An intelligent model approach for dynamic prediction of magnetized Jeffrey blood flow carrying penta-hybrid nanoparticles in a catheterized electrified arterial annulus

This research paper presents an artificial intelligence (AI) framework to predict magnetized penta-nanoparticle-enhanced Jeffrey blood flow dynamics in a catheterized electrified arterial annulus. The work addresses critical gaps in modeling non-Newtonian blood rheology with multi-physics interactions. Employing Jeffrey's fluid model to encapsulate the non-Newtonian rheological properties of blood mixed with nanoparticles. This analysis combines diverse factors influencing heat sources, Joule heating, interfacial nanolayers, and porous media drag. The flow system is streamlined via lubrication theory and Debye-Hückel linearization and then solved using homotopy perturbation method (HPM). Visualization of indispensable flow metrics is conducted using tools in Mathematica and Matlab. Computational results indicate electro-osmotic forces significantly alter the streaming patterns of penta-hybrid nanoparticle-infused blood in catheterized arterial geometry. Blood temperature lowers in the catheterized regions for the expanded thickness of nanolayer, and the axial blood pressure gradient elevates with an upsurge in the electro-osmotic factor while wall shear stress (WSS) abates. Heat transfer coefficient (HTC) improves with thicker nanolayers. AI-driven artificial neural network (ANN) model achieves 97–100% accuracy in predicting WSS and HTC. The research findings highlight potential improvements in patient-specific treatment strategies and contribute to the broader field of biomedical engineering by enhancing the efficacy and precision of non-invasive therapies.