Computational fluid dynamics based Taguchi analysis on shear stress in microfluidic cerebrovascular channels

The cerebrovascular blood vessels feed necessary agents such as oxygen, glucose, and so forth. to the brain which maintains the smooth functioning of the human body. However, the blood-brain barrier as a vascular border restricts the entry of drugs that can be necessary for the treatment of neurological disorders. The fluid shear stress in the cerebrovascular blood vessels may regulate the drug delivery in the interface between the cerebrovascular blood vessels and the brain. The intensity of influence by various factors that affects the shear stress in the cerebrovascular blood vessels is scarcely addressed in the present study. A hybrid approach of computational fluid dynamics and Taguchi analysis is proposed to evaluate the influence of various geometrical and operating factors on the shear stress in the microfluidic cerebrovascular channel. Furthermore, the non-Newtonian behavior of blood flow is considered to evaluate the shear stress in the microfluidic cerebrovascular channel. The Newtonian and six non-Newtonian fluids models of Carreau, Carreau-Yasuda, Casson, Cross, Ostwald-de Waele, and Herschel-Bulkley are numerically tested under various conditions of the flow rate, width, and height of the channel to find the viscosity influence on the shear stress. The Taguchi analysis consisting of range and variance analyses is applied to the L₁₆ orthogonal array to evaluate the effect of various factors on shear stress in terms of influence order, range, F value, and percentage contribution. The parameters for the considered six non-Newtonian fluid models are proposed to accurately map the viscosity behavior with shear strain compared to the actual blood flow behavior. The Newtonian, Carreau, and Carreau-Yasuda non-Newtonian fluid models are found accurately with maximum errors between the experimental and numerical shear stress results as 2.17%, 1.30%, and 1.48%, respectively. The shear stress decreases with an increase in the width and height of the channel and a decrease in the viscosity for all flow rates. The porosity is evaluated as a highly influential factor followed by the flow rate, width, and height of the channel in decreasing order based on their effects on the shear stress. The modified shear stress equation is proposed with an accuracy of 0.96 by integrating the effect of porosity in addition to width, height, flow rate, and viscosity. The in-vitro microfluidic cerebrovascular model could be designed and manufactured based on the proposed results on influence order, F value, and percentage contribution of various factors in direction of achieving the in-vivo level shear stress.