Equating emphysema scores and segmentations across CT reconstructions: A comparison study

摘要

The extent of pulmonary emphysema is commonly quantified on computed tomography (CT) as the proportion of voxels within the lung region below a fixed attenuation threshold. However, this measure is known to be sensitive to several factors, including changes in imaging protocol. The sensitivity restricts the comparison of emphysema scores between studies, between clinical scans without a standard imaging protocol, and in longitudinal data sets with evolving imaging protocols. Recently proposed image processing methods equate emphysema scores across imaging protocols. This study evaluates the performance of two state-of-the-art emphysema segmentation methods and an approach to correct emphysema scores. Emphysema scores and masks generated on sharp reconstructions are compared to ones generated by standard thresholding on smooth reconstructions, for 22 pairs of CT reconstructions.