Artificial Neural Network Model for Radiative Heat Transfer in a Magnetized Tapered Stenosed Artery

Blood flow dynamics in stenosed arteries are significantly shaped by the interplay between the magnetic field and radiative heat transfer, which affects both the precision of diagnosis and treatment approaches. The artificial neural network (ANN) model outperforms conventional numerical techniques in solving highly nonlinear governing equations with improved adaptation to parameter changes, faster convergence, and higher computational efficiency. This serves as motivation for the proposal of an AI-driven framework to model blood flow in an inclined, tapered, stenosed porous artery using physics-informed neural networks (PINNs), a specialized class of ANNs that directly integrate physical rules into the learning process. In contrast to traditional ANNs, which only use data-driven training, PINNs incorporate governing laws into the neural network architecture to guarantee both physical consistency and computational efficiency. Under various magnetic fields and physiological situations, such as extreme arterial tilt and stenosis geometry, the ANN model accurately captures the impacts of radiative heat transfer, which lowers velocity and temperature profiles, according to the observed results. Furthermore, the ANN model offers a deeper knowledge of hemodynamic and thermal interactions by exceeding the conventional method (differential transform method) in forecasting wall shear stress and volumetric flow rates. The ANN model outperforms the conventional numerical method in terms of accuracy, as demonstrated by the comparison table and unambiguous graphical representations. By closing the gap between computational efficiency and physiological complexity, this approach enables one to create a more accurate, flexible, and scalable model that transforms the investigation of vascular heat transport and its impact on disease detection and treatment.