Detection of artificial pulmonary lung nodules in Ultralow-Dose CT using an ex vivo lung phantom

Purpose:

To assess the image quality of 3 different ultra low-dose CT protocols with model based iterative reconstruction (IR) on pulmonary nodule depiction in a ventilated ex vivo-system.

Methods:

Two porcine lungs were inflated inside a dedicated chest phantom and prepared with n = 25 artificial nodules (0.5 – 1 ml). The artificial chest wall was filled with water to simulate the absorption of a human chest. Images were acquired with a 2 × 192-row detector CT using normal, low-dose and 3 different ultra low-dose (ULD) protocols (effective dosis: respectively 4 mSv). A different tube voltage was used for each ULD protocol: 70 kV, 100 kV with tin filter and 150 kV with tin filter. Tube current was automatically adapted. Nodule delineation was assessed by two observers (scores 1 – 5, 1 = unsure, 5 = high confidence).

Results:

The diameter of the 25 detected artificial nodules ranged from 8 to18.6 mm (mean value 13.4 mm, standard deviation 2.9 mm; n = 20 nodules were solid and n = 5 part-solid). The delineation of nodules due to image quality was lowest in images acquired with ultra low-dose protocol with tin filter at 150 kV (mean value of Score: 3.32) compared to normal protocol (gold standard: 100 kV, 140 mAs, without tin filter, effective dosis: 4 mSv) (p = 0.001). Surprisingly, ultra low-dose protocols at 70 kV showed a better delineation (mean value of Score: 3.9) compared to ultra low-dose protocol at 100 kV with tin filter (p = 0.81). A low score of 2 (impaired diagnostic confidence) was rated once with ultra low-dose protocol at 70 kV due to diagnostically relevant artifacts close to diaphragm and twice with at ultra low-dose protocol at 150 kV with tin filter due to low contrast and high image noise.

Conclusion:

The ex-vivo study in patient scale demonstrates the good image quality of ultra low-dose CT and the feasibility of ultra low dose screening.

*Presenting author