Abstract
Four-dimensional flow magnetic resonance imaging-based pulse wave velocity (4D flow PWV) estimation is a promising tool for measuring regional aortic stiffness for non-invasive cardiovascular disease screening. However, the effect of variations in the shape of flow waveforms on 4D flow PWV measurements remains unclear. In this study, 4D flow PWV values were compared using cross-correlation algorithm with different interpolation times (iTs) based on flow rate and beat frequency. A critical iT (iT(Crit)) was proposed from in vitro study using flexible and stiff phantom models to simultaneously achieve a low difference and a low computation time. In vivo 4D flow PWV values from six healthy volunteers were also compared between iT(Crit) and the conventionally used interpolation time of 1 ms (iT(1 ms)). The results indicated that iT(Crit) reduced the mean difference of in vitro 4D flow PWV values by 19%, compared to iT(1 ms). In addition, iT(Crit) measured in vivo 4D flow PWV, showing differences similar to those obtained with iT(1 ms). A difference estimation model was proposed to retrospectively estimate potential differences of 4D flow PWV using known values of PWV and the used iT. This study would be helpful for understanding the differences of PWV generated by physiological changes and time step of obtained flow waveforms.