We create personalised patient-based and species-specific geometric models. Our models are derived using medical imaging data (computer tomography or magnetic resonance imaging) and computational algorithms and are the most realistic lung models of their kind. We create geometric models of the lungs, lobes, airways, and the pulmonary circulation (arteries and veins).

We have quantified the mean lung shape and its principal modes of shape variation in healthy subjects across a seven-decade age span. We have also established correlation between lung shape with age, sex, and body mass index (BMI). In effect, we have developed a unique statistical shape model - the first ever quantification of lung shape that considers age, and is the only study to demonstrate clear relationships between lung shape and measurements of lung function. Age-related shape differences in our model were indicative of chest wall remodelling and age-related decline in lung tissue elasticity.

Distribution of lung tissue within the chest cavity plays a significant role in the delivery of both blood and air to the gas exchange regions of the lung. Various factors such as shape change (chest movement in one patient versus no chest movement in another), parenchymal density (eroded tissue versus healthy tissue), gravitational direction (patient breathing when lying on the back versus lying down on ones’ tummy) and level of inflation (shallow breathing versus deep inspiration) influence the distribution of lung tissue. Using our computational model, we found that the pleural cavity shape change affects regional lung tissue distribution, and further impacts regional elastic recoil pressure.

We have developed a computational model to represent the distribution of ventilation within the lungs. This model links the resistance to airflow within the conducting airways to the deformation of the alveolar/acinar lung tissue, creating what we refer to as the ‘balloon on sticks’ model. This model links the mechanical deformation of lung tissue to airflow within the airways and is the most advanced and realistic ventilation model of its kind.

We solve equations representing fluid flow to solve for the pressure and flow distribution through the full pulmonary circulation – including arteries, veins, and capillaries. These models have and are being applied to understand pulmonary embolism and pulmonary hypertension.

After simulating ventilation (V) and blood flow (Q) in our anatomically-based models, we use this V/Q ratio to determine the amount of gas exchange (oxygen from air to blood and carbon dioxide from blood to air). To do this, we solve equations that represent the exchange of these gases across the blood-gas diffusion barrier.

The airways are lined with a thin layer of liquid, the presence of this liquid ensures the healthy function of the cells and mucociliary transport function. We have developed a predictive model of dynamic heat and moisture transport. This has been applied to studying the ideal conditions for air (temperature and humidity level) being delivered to ventilated patients.

Clinical measurements of lung function provide averaged whole lung measures of lung function. When disease occurs and these clinical measures change, it is difficult to know where disease changes are occurring. We have models that simulated forced expiration and multiple breath washout and are using these to improve our understanding of clinical measurements of lung function.