Translational Research: Multi-Scale Models of the Pulmonary Circulation in Health and Disease - Multiscale Computer Modeling in Biomechanics and Biomedical Engineering - page 256

delicate parenchymal tissue [ 1 ]. The in vivo lung is maintained in contact with the
wall of the thoracic cavity by negative pressure in the intrapleural 'space' [ 2 ]. Ex
vivo experimentation with lung tissue is difficult to reconcile with in vivo behavior
as when removed from the body the lung recoils to considerably smaller volumes
than are encountered in breathing, and the lung changes in shape. The advent of
modern imaging has provided opportunity to study the lung in situ, however this is
again complicated by lung deformability and its air content. In awake humans the
lung is typically functioning in an upright posture, and because the delicate tissue
is readily deformable there is a substantial gradient of tissue density along the
cranial-caudal axis [ 3 ]. Magnetic resonance imaging (MRI) and computed
tomography (CT) are currently constrained to horizontal postures, i.e. supine or
prone. The lungs are typically at a smaller volume in these postures than when
upright [ 4 ], and the density gradient is in a different axis. This impacts on regional
lung expansion [ 5 ], pulmonary blood volume and flow distribution [ 6 ], the rela-
tionship between ventilation ( V) and perfusion ( Q)[ 6 ], and so gas exchange
function [ 7 ]. The supine or prone human lung could therefore have important
differences in function to upright. Typically, MRI is preferable as an imaging
modality in comparison to CT because the latter involves ionizing radiation.
However, because MRI images protons in water and the lung tissue contains
mostly air, its effective use in lung is restricted to tagging of blood for quantifying
regional blood flow [ 8 , 9 ], or imaging inhaled hyper-polarized gases [ 10 ]. That is,
MRI is not yet effective at imaging lung tissue. High resolution CT provides
excellent anatomical definition of the lung but its use is restricted in vulnerable
populations, and it is not appropriate for most longitudinal studies. High resolution
imaging data for the normal human lung is therefore typically restricted to one or
two volumes, or one or both horizontal postures at a single volume. The complex
interaction between airspaces, blood vessels and tissue in the lung and its move-
ment in breathing further complicates diagnostic imaging and reconciliation of
experimental data at different spatial and temporal scales.
These limitations on imaging and direct experimental measurement of lung
structure and function provide strong motivation for developing mathematical
models of the lung, including the pulmonary circulation, to interpret experimental
results between postures, lung volumes, or different species, or at different spatial
scales. A multi-scale approach to modeling the pulmonary circulation provides a
means to bridge the gap between mechanisms at the genetic, cellular, tissue, vessel
and organ level and to provide new insights into pulmonary function in health and
disease.
Conceptual models of the pulmonary circulation have provided the basis for
understanding its function since at least the 1960s. An example of an early con-
ceptual model of the pulmonary circulation is the 'zonal model' of pulmonary
blood flow [ 11 ]. This model explained how the interaction between pulmonary
arterial, venous and alveolar pressure act to introduce a gravitationally dependent
distribution of blood flow in the lungs. Although there is currently extensive
debate on the magnitude of the zonal effect in comparison to other gravitational