Physics-informed machine learning based design for 4D-printed biomedical devices

This PhD topic is funded by the French Research Agency (ANR) and is part of a collaborative research project focused on Multifunctional and multi-sensitive copolymers for 4D printed structures in biomidical field. In this project a novel stimuli-responsive material will be synthesized and tuned for 3D/4D printing by a partner involved in chemistry. This PhD proposal will leverage the obtained experimental data related to original material properties depending upong thermal and photo stimulations through the development of numerical models for simulation. Numerical simulation enables precise analysis and prediction of the material’s response to the targeted stimuli. We will also leverage the performance of the synthesized active material and its numerical models to design a deployable stent, able to hold force of a biliary or intestinal duct section. To achieve this, we will address the inverse design problem using physics-informed machine learning that consists of determining the optimal structure and material distribution to fulfil the desired and sequenced shape changes.

ICB UMR 6303 CNRS lab at Belfort-Montbéliard University of Technology is composed of six research departments in which CO2M department – leads research efforts on design, modeling and optimization of mechanical systems. In such a department, three issues are currently addressed:

• Mechanical modeling and optimization.
• Manufacturing processes and techniques optimization.
• Advanced design of mechanical systems.

Expected soft skills: autonomy, divergent thinking, curisosity.

Expected technical skills: Mechanical Modeling, Simulation, Design, 3D printing, Artificial Intelligence, Programming.

Expected communication skills: English