The goal of this PhD is to modernize and leverage numerical Abdominal Aortic Aneurysm (AAA) simulations with Artificial Intelligence (AI) to make their use easier in the medical community and the procurement of the biomechanical patient-specific data possible within less than one minute, which is a real challenge. On the fundamental science side, and as mentioned above, our objective is to investigate state-of-the-art data-driven identification algorithms for high-dimensional spatio-temporal problems from previous high-fidelity simulations, possibly readjusted with patient data/measurements. We hope they will enable live visualizations of quantities and fields of interest (stress fields, risk assessment according to the evolution) for decision-making (surgical decision, surgical procedure).
Current computational Physics-based solvers can provide accurate results but are very time-consuming. For medical use, a breakthrough is needed to lower the computational time by at least 3 or 4 orders of magnitude. We believe that a smart synergy between Full-Order models (FOM) and AI-based strategies can take up this challenge.
This exploratory project adopts an interdisciplinary approach including biomechanics, scientific computing, advanced AI/ML methods, data science and medical applications.
UTC-CNAM supervisors:
- F. De Vuyst (UTC, principal supervisor) is Prof. PU1 CNU 26, affiliated with the Biomechanics and Bioengineering lab. His areas of expertise include numerical analysis, computational methods for PDEs, and general approximation methods. Part of his research focuses on MOR techniques and potential synergies between ROM methods and ML for fluid mechanics and fluid-structure interaction problems. He is responsible for the MSc Degree program "Complex System Eng." at UTC where he teaches SML. For the project, he will contribute his knowledge and skills in ROM (e.g. geometry reparametrization) and SML (AE, possibly GNN for mesh and field interpolation).
- I. Mortazavi (CNAM, co-supervisor) is Prof. PUEX1 CNU 26, affiliated with the M2N lab (former Director), expert in computational methods for Fluid Mechanics, especially for Navier-Stokes equations and turbulence modeling. He is also strongly involved in dimensionality reduction techniques and ROMs (POD, Spectral POD, (E)DMD, Koopman theory, quadratic manifold learning, ...).
- A.-V. Salsac (UTC, co-supervisor) is DR2 CNRS in Section 10, specialized in biofluids applied to the study of hemodynamics from the large blood vessels to the microcirculation, and of endovascular treatments. She received various prestigious prizes and projects (e.g. Medal of the National Order of Merit in 2016, ERC Consolidator grand in 2018, co-PI of a CAP of PostgenAI@Paris in 2024). Her research combines highly fundamental studies with applied aspects (strong collaborations with clinicians and industries such as ANSYS, Guerbet).
