Volume estimation of multidensity nodules with thoracic computed tomography

Area of Science:

• Medical Imaging
• Radiology
• Quantitative Imaging

Background:

• Accurate lung nodule volumetry is crucial for cancer diagnosis and treatment monitoring.
• Multidensity nodules, such as those with ground glass opacities and solid components, present unique challenges for volume estimation.
• Understanding the impact of imaging parameters on volume estimation accuracy is essential for reliable clinical interpretation.

Purpose of the Study:

• To assess the accuracy of a model-based volume estimator for multidensity lung nodules in a phantom CT study.
• To investigate the influence of various computed tomography (CT) acquisition and reconstruction parameters on volume estimation bias and error.

Main Methods:

• Eight multidensity spherical objects with varying densities and diameters were manufactured.
• Objects were scanned using an anthropomorphic phantom with diverse CT acquisition and reconstruction settings.
• A model-based volume estimator was employed to measure the volumes of the entire object and its inner core.

Main Results:

• Volume estimation bias ranged from [Formula: see text] to 6.6% for the entire object and [Formula: see text] to 5.7% for the inner core.
• Standard deviations for volume estimation error were 1.5%–7.6% (entire object) and 2.0%–17.7% (inner core).
• Reconstructed slice thickness significantly affected volumetric error; exposure and reconstruction kernel did not.

Conclusions:

• Model-based volume estimation of multidensity lung nodules is feasible but subject to bias and error.
• Inner-core volume estimation is more prone to error due to the smaller size of the core.
• Reconstructed slice thickness is a critical parameter influencing the accuracy of lung nodule volumetry in CT.