Original ContributionAn Automatic Angle Tracking Procedure for Feasible Vector Doppler Blood Velocity Measurements
Introduction
Blood velocity is typically investigated through methods exploiting the well known Doppler effect. According to this effect, a flow with velocity magnitude |v| impinged by a planar ultrasound (US) wave with center frequency fo, generates echoes characterized by a frequency shift:where c is the velocity of US and θ is the beam-to-flow angle. By measuring the frequency fd, the axial component, |v|cosθ, of the velocity magnitude is estimated. However, unless θ is known, it is not possible to estimate the individual contribution of |v| or θ to this component.
The lack of knowledge of the beam-flow angle represents a crucial source of uncertainty in US blood velocity measurements. Because interpretative criteria for carotid stenosis are heavily based on Doppler velocities, errors in Doppler angle correction may lead to serious errors in diagnosis (Logason et al., 2001, Grant et al., 2003). In particular, peak velocity measurements, which are frequently used in diagnosis and grading of internal carotid artery (ICA) stenosis (Hunink et al., 1993, Lewis and Wardlaw, 2002), are seriously affected by angle errors associated to human factors (Lui et al. 2005). As many as 35% of the applications received by the Intersocietal Commission for Accreditation of Vascular Laboratories demonstrate improper angle correction techniques, making this issue one of the most common causes for delayed decisions (DeJong 2000). However, significant errors are made even by experienced vascular technologists in accredited laboratories (Lui et al. 2005).
The Doppler angle ambiguity is still usually faced by manually aligning, in B-mode imaging, the angle cursor with the vessel wall (Evans and McDicken 2000), following an approach that can be accurate only when the flow direction is parallel to the vessel axis. The research in this field is quite intense, and sophisticated methods based on, e.g., speckle tracking (Trahey et al. 1987), transverse oscillation (Jensen and Munk 1998) and directional beamforming (Jensen and Bjerngaard 2003) have been proposed. However, there is still great interest for the classic approach based on the combination of Doppler measurements related to multiple US beams intersecting in the region of interest with known interbeam angle (Dunmire et al., 2000, Steel et al., 2004). The frequencies generated from the Doppler sample volume (DSV) intercepted by two (or more) transducers or by multiple apertures realized on a single linear array transducer are first estimated. Through a trigonometric combination of the Doppler equations related to each beam, both the velocity magnitude and the flow direction are derived. However, the classic fluctuation in the Doppler frequency estimate is subject to magnification of bias, especially at small interbeam angles (Steel and Fish 2002).
An original dual-beam vector Doppler method, which does not need any trigonometric combination of independent frequency measurements, has been recently introduced (Tortoli et al. 2006). Here, the flow direction is identified by transversely orienting a reference beam. This condition is recognized by exploiting a unique signature of “transverse” Doppler spectra, i.e., the symmetry around zero frequency (Newhouse et al., 1987, Tortoli et al., 1993). Once the reference beam has been finely oriented, a second (measuring) beam can be used to directly estimate the true flow velocity at known beam-flow angle.
Although the technique has been thoroughly validated in vitro and in vivo (Ricci et al. 2009), for its clinical application it is necessary that the desired transverse angle be automatically obtained. In this paper, we present an automatic angle-tracking method and its implementation in the ULtrasound Advanced Open Platform (ULA-OP), an experimental imaging/Doppler system connected to a linear array probe (Tortoli et al. 2009). This system was programmed to independently control two US beams and perform in real-time all tasks involved by the proposed procedure. The operator should only take care of locating the DSV in a suitable position so that, independently of the initial probe orientation, the reference and measuring beams were automatically steered to intercept the DSV along directions optimal to identify the flow direction and the flow velocity magnitude, respectively. Results of in vitro experiments conducted in both steady and pulsatile flow conditions and of in vivo reproducibility tests made on the common carotid arteries of 13 volunteers are reported. Major benefits and limitations of the proposed procedure in view of its possible clinical application are considered in the Discussion.
