| Pulmonary emphysema is a chronic obstructive lung disease that affects over 60 million people worldwide. It is characterized by the progressive destruction of lung tissue and the subsequent decline in lung function. However, the link between microscopic structural changes in the lung due to emphysema and the loss of function is not very well understood. In this thesis, we hypothesized that in addition to the amount of tissue lost, the spatial pattern of destruction plays a major role in determining the rate of decline of lung function and that by characterizing the changes in the microscopic structure, it may be possible to predict changes in function. To test these hypotheses, first, we developed a new technique to stain lung tissue with silver and image the microscopic structure of lung tissue in three dimensions. We used this technique to quantify changes in the alveolar volume distribution in an elastase-induced mouse model of emphysema one week and two weeks following the initial treatment. We found that early structural changes in emphysema were marked by a faster increase in variability of airspace sizes compared to its mean. To examine what kind of processes may have led to the experimentally observed structural changes, we designed and developed a 3D computational model of lung tissue where we mimicked the process of tissue destruction using different mechanisms. We found that different destruction mechanisms are characterized by different spatial patterns which in turn have significant effect on rate at which the stiffness of the network declines. Hence, by identifying correlations in the destruction process, we can discern the mechanisms that lead to the progressive destruction of alveolar walls in emphysema. Further, using the method of principal component analysis, we were able to develop a general framework to relate microscopic structural changes as characterized by the cell volume distribution to functional changes. We showed that for spatially correlated destruction processes it is possible to predict changes in stiffness of the network from measurements of structural changes. Our approach has unique potential for early detection of the disease as well as in testing the efficacy of potential therapeutic interventions. |
