Validation study of an interpolation method for calculating whole lung volumes and masses from reduced numbers of CT-images in ponies

• Published in: The veterinary journal (1997) Vol. 202; no. 3; pp. 603 - 607
• Main Authors: Reich, H., Moens, Y., Braun, C., Kneissl, S., Noreikat, K., Reske, A.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.12.2014
• Subjects: Anaesthetised ponies; Animals; Horses; Image Processing, Computer-Assisted - methods; Interpolation validation; Lung - diagnostic imaging; Lung - physiology; Lung Volume Measurements - veterinary; Lung volumes and masses; Quantitative imaging; Reproducibility of Results; Thoracic computed tomography; Tidal Volume; Tomography, X-Ray Computed - veterinary; Interpolation validation; Thoracic computed tomography; Quantitative imaging; Lung volumes and masses; Anaesthetised ponies
• ISSN: 1090-0233; 1532-2971
• DOI: 10.1016/j.tvjl.2014.10.019

•Quantitative thoracic CT-analysis is a good tool to evaluate lung tissue aeration.•Manual outlining of regions of interest for volume calculations is time intensive.•An interpolation method using only 10–14 CT-images was evaluated.•Results validate the technique with 12 CT-images as time efficient and reliable. Quantitative computer tomographic analysis (qCTA) is an accurate but time intensive method used to quantify volume, mass and aeration of the lungs. The aim of this study was to validate a time efficient interpolation technique for application of qCTA in ponies. Forty-one thoracic computer tomographic (CT) scans obtained from eight anaesthetised ponies positioned in dorsal recumbency were included. Total lung volume and mass and their distribution into four compartments (non-aerated, poorly aerated, normally aerated and hyperaerated; defined based on the attenuation in Hounsfield Units) were determined for the entire lung from all 5 mm thick CT-images, 59 (55–66) per animal. An interpolation technique validated for use in humans was then applied to calculate qCTA results for lung volumes and masses from only 10, 12, and 14 selected CT-images per scan. The time required for both procedures was recorded. Results were compared statistically using the Bland–Altman approach. The bias ± 2 SD for total lung volume calculated from interpolation of 10, 12, and 14 CT-images was −1.2 ± 5.8%, 0.1 ± 3.5%, and 0.0 ± 2.5%, respectively. The corresponding results for total lung mass were −1.1 ± 5.9%, 0.0 ± 3.5%, and 0.0 ± 3.0%. The average time for analysis of one thoracic CT-scan using the interpolation method was 1.5–2 h compared to 8 h for analysis of all images of one complete thoracic CT-scan. The calculation of pulmonary qCTA data by interpolation from 12 CT-images was applicable for equine lung CT-scans and reduced the time required for analysis by 75%.