Applied mathematical modelling of pulmonary function tests

Thesis

The lungs are an incredibly complex pair of organs, whose function is strongly driven by both shape and structure. However, in standard clinical practice lung function is assessed through analysis of 'at-the-mouth' measurements, which by their nature represent a smoothing of behaviours throughout the lungs. Due to this there is a lack of precision in the understanding of how different pulmonary function test outputs respond to different types of lung disease. Within this thesis we apply mathematical and computational modelling to analyse and improve understanding of these responses, and to generate hypotheses that are suitable for further clinical investigation. In the early chapters physics-based models of ventilation, gas transport and frequency propagation within the lungs are developed, and then implemented in efficient computing environments. These models are then applied to the exploratory analysis of indices derived from three pulmonary function tests: the multiple-breath washout; the forced oscillation technique; and magnetic resonance imaging, under a variety of bronchoconstrictive patterns. Following preliminary analysis, we provide a detailed validation of the washout and frequency propagation models, and then show how these models can help create clinical insight as to how different test indices can be most effectively interpreted. In the later chapters we extend upon earlier models, to incorporate flow and gas transfer processes throughout the pulmonary bloodstream. By embedding these models in high-performance computing environments, we create a tractable way to simulate function of the entire pulmonary system. Following implementation of the pulmonary model, we apply it to analyse how gas choice may affect the multiple-breath washout. Finally, we close the thesis with a discussion of the role of mathematical modelling in clinical medicine, of the clinical relevance of results within this thesis, and of the potential for future work.

Authors: Foy, B

• Alternative title: Computational models of airflow, blood flow and gas transport, applied to simulation of clinical pulmonary tests
• DOI: 10.5287/ora-mnw57kago
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford