Estimating aortic blood pressure from femoral pressure measurements via an arterial transfer function: A Gaussian decomposition approach.

This paper presents a method of identifying the model parameters of a validated arterial transfer function model, to estimate central blood pressure (P ao,est ) from femoral pressure measurements (P fe,mea ), alone. The proposed method avoids the need for additional patient specific measurements to individualize the model, which can be clinically challenging to measure and impossible to obtain retrospectively. The proposed method method decomposes each P fe,mea wave into its constituent forward and reflected waves by fitting a set of Gaussian waves to the measurement. Reflected waves have been shown to exert a similar influence on central and peripheral pressures and thus can be exploited to identify all model parameters, including central-peripheral pulse transit time (PTT). The P ao,est method was implemented using data from seven (7) porcine endotoxin experiments where septic shock was induced and subsequent treatment was performed. Systolic pressure, pulse pressure and the root-mean-squared errors, relative to direct, invasive measurements of aortic pressure were used to assess method accuracy. Peripheral pressure accuracy is also assessed due to its prevalence as a surrogate for central BP in clinical environments. Finally, two alternate transfer function methods were considered; a population average based generalized model (P GTF,arx ) model, and an optimal method (P ao,rmse ) using the measured aortic pressure to minimize the model RMSE for each section of data.