Section snippets
Automatic angle-tracking dual-beam method
In the recently proposed dual-beam method (Tortoli et al. 2006), two transducers, hereinafter referred to as “reference” and “measuring” transducers, respectively, inspect the same DSV with controllable steering angles, α1 and α2, and interbeam angle, δ (Fig. 1). If the Doppler spectrum associated to the reference transducer is continuously calculated, it is possible to monitor its mean frequency, fdr, and the Spectral Symmetry Index (SSI), defined as:in which P+ and P-
Experimental set-up
All experiments reported here have been obtained using ULA-OP together with the 192-element linear array probe LA523 (Esaote spa, Firenze, Italy), which is characterized by a pitch of 0.245 mm. In PW Doppler mode, groups of 64 elements were excited with Hamming-weighted bursts of 5 cycles at 6.25 MHz, apodized by a Hanning window. By controlling the system bandwidth, the length of the TX burst and of the RX gate, the DSV length was fixed at about 2 mm for the reference beam and 1 mm for the
Discussion
The automatic tracking angle procedure proposed in this paper is based on unique symmetry features shown by the Doppler spectrum when the interrogating US beam is perpendicular to the flow. Hence, it was significant to quantitatively evaluate at which extent the spectral symmetry exhibited around zero frequency is correlated to the closeness of the Doppler angle to 90°. Previous in vitro experiments (Ricci et al. 2009) already showed that angle errors as low as 2° were sufficient to yield SSIs
Conclusion
The results of the in vitro and in vivo test demonstrate that the tracking method is capable of automatically finding the correct reference line orientation without degrading the repeatability performance of the dual-beam technique detailed in Ricci et al. (2009). The procedure discussed here does not aim at solving the inherent limitations of the technique, which is intrinsically 2-D and may be not applicable in vessels that do not allow achieving the needed transverse beam orientation because
Acknowledgments
This work was partially funded by Fondazione Cure, Milan, Italy. The authors wish to thank all the staff at the Microelectronic Systems Design Laboratory (Florence, Italy), and in particular Dr. Luca Bassi, for great technical contribution to this work. Special thanks are also due to Daniele Righi, M.D., for valuable discussions about the possible clinical impact of the proposed procedure.
References (20)
- et al.
Cross-beam vector Doppler ultrasound for angle-independent velocity measurements
Ultrasound Med Biol
(2000) - et al.
Which Doppler velocity is best for assessing suitability for carotid endarterectomy?
Eur J Ultrasound
(2002) - et al.
The importance of Doppler angle of insonation on differentiation between 50–69% and 70–99% carotid artery stenosis
Eur J Vasc Endovasc Surg
(2001) - et al.
Validation of a new blood-mimicking fluid for use in Doppler flow test objects
Ultrasound Med Biol
(1998) - et al.
Accuracy and reproducibility of a novel dual-beam vector Doppler method
Ultrasound Med Biol
(2009) - et al.
Angle-dependence and reproducibility of dual-beam vector Doppler ultrasound in the common carotid arteries of normal volunteers
Ultrasound Med Biol
(2004) - et al.
Transverse Doppler spectral analysis for a correct interpretation of flow sonograms
Ultrasound Med Biol
(1993) - DeJong MR. Developing Angle Correction Methods for the Laboratory. ICAVL newsletter 2000, Spring Issue. Available at:...
- et al.
Doppler ultrasound: Physics, instrumentation and signal processing
(2000) - et al.
Carotid Artery Stenosis: Gray-Scale and Doppler US Diagnosis—Society of Radiologists in Ultrasound Consensus Conference
Radiology
(2003)
Cited by (49)
Respiratory variability of peak velocities in the common femoral vein estimated with vector flow imaging and Doppler ultrasound
2018, Ultrasound in Medicine and BiologyPitfalls of Doppler Measurements for Arterial Blood Flow Quantification in Small Animal Research: A Study Based on Virtual Ultrasound Imaging
2016, Ultrasound in Medicine and BiologyCitation Excerpt :In these cases, the maximum frequency envelope might seemingly result in an improved velocity measurements because the complex flow field is not entirely captured by this 1-D Doppler technique. Several strategies have been proposed to overcome these angle correction issues, ranging from the implementation of multidimensional velocity estimation (vector Doppler [Fox 1978], transverse oscillation [Jensen and Munk 1998], speckle tracking [Trahey et al. 1987]) to improvements in 1-D Doppler signal processing such as the automated angle tracking procedure (Tortoli et al. 2010), but to the best of our knowledge these strategies have not yet been explored in small animal imaging. During in vivo scanning in mice, the imaging depth of the target artery can be considered quite flexible, with the ratio of the gel layer applied on the probe to the actual imaging depth higher in mice than humans.
Flow velocity mapping using contrast enhanced high-frame-rate plane wave ultrasound and image tracking: Methods and initial in vitro and in vivo evaluation
2015, Ultrasound in Medicine and BiologyComparison of carotid artery blood velocity measurements by vector and standard doppler approaches
2015, Ultrasound in Medicine and BiologyCitation Excerpt :Some of these methods have also been tested in vivo. Among the dual-beam approaches, the one capable of tracking the Doppler angle by using the transverse Doppler principle (Tortoli et al. 1993) has recently been implemented in the ULtrasound Advanced Open Platform (ULA-OP) (Tortoli et al. 2010) to allow repeatable PSV measurements in the common carotid arteries (CCAs) of 13 healthy patients. In Pedersen et al. (2012), a modified commercial scanner integrating the transverse oscillation vector technique was used to estimate multiple parameters (including PSV and end-diastolic velocity) in the CCAs of 16 healthy volunteers.
Blood flow image by multi-angle composite ultrasonic Doppler vector
2022, Wuli Xuebao/Acta Physica Sinica