Extrapolation in the analysis of lung aeration by computed tomography: a validation study

Area of Science:

• Veterinary Radiology
• Pulmonary Imaging
• Comparative Anatomy

Background:

• Computed tomography (CT) is the gold standard for lung aeration and mass quantification.
• Current quantitative CT methods involve significant radiation exposure and manual processing.
• An extrapolation method using 10 CT slices was previously developed to reduce these burdens.

Purpose of the Study:

• To validate a 10-slice CT extrapolation method for lung quantification in pigs and sheep.
• To assess the method's robustness across different thoracic anatomies.
• To determine the minimum number of slices required for accurate extrapolation.

Main Methods:

• Quantitative CT analysis of 168 ovine and 55 porcine whole lungs.
• Comparison of extrapolated lung volume and mass with whole-lung analysis results.
• Bland-Altman analysis to assess bias and limits of agreement (LOA) for varying slice numbers (5-15).

Main Results:

• Extrapolation from 10 slices showed minimal bias (<1%) for lung compartments and acceptable LOA (±2.5%) in both species.
• Total lung volume and mass showed low bias (sheep: 18.4 ml, 4.2 g; pigs: 5.1 ml, 1.6 g) with 10 slices.
• Accuracy decreased significantly with fewer than 10 reference CT slices.

Conclusions:

• The 10-slice CT extrapolation method is accurate and robust across different thoracic anatomies (ovine and porcine).
• This validated method offers a promising approach for reducing radiation exposure and analysis time in quantitative CT.
• The findings support the clinical and experimental utility of this optimized quantitative CT technique.