Program Information
Strong Evidence for Physiologic Correlation of 4D-CT Ventilation Imaging with Respiratory-Correlated Gallium 68 PET/CT in Humans
J Kipritidis1*, S Siva2, J Callahan2, M Hofman2, P Keall1, (1) Radiation Physics Laboratory, Sydney Medical School, The University of Sydney, Sydney NSW 2006, Australia (2) Division of Radiation Oncology and Cancer Imaging, Peter MacCallum Cancer Centre, Melbourne University, Melbourne VIC 3002, Australia
TU-A-WAB-8 Tuesday 8:00AM - 9:55AM Room: Wabash BallroomPurpose: Four-dimensional CT (4D-CT) ventilation imaging has been developed as a complement to nuclear medicine (NM) based methods, but has not yet been validated against a quantitative NM based standard. We are performing the first such study with positron emission tomography using ⁶⁸Ga-labelled nanoparticles (Galli-PET).
Methods: Respiratory correlated Galli-PET ventilation images are obtained for 10 lung cancer patients using a combined PET/CT scanner with Real-time Position Management system. We begin by calculating the exhale-to-inhale 4D-CT ventilation image (V) by deformably registering the 4D-CT maximal inhale phase to the maximal exhale phase, and performing quantitative analysis of the deformation vector fields (DVFs). We then simulate a motion-blurred or time-integrated 4D-CT ventilation image. This facilitates comparison with time-integrated Galli-PET, which exhibits better signal than for gated images. We study the correlation of V with gated Galli-PET at exhale, and between time-integrated 4D-CT and Galli-PET ventilation using four distinct ventilation metrics. Ventilation images were compared using the voxel-wise spearman correlation (r) within the segmented lung volumes, and dice similarity (d) for the lowest functioning (25th percentile) regions of interest.
Results: Averaged over the ten patients and four ventilation metrics in this study, the comparison of V with gated Galli-PET yielded r = 0.38 and d = 0.40. Time-integrated images exhibited better agreement for all ventilation metrics, yielding r = 0.53 and d = 0.60 on average.
Conclusion: Our results demonstrate moderate correlation between 4D-CT and NM-based ventilation images. The comparison between time-integrated images gives better agreement than for gated images as a result of the improved Galli-PET signal. Further validation of 4D-CT ventilation imaging could facilitate improved access to functional lung imaging and functional lung avoidance in radiotherapy.
Funding Support, Disclosures, and Conflict of Interest: This work is supported by an NHMRC Australia Fellowship, NHMRC project grant 1034060 and US NCI P01CA116602.
Contact Email: