Key result
POD-ANN models accelerate hemodynamic computations ~100,000x versus full order models with minimal error.
Why the study?
Previous reduced order models for coronary artery disease lacked integration of finite volume methods in patient-specific configurations, data-driven techniques, and free form deformation mesh strategies.
A novel machine learning-based reduced order model provides an efficient computational approach for evaluating patient-specific hemodynamics in coronary artery bypass grafts.
May enable faster patient-specific CABG hemodynamic assessment; extends reduced-order modeling but leaves open prospective clinical validation.
In this work the development of a machine learning-based Reduced Order Model (ROM) for the investigation of hemodynamics in a patient-specific configuration of Coronary Artery Bypass Graft (CABG) is proposed. The computational domain is referred to left branches of coronary arteries when a stenosis of the Left Main Coronary Artery (LMCA) occurs. The method extracts a reduced basis space from a collection of high-fidelity solutions via a Proper Orthogonal Decomposition (POD) algorithm and employs Artificial Neural Networks (ANNs) for the computation of the modal coefficients. The Full Order Model (FOM) is represented by the incompressible Navier-Stokes equations discretized using a Finite Volume (FV) technique. Both physical and geometrical parametrization are taken into account, the former one related to the inlet flow rate and the latter one related to the stenosis severity. With respect to the previous works focused on the development of a ROM framework for the evaluation of coronary artery disease, the novelties of our study include the use of the FV method in a patient-specific configuration, the use of a data-driven ROM technique and the mesh deformation strategy based on a Free Form Deformation (FFD) technique. The performance of our ROM approach is analyzed in terms of the error between full order and reduced order solutions as well as the speedup achieved at the online stage.
No takes yet. Share an insight, caveat, or question.
Siena et al. (2022) studied Coronary artery disease (LMCA stenosis). Proper Orthogonal Decomposition-Artificial Neural Network (POD-ANN) Reduced Order Model vs. Full Order Model (Navier-Stokes equations discretized using Finite Volume technique) was evaluated on Relative error between full order and reduced order solutions and computational speedup. The POD-ANN reduced order model achieved a computational speedup of at least 10^5 compared to the full order model while maintaining a time-averaged error of 2-5% for hemodynamic variables.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: