonne Cluster for Artificial Intelligence (SCAI). Launched in January 2025, PostGenAI@Paris is a five-year initiative funded under grant ANR-23-IACL-0007.

The research project will be part of the « AI, Images and Models for Medicine » Collaborative Acceleration Program (CAP 2), which is one of the programs developed within the PostGenAI@Paris cluster.