Development of a multi-scale model of the human pulmonary circulation

Introduction

These movies illustrate the development of a multi-scale model of the human pulmonary circulation. We have constructed models from the level of the alveolar-capillary network - where blood is modelled as a non-Newtonian fluid - to models of blood flow in the larger scale arterial and venous networks via solution of the Navier-Stokes equations. The following movies illustrate the progression of these models.

Alveolar geometry

This movie demonstrates the 3D Voronoi meshing technique within a unit cube. Certain Voronoi 'cells' are then allocated as duct cells (shown in blue). All adjacent Voronoi cells are then considered to be alveoli (shown in grey). A single alveolar sac, consisting of 19 alveoli (shown in pink), is isolated for analysis and creation of the capillary model.

Capillary geometry

This movie displays the method use to create the capillary geometry. A 2D Voronoi meshing technique is applied on the surface of a unit sphere. The Delaunay triangulation is then formed and the Voronoi mesh is created. This mesh is then projected down onto the alveolar surface of a single alveoli. This entire alveolar sac capillary mesh is also displayed.

Capillary blood flow

This movie initiates at the parenchymal (alveolar model) level and zooms out to display the alveolar sac capillary model. A pulsatile pressure solution within the capillary model is then displayed in the movie. The flow rate solution is displayed (mm³ s^-1) and ranges from 8.9301e-08 (red) to 0 (dark blue). This solution is obtained by imposing a time-dependent pressure boundary condition at the arteriole inlet vessels.

Lobar geometry fitting

This movie demonstrates the fitting of the lobar models to MDCT (Multi-Detector row x-ray Computed Tomography) data. MDCT slices are stacked to produce rendered iso-surfaces of the lung surface. The scan data is of high enough resolution to detect the fissures and enable a description of geometry of each of the five separate lobes. Data points are generated on the iso-surfaces. A geometry fitting procedure is then used to fit initially linear elements to the data points. This movie was created by Dr Merryn Tawhai.

Vascular geometry

This movie demonstrates the volume-filling branching (VFB) algorithm used to generate the additional arterial and venous vessels unidentifiable from the MDCT image data. The algoithm begins from the terminal locations of the vessels identified from the MDCT and creates a volume-filling branching structure into the lobar volumes.

Pulsatile pressure solution

This movie displays the pulsatile pressure solution through the arterial network. A pulsatile pressure boundary condition is applied at the pulmonary trunk inlet and all terminal pressures are constant over time. The solution spectrum ranges from -0.04 to 3.2 kPa.

Pulsatile velocity solution

This movie illustrates the pulsatile velocity solution through the arterial network. The solution spectrum ranges from -50 to 400 mm s^-1.

IUPS Physiome Project