Subject-specific estimation of aortic blood pressure via system identification: Preliminary in-human experimental study

This paper demonstrates in-human validity of a novel subject-specific approach to estimation of central aortic blood pressure from peripheral circulatory signals. In this “Individualized Transfer Function” (ITF) approach, the unknown circulatory dynamics of cardiovascular system are determined via system identification by characterizing its parallel tube-load model representation based on circulatory signals measured at upper and lower extremity locations. Then a stable input de-convolution algorithm is used to estimate central aortic blood pressure as unknown common input signal to the parallel tube-load model of cardiovascular system. Using experimental data collected from five patients undergoing cardiac surgery with cardiopulmonary bypass, the validity of the ITF approach was established by demonstrating that 1) the tube-load model can reproduce cardiovascular hemodynamics with fidelity, and 2) the ITF approach can estimate aortic blood pressure very accurately. In comparison with direct radial and femoral BP measurements, the ITF approach resulted in significant reductions in errors associated with estimation of central aortic blood pressure, including 27-42% reduction in root-mean-squared blood pressure waveform errors as well as 64-70% and 75-81% reductions in systolic and pulse blood pressure errors.