The novel theoretical tetra-hybrid nanofluid (tehnf) prototype presented in this paper focuses on enhancing heat transmission. The tetra-hybrid Cu, Ag, \({{Al}_{2}O}_{3},\) and TiO2 are immersed in blood. The ordinary differential equations (ODEs) derived from the transformation of the fluid flow and temperature equations are solved using similarity variables in combination with the Keller box method (KBM). This approach effectively reduces the complexity of the problem, enabling accurate numerical solutions. The Keller box technique is employed to handle the resulting system of equations with stability and precision. According to the results, the tehnf exhibits a higher thermal conductivity than Trihnf. When thermal conduction and radiation impact are increased, the heat transference rate rises, aiding in the removal of harmful plaque from the blood passing through arteries. Increasing the flow power index of the non-Newtonian Sisko model leads to improved convective heat transfer. Rigid arteries that allow blood to flow through them are opened by the extra heat created by thermal radiative flux amplification. Artificial neural networks (ANNs) were utilized in this research project to investigate the impacts of heat production. Models of multi-layer perceptron networks with ten neurons in the hidden layers employ the Levenberg–Marquardt Training Algorithm (LMTA). Performance is confirmed using error analysis results, and training/testing procedures are evaluated to examine the approximate solution validation.
