Indirect measurement of average alveolar dimension using dynamic phase-contrast imaging

For some lung diseases, assessment of alveolar dimension could add critical information to inform patient care and disease progression. However, current clinical imaging techniques, such as computed tomography, lack the resolution required to measure these small structures in patients. While the gold standard imaging modality for measuring alveoli is micro-CT, this technique is not possible in clinical use due to the size of the patients and the radiation dose. An alternative imaging modality is phase-based contrast imaging, which would deliver a lower dose to patients and increase the size limit. Phase contrast X-ray imaging has previously been combined with particle image velocimetry (PIV) to measure lung motion, another indicator of lung disease. Thus it was hypothesized that average alveolar size could also be measured indirectly using PIV. In the work reported here, we show that average alveolar size shows a high correlation to the mathematical divergence of the velocity vector field that results from the speckle pattern produced by phase imaging of mouse lungs. This correlation is linear with p<0.006. If this correlation holds in human lungs, it could potentially be calibrated to indirectly measure average alveolar size in human patients using some of the grating-based phase-contrast imaging methods that are showing great promise in clinical use.