A 4D CT lung ventilation scan displaying coronal slices of the lungs. Courtesy of Tokihiro Yamamoto, Stanford University.
Lung ventilation is defined as "the rate at which a gas leaves or enters the lungs" and is proportional to the relative change in air volume over time within a given region of the lung. The new modality provides a measure of the latter by quantifying the extent of lung deformation between respiratory phases in 4D CT scans.
Specifically, deformable registration is used to register image frames from different phases of the respiratory cycle to one another. Once registered, the deformation between phases is quantified in the form of a deformation vector field. With further quantitative analysis, a ventilation index is calculated and displayed as a colour scale plot. As a post-processing technique, the modality provides 'free' information on the respiratory function of individuals who have already received a 4D CT scan.
Early days for CT ventilation imaging
The technique is taking its first steps towards clinical use. "Little [clinical] validation has been performed to date and that's the first and necessary step in seeing if this is going to be a useful modality," said Keall. A study led by Keall while at Stanford University (Stanford, CA) provided promising results. CT ventilation imaging demonstrated significant differences between the ventilation of emphysematous and non-emphysematous lung.
Another collaborative ongoing study at Stanford is testing the primary hypothesis that CT ventilation imaging correlates with SPECT ventilation imaging. Promisingly, early results appear to show a correlation between the two modalities. However, as yet unexplained instances of ventilation mismatches between the modalities have also been observed, suggesting a greater understanding is needed before CT ventilation imaging can be introduced to the clinic.
Keall identified the accuracy of deformable registration algorithms and the choice of a representative ventilation metric as two additional areas requiring particular attention. Illustrative examples demonstrated how different registration algorithms and ventilation indices can produce quite different ventilation images.
"It's early days for this modality, but there are several groups aiming to get high-quality data and test out this new functional imaging modality. Time will tell if it is an important tool," said Keall.