Nanofluid flow and heat transfer are important and their study in the biomedical application, especially drug delivery and thermal therapy is of paramount importance because it can enhance the medical results of diagnosis and treatment. The paper will set out to explore heat transfer and fluid flow characteristics of blood-based nanofluids to which both Uranium dioxide ( \(\:U{O}_{2}\) ) and Thorium dioxide ( \(\:Th{O}_{2}\) ) nanoparticles have been dispersed in the presence of mixed convection and thermal radiation over a stretching surface. The flow is considered steady, laminar, and two-dimensional with blood being a couple-stress non-Newtonian fluid and Rosseland approximation of thermal radiation being adopted. Similarity transformations are used to reduce the governing partial differential equations to nonlinear ordinary differential equations which are solved numerically using the MATLAB bvp4c solver, and an artificial neural network with the Levenberg-Marquardt algorithm is trained on numerical data to allow fast predictions. The findings indicate that a larger exponent of velocity, couple stress, and temperature exponent will decrease the fluid velocity, and a larger exponent of radiation and volume fraction will enhance the distribution of temperature, and Blood- \(\:U{O}_{2}\) exhibits better thermal performance than Blood- \(\:Th{O}_{2}\) . The ANN has very high predictive errors of less than 0.17% and is 425 times faster than the numerical solver which makes it an efficient and trustworthy model to analyze nanofluid flow in biomedical and thermal engineering.
Graphical Abstract
