Author: Hilary Louisa Byrne, Paul J. Keall, John Kipritidis, Jeremy Lim 👨🔬
Affiliation: Northern Sydney Cancer Centre, University of Sydney, Image X Institute, Faculty of Medicine and Health, The University of Sydney 🌍
Purpose:
Non-contrast CT ventilation imaging (CTVI) has been developed as a cost-effective and accessible alternative to PET/SPECT V/Q imaging for visualizing lung function. However, the sensitivity of CTVI to variations in image acquisition and reconstruction parameters has not been extensively studied. Here, we investigate the impact of reconstruction kernel, iterative reconstruction strength, and imaging dose on CTVI using a custom-designed phantom.
Methods:
A pair of patient-derived inhale and exhale lung phantoms was fabricated using a novel 3D printing technique (PixelPrint, Philadelphia) that produces realistic tissue structures with high spatial and Hounsfield Unit (HU) accuracy. The phantoms were scanned on a Siemens Biograph Vision Quadra CT scanner using different combinations of reconstruction kernels suited for lung imaging: QR40, BL57, and BR60; iterative reconstruction (ADMIRE) strengths: 3 and 5; and imaging doses: standard dose (CTDIvol = 6.3 mGy) and low dose (CTDIvol = 1.8 mGy). A baseline scan (QR40, ADMIRE 3, standard dose) was selected for comparison against other scan settings. CT ventilation images were created using the Jacobian (CTVIJac) and HU methods (CTVIHU). CT image similarity was assessed using root mean square error (RMSE) while ventilation image similarity was assessed using the Pearson correlation coefficient.
Results:
The maximum RMSE of values within the lung was 46 HU while both CTVIJac and CTVIHU exhibited consistently high correlations ranging from 0.99 to 1.00 and 0.98 to 1.00, respectively, across all combinations of reconstruction kernels, iterative reconstruction strengths, and imaging dose.
Conclusion:
CTVI is robust to variations in reconstruction kernel, iterative reconstruction strength, and imaging dose. These findings suggest that CTVI is a reliable method for assessing lung function across diverse image acquisition and reconstruction settings, supporting its clinical applicability and reproducibility. The robustness of low dose CTVI highlights its potential for reducing radiation exposure without compromising the reliability of lung function assessment.