Dennis Onyeka Frank, North Carolina State University


Parameter Estimation and Sensitivity Analysis of a Reduced Endotoxin Challenge Model

Abstract: The initial reaction of the body to bacterial infection or severe tissue trauma is an acute inflammatory response, such response helps to annihilate threat posed by gram-negative bacteria (endotoxin) and thus restore health. Meanwhile, an uncontrolled inflammatory response can cause tissue damage, and ultimately death. In a previous work by Roy et al, an 8-state ordinary differential equation (ODE) model of the acute inflammatory response system to endotoxin challenge was developed. Endotoxin challenges at 3, 6 and 12mg/kg were administered to rats, and experimental data for pro- inflammatory cytokines such as interleukin-6 (IL - 6) and tumor necrosis factor (TNF), as well as anti-inflammatory cytokine such as interleukin-10 (IL-10) were obtained. Endotoxin challenges at 3 and 12mg/kg were used to calibrate the model, and model validation was performed by comparing the model predictions at an endotoxin challenge level of 6 mg/kg with experimental data from rats at the same level. In this work, we developed a reduced ordinary differential equation (ODE) model of the acute inflammatory response system to endotoxin challenge by categorizing state variables with similar behavior or functions into the same group. Furthermore, we proceeded with parameter estimation by using model sensitivity analysis at endotoxin challenges of 3 and 12mg/kg respectively to identify those parameters that were sensitive. We then introduced the subset selection method of ``SVD followed by QR factorization with column pivoting" for parameter identifiability. Finally, model comparison and validation were done by comparing curve fittings of the original ODE model and, the reduced ODE model against experimental data, as well as using Akaike's Information Criterion (AIC).

Advisor: Hien Tran (North Carolina State University)