Phase‐contrast velocimetry with hyperpolarized ³He for in vitro and in vivo characterization of airflow

Abstract

This paper describes a technique that combines radial MRI and phase contrast (PC) to map the velocities of hyperpolarized gases ( 3 He) in respiratory airways. The method was evaluated on well known geometries (straight and U-shaped pipes) before it was applied in vivo. Dynamic 2D maps of the three velocity components were obtained from a 10-mm slice with an in-plane spatial resolution of 1.6 mm within 1 s. Integration of the in vitro throughplane velocity over the slice matched the input flow within a relative precision of 6.4%. As expected for the given Reynolds number, a parabolic velocity profile was obtained in the straight pipe. In the U-shaped pipe the three velocity components were measured and compared to a fluid-dynamics simulation so the precision was evaluated as fine as 0.025 m s ؊1. Ventilation and particle-deposition studies are important parts of functional and physiological explorations of the lungs. Normal or altered geometries of the bronchial tree directly affect airflow distributions in the lungs (1). Particle deposition is involved in inhalation toxicology, and its study is motivated to achieve a better understanding of its effects and devise new therapies based on drug inhalation (2,3). Knowledge of flow patterns in large tracheobronchial airways is required to understand flow distribution, gas mixing, and inhaled particle deposition.Several experimental and numerical studies have focused on the velocity patterns in large airways. Experimentally, velocity fields in various airway models were obtained with the available techniques for measuring gas velocity, including hot-wire anemometry, laser Doppler anemometry, and particle-image velocimetry (4 -6). Numerical simulations were also performed with computational fluid dynamics (CFD) (2,6 -8). Both approaches independently yielded basic results that led to a better understanding of gas-flow dynamics in the bronchi. However, in vivo, functional respiratory tests provide only global information on the airflow, and some invasive techniques introduce sensors to probe flow properties locally but only at a few specific points (9). Nevertheless, noninvasive direct measurements of gas velocities have never been obtained in living subjects, since none of the abovecited measurement techniques are able to do so.Recently it was shown that ventilation dynamics in the lungs of small animals and humans can be monitored by hyperpolarized (HP) gas MRI (10) using EPI (11), fast gradient-echo (12), spiral (13,14), or radial sequences (15,16). However, quantifying the gas-flow rate from signal dynamics is not straightforward because it is difficult to separate flip angle and inflow effects (12,(17)(18)(19). Moreover, additional signal losses result from oxygen-dependent longitudinal relaxation times that become relevant over long acquisition times (20), and from short effective transverse relaxation times, while the gas diffuses through the magnetic field gradients within the airways (21,22).In proton MRI, phase contrast (PC) has been widely used to map blood v...

Multiscale Modeling of the Respiratory Tract

“…Note that most of the numerical studies in realistic airways geometries we are aware of (see [9], [11], [12], [14]) use commercial softwares, with Neumann or Dirichlet boundary conditions at the inlet and outlets, to perform their flow simulations. Nervertheless most "general purpose" softwares are not designed to take into account easily natural dissipative boundary conditions, nor the coupling with the spring model.…”

“…We take v = u as a test function in (12). By writing the convective terms, thanks to an integration by parts, as a flux of kinetic energy at the inlet and at the outlets, and remembering that Sẋ = − Γ 0 u · n, we obtain…”

A computational study of the respiratory airflow characteristics in normal and obstructed human airways

“…However, because PFT measures airflow in the bulk phase, it is insensitive to regional differences in airflow patterns, especially in the lower airways. Similarly, although hyperpolarized noble gas magnetic resonance imaging can provide static and dynamic ventilation maps of the lungs [8], it is still not possible to obtain details of airflow characteristics, such as velocity contours, in the lower airways [9].…”

Intrapulmonary 3He Gas Distribution Depending on Bolus Size and Temporal Bolus Placement

“…They look at phase contrast MRI using hyperpolarized 3 He, which was first tested in an airway model and showed the expected parabolic flow profile in a straight tube as well as feasibility of human application. 27 Drawbacks of the present study were that the results only hold true for healthy lung tissue. There might be airway dynamics in pulmonary disease, which alter the influences of bolus placement, especially in regions, which underlie ventilation redistribution.…”