AI-assisted hemodynamic analysis of magnetohydrodynamic Au–blood nanofluid flow in stenosed artery

It illustrates an AI-Assisted Hemodynamic analysis magnetohydrodynamic (MHD) movement of an Au -blood nanofluid in stenosed artery. The arterial geometry is simulated to represent arterial stenosis, and the impact of a magnetic field and viscous dissipation on the blood flow development is examined. The mathematical model is constructed for the analysis of velocity and temperature of blood dynamics in stenosed artery by the consideration of magnetic and viscous dissipation effects. The N subsequent nonlinear models are solved by using a collocation-based boundary value problem solver (bvp5c) in MATLAB. The findings show that, as Pr is increased, there is a reduction in the thickness of the convective thermal boundary layer, whereas the values of Ec were high and thus, a large contribution to the temperature field is produced by viscous heating. Addition of gold nanoparticles enhances the thermal conductivity of blood, which is effective and thereby increases heat transfer in the stenosed area. The core scientific insight of this study is that the combined action of magnetohydrodynamics and Au nanoparticles enables controlled thermal enhancement through Lorentz-force-modulated flow and improved conductivity, while the ANN model efficiently captures the nonlinear interactions among governing parameters, providing a fast and reliable predictive framework. Moreover, an ANN model (FFNN) is also formulated and trained with the help of the numerical data. The ANN model demonstrates high predictive accuracy with RMSE values of 0.0013 for velocity and 0.0022 for temperature, indicating minimal numerical deviation. Correlation coefficients of 1.0000 for velocity and 0.9999 for temperature confirm excellent agreement with bvp5c results. Novelty of this study is the development of a hybrid numerical–ANN model for MHD Au–blood nanofluid flow in stenosed arteries, enabling controlled thermal enhancement and providing a reliable, computationally efficient predictive tool for biomedical heat transfer and magnetic therapy applications.