Reconstructing Haemodynamics Quantities of Interest from Doppler Ultrasound Imaging

Felipe Galarce,Damiano Lombardi,Olga Mula
DOI: https://doi.org/10.1002/cnm.3416
2021-01-20
Abstract:The present contribution deals with the estimation of haemodynamics Quantities of Interest by exploiting Ultrasound Doppler measurements. A fast method is proposed, based on the PBDW method. Several methodological contributions are described: a sub-manifold partitioning is introduced to improve the reduced-order approximation, two different ways to estimate the pressure drop are compared, and an error estimation is derived. A test-case on a realistic common carotid geometry is presented, showing that the proposed approach is promising in view of realistic applications.
Numerical Analysis
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to reconstruct the Quantities of Interest (QoI) in hemodynamics from ultrasonic Doppler imaging. Specifically, the research aims to use Doppler ultrasonic measurement data to estimate the following important hemodynamic parameters: 1. **Complete 3D velocity field and its related quantities**: for example, the maximum velocity. 2. **Pressure and pressure drop**: these are key indicators for evaluating the severity of arterial stenosis. Direct measurement of pressure or pressure drop usually requires invasive means (such as inserting a catheter), so non - invasive ultrasonic measurement methods are of great value. 3. **Vorticity**: especially, it is very important to monitor vorticity around the heart chambers and heart valves, because excessive vorticity may lead to hemolysis. 4. **Wall shear stress**: this is related to the mechanical stress of blood on endothelial cells and is crucial in the study of aneurysms and plaque formation. To achieve this goal, the authors propose a fast algorithm based on the Parameterized Background Data - Weak (PBDW) method. This method combines physical models (such as the incompressible Navier - Stokes equation) and experimental data (such as ultrasonic images), improves computational efficiency by reducing the dimension of the model, and can provide reliable QoI estimates in near - real - time. In addition, the article also introduces several improvement techniques, including sub - manifold partitioning to improve reduced - order approximation, two different pressure drop estimation methods, and the derivation of error estimation. In summary, the core problem of this paper is to develop an efficient and accurate method for extracting valuable hemodynamic information from non - invasive Doppler ultrasonic images.