Funded PhD: Machine Learning in Cardiovascular Haemodynamics

Outline: Haemodynamics modelling is essential for prognostics of cardiovascular diseases. It is conventionally simulated using finite element methods and can take days to produce high fidelity, clinically relevant predictions, which challenges their deployment. The technique of Physics Informed Neural Networks (PINNs) has since shown promising results in surrogate modelling of physical processes like elasticity, wave propagation, blood flow, etc. The aim of this project is to develop PINNs to model the blood flow in aorta to be validated using the experimental measurements like 4D flow MRI dataset or the insilico dataset.   

Qualifications: Aspiring candidates must have the motivation to work at the confluence of Mathematics, Mechanics, Statistics and Computer Science to develop novel theoretical models and numerical methods for Haemodynamics. Subject knowledge of Partial Differential Equations, Fluid Mechanics, Numerical Analysis, Python, C/C++, machine learning is desirable. Experimental skills will also be beneficial.

To Apply:  Applications, to include a    

1. two-page CV,   
2. motivation letter (including details of the past experience/courses relevant for this project), 
   
3. contact details of two referees, and
4. transcripts

should be sent, via e-mail (in one PDF only) to bharat.tripathi@universityofgalway.ie.