Comparison of Linear Methods for the Determination of Propagation Coefficient in Arterials Systems: Numerical Investigation

Abstract

To get a deeper insight into the factors affecting the vascular hemodynamic, the propagation coefficient (γ) should be calculated. However, results from estimated method of this quantity are conflicting. Using numerical tools, three methods permitting an estimation of this complex number were investigated. We studied the influence of peripheral resistance, fluid and wall viscosities, stiffness, cross-sectional area, vessel length and measurements errors on the accuracy of these methods. Results obtained from this analysis demonstrated that the three methods provide exact value of propagation coefficient when analysing accurate flow velocity and diameter data. Conversely, in realistic condition (i.e. inaccurate data) significant errors related to the degrees of inaccuracy within the data arise. These errors are systematically larger on the values of propagation coefficient obtained by the three-point method than the error on the values calculated by two-point methods. Hence, we believe that difference in sensitivity of each method to the measurements errors may be the main source of the disparity of results reported in literature. In addition, the small value of attenuation and time lag seem to be the sources of the large difference between calculated and theoretical values of this parameter. Our finding doesn't support previous works which attributed the disparity of results to the high reflection or/and to the failure of the methods themselves. In realistic condition, it seems that two-point methods are more reliable than the three-point method for the estimation of the true propagation coefficient.