Robust Transverse and Axial Flow Measurement Using Constrained Dynamic Time Warping and Cumulative Decorrelation of Intravascular Ultrasound Signals

The accurate measurement of blood flow velocity is vital for diagnosing and treating various medical conditions. However, precisely measuring transverse and axial blood flow velocities in small, deeply embedded blood vessels remains a significant challenge. This article presents a robust approach that leverages cumulative signal decorrelation for intravascular ultrasound (IVUS) applications. Within the vessel, movement of blood cells induces a phase discrepancy in the signals acquired by microprobes, manifesting as a time lag between corresponding segments of a set of radio frequency (RF) signals. A triangular constrained dynamic time warping (DTW) model is proposed to calculate the time lag, coupled with an empirical formula relating cumulative decorrelation to RF echo signals. This integration enables the estimation of both axial and transverse blood flow velocities, overcoming the limitations of traditional decorrelation methods, such as sensitivity to signal-to-noise ratio (SNR) and cycle hopping, thereby enhancing the accuracy and stability of blood flow velocity measurements. The technique was validated using a 50-MHz IVUS system to measure flow velocities within a microchannel with a 3 mm inner diameter, demonstrating robustness independent of the probe angle. This method effectively measures various blood flow parameters, including transverse and axial velocities and velocity profiles, across a wide range up to 75.5 cm/s. Notably, the spatial resolution achieved for both transverse and axial velocities is up to $10~\mu $ m, providing an opportunity for highly accurate wall shear rate estimation.