Virtual Spirometry: Utilizing an abdominal bellows system for quantitatively gating in 4D CT acquisition

Abstract

Purpose: To determine if an abdominal bellows system can be utilized in place of a spirometer for quantitatively gating in 4D CT. While a standard metric in quantitatively gated 4D CT, spirometry presents significant challenges (signal drift, air leakage). When using non-spirometry breathing surrogates (e.g., bellows systems), questions of how to convert the breathing signal into a physiologically quantifiable signal have to be discussed. Methods And Materials: Based on an established 4D CT protocol, 41 patient data sets are acquired while measuring the breathing cycle using both spirometry and an abdominal pressure bellows system. The bellows internal pressure is normalized into a tidal volume signal by correlating it against drift and phase shift corrected spirometry. Assumptions such as stable phase relation and proportionality of the signals are examined. The air volume of the lung (internal air content) is determined by segmentation of the image data and compared to the bellows-based tidal volume to validate the normalized bellows signal. Results: For 14/41 data sets the proportionality factor of spirometry and bellows signal varied considerably during data acquisition, caused by spurious spirometry signals. Introducing a 2% correlation threshold (i.e., for time samples with correlation values < 0.98 for spirometry and bellows signal the spirometry data are removed from further analysis), bellows and spirometry signals correlated well (stable phase relation and proportionality). Moreover, the measured ratio of internal air content and bellows-based tidal volume was 1.11±0.08, matching very well the theoretical value of 1.11 deduced from air density differences in room air and air inside the lungs. Conclusion: Our data support the use of non-spirometry breathing surrogates like bellows systems for quantitative 4D CT - as long as the entire lungs are scanned to provide the internal air content which could be used to convert the non-spirometry signal into a physiologically quantifiable signal.