https://doi.org/10.1140/epjp/s13360-025-06685-3
Regular Article
Machine learning-assisted hemodynamic analysis: physics-informed neural network for flow simulation and hybrid ANN–PSO technique for wall shear stress optimization in variable gravity
Department of Mathematics, National Institute of Technology Jamshedpur, 831014, Jamshedpur, Jharkhand, India
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
25
April
2025
Accepted:
23
July
2025
Published online:
19
August
2025
In this study, blood flow is analyzed under simulated gravity variations, using a robust machine learning approach. It has been experimentally proved that the changes in gravitational forces affect vascular resistance and cardiac output. However, accurate modeling of blood flow under variable gravity conditions remains limited. The present study addresses this gap. A deep learning-based physics-informed neural network (PINN) is employed to solve the governing equations of blood flow under these conditions. The PINN model is validated against the Keller box method (KBM), showing minimal discrepancy. Absolute error simulations demonstrate that higher numbers of hidden layers along with neurons produce more accurate results that closely match KBM output. Graphical representations of blood flow velocity and wall shear stress (WSS) are provided for different influencing parameters. Simulation results reveal that increased amplitudes of the pressure gradient enhance blood flow velocity, while an elevated threshold heart pulse frequency reduces it. Furthermore, a hybrid artificial neural network and particle swarm optimization technique is used to optimize WSS. The key parameters, including the amplitudes of various components of the pressure gradients and heart pulse frequency, are considered as input variables for optimization. The outcomes showcase both the high precision of the PINN model and an efficient optimization approach to minimize the WSS under varying gravitational scenarios.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
