Coronary artery disease represents one of the leading causes of death worldwide and, due to the huge number of people affected, it deserves particular consideration in several research areas.
In this talk, a machine learning reduced order model is proposed for the investigation of the haemodynamics in a patient-specific configuration of coronary artery bypass grafts. The full-order model is represented by incompressible Navier-Stokes equations, discretized using a finite volume technique. Both physical and geometrical parametrization are taken into account. A complete decoupling between two phases (offline and online) is the key to the good functionality of the reduced framework. The method extracts a reduced basis space via proper orthogonal decomposition and employs artificial neural network for the computation of the reduced coefficients.
Finally, several numerical results are performed to investigate the accuracy and speed-up of our approach.
Zoom link
Topic: AJS Pierfrancesco Siena - A machine learning-based reduced order model for the investigation of the haemodynamics in coronary artery bypass grafts
Time: Dec 2, 2022 14:00 Rome
Join Zoom Meeting
https://sissa-it.zoom.us/j/82004339944?pwd=QUFKWXduNURiZlNQK2NZWkRNWGZlUT09
Meeting ID: 820 0433 9944
Passcode: 601683