Neural Network-Based Sensitivity Analysis of EMHD Carreau-Gold Bionanomaterial Flow under Radiative-Reactive Conditions

Cardiovascular diseases and cancer remain critical threats to human health. The present study elucidates the biomedical application of Carreau bionanomaterial flow over a stretching cylinder, considering gold nanoparticles and external electro-magnetic fields that align with sustainable development goal (SDG) “Good Health and Well-Being”. The shear-thinning nature of the Carreau bionanomaterial has been considered for its similarity with human blood. The governing equations, derived from conservation principles and a modified Buongiorno framework, accounting for radiation, higher-order chemical reaction, convective heating, nonuniform heat source, and second-order slip, are transmuted through similarity transformations to a coupled nonlinear set of ordinary differential equations. These equations are solved numerically using MATLAB's bvp5c solver and validated against limiting cases. Results indicate that stronger magnetic fields and weaker electric fields retard flow, facilitating improved blood-flow regulation, and targeted drug delivery. Neural network-driven sensitivity analysis and a multiple linear regression model, based on the modified Garson algorithm, highlight the relative significance of pertinent factors on the heat transfer rate. The results reveal that the Biot number and volume fraction of gold nanoparticles exhibit the strongest positive and negative associations, respectively.