Performance of the Transverse Oscillation method using beamformed data from a commercial scanner

Abstract

Abstract-Blood velocity estimates using conventional color flow imaging (CFI) or Doppler techniques are angle dependent. One of the proposed techniques to overcome this limitation is the Transverse Oscillation (TO) method, which also estimates the lateral velocity components. The performance of this is evaluated on a commercial platform. Beamformed data are acquired using a commercial BK Medical scanner as opposed to the previously reported results obtained with the experimental scanner RASMUS. The implementation is evaluated using an in-house circulating flow rig by calculating the relative mean standard deviation and bias of the velocity components. The relative mean standard deviation decreases as the number of shots per estimate increases and a value of 5% is obtained for 64 shots per estimate. For a center frequency of 5 MHz at 60• , 75• , and 90• , the relative mean bias varies from 21% to 27% and is lowest at a transmit focal depth close to the center of the vessel. The present performance is comparable with the results from the experimental scanner and simulations. It is obtained with only few changes to the conventional CFI setup and further optimization can improve the performance. This illustrates the feasibility of implementing the TO method on a commercial platform for real-time estimation.

Vector Flow Imaging Compared with Conventional Doppler Ultrasound and Thermodilution for Estimation of Blood Flow in the Ascending Aorta

“…According to simulations, the TO method has an almost constant relative bias with insonation angles between 60° and 90°, whereas the relative SD is reduced from 7% to 5% when increasing the insonation angle from 60° to 90°. 25 As the average beam-to-flow angle in vivo was 76.3° compared with 90° in the flowrig, a slightly higher variation with an almost identical bias of the vector velocities was expected in vivo when compared with the flowrig measurements.…”

“…TO has previously been validated with MRI, ultrasound dilution technique (UDT), and spectral Doppler. [17][18][19]24 The TO method implemented on the commercial scanner has also been validated on a flowrig with constant flow, 25 and studies have investigated the feasibility of TO estimation of the heart intraoperatively on a few patients. 22,23 In this study, TO was evaluated in a flowrig with both constant and pulsatile flow to investigate the systematic bias of TO in a controlled setup, and to distinguish different factors contributing to the bias.…”

“…According to previous flowrig TO experiments, the beam-to-flow angle is of less importance when using 5 MHz center frequency TO emission, thus, only 90° insonation angle was investigated. 25…”

“…Further details on the volume flow calculation from vector estimates are given by Hansen et al 24 Flow Phantom Setup MHz center frequency TO emission, thus, only 90° insonation angle was investigated. 25 TO was evaluated for increasing flow rates with both constant and pulsatile flow. Vector velocities at each flow rate setting were recorded twice to assess the precision of TO.…”

Transverse oscillations for phased array vector velocity imaging

“…The Transverse Oscillation (TO) method by Jensen and Munk [2] and the similar approach by Andersen [3] are two of them. The TO method has shown intriguing in vivo results of blood vector velocities in superficial vessels using linear arrays [4], and is a technique close to a commercial breakthrough [5]. However, the current implementation limits the scan depth, and hence, prevents imaging of deeper lying vessels and organs, e.g.…”