Duo Liu MD, Abstract Co-Author: Nothing to Disclose

Yoshinori Funama PhD, Abstract Co-Author: Nothing to Disclose

Kazuo Awai MD, Presenter: Nothing to Disclose

Ori Hay, Abstract Co-Author: Nothing to Disclose

Yoshiharu Nakayama MD, Abstract Co-Author: Nothing to Disclose

Yasuyuki Yamashita MD, Abstract Co-Author: Nothing to Disclose

Natsuko Sato BS, Abstract Co-Author: Nothing to Disclose

Masahiro Hatemura RT, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To evaluate the effect of the radiation dose on the detection and characterization of ground-glass opacity (GGO) in the lungs by high-resolution CT (HRCT)

A chest CT phantom that featured simulated GGO was scanned with a 40-detector row CT scanner (Brilliance 40, Philips). The detector row width was 1.25 mm, pitch 0.719, rotation time 0.5 sec, reconstructed slice thickness and interval 1.25 mm. The tube current and rotation time were 15, 22.5, 45, 90, and 180 mAs. The CT number of the simulated lungs and GGO in the phantom was -900 and -800 HU, respectively. Simulated GGO measuring 6, 8, 10 mm in diameter were placed at the lung apex-, tracheal bifurcation-, and lung base level. We carried out observer performance tests using receiver operating characteristic (ROC) analysis to evaluate the detectability of the GGO at each mAs setting used. For ROC analysis we extracted 10 x 30 mm areas with or without GGO from the chest phantom image scanned at the different mAs. We then prepared a total of 450 images (5 mAs settings x 30 images with and without GGO x 3 scan levels). The detectability of GGO was estimated with the area under the ROC curves (Az values). Using a 5-point scale, we also visually evaluated the graininess of lung fields, the homogeneity and marginal sharpness of the GGO, and overall image quality.

On ROC analysis there was no statistical difference in the detectability of GGO among images obtained with 22.5, 45, 90 mAs (Az = 0.91, 0.94, and 0.96, respectively), and 180 mAs (Az = 0.98) [Dunnet’s test, p= 0.07, 0.48, and 0.84, respectively]. The mean Az value at 15 mAs (0.68) was significantly lower than at 180 mAs (p0.05 ). The graininess of the lung at 15 and 22.5 mAs was significantly lower than at 180 mAs (p<0.01 and p=0.01)

Except for lung graininess, HRCT scans obtained at 22.5 mAs and 180 mAs were comparable with respect to the detectability of GGO and image quality.

O.H.,N.S.: Ori Hay and Natsuko Sato are employees of Phillips medical systems

To evaluate the effect of the radiation dose on the detection and characterization of ground-glass opacity (GGO) in the lungs by high-resolution CT (HRCT)

A chest CT phantom that featured simulated GGO was scanned with a 40-detector row CT scanner (Brilliance 40, Philips). The detector row width was 1.25 mm, pitch 0.719, rotation time 0.5 sec, reconstructed slice thickness and interval 1.25 mm. The tube current and rotation time were 15, 22.5, 45, 90, and 180 mAs. The CT number of the simulated lungs and GGO in the phantom was -900 and -800 HU, respectively. Simulated GGO measuring 6, 8, 10 mm in diameter were placed at the lung apex-, tracheal bifurcation-, and lung base level. We carried out observer performance tests using receiver operating characteristic (ROC) analysis to evaluate the detectability of the GGO at each mAs setting used. For ROC analysis we extracted 10 x 30 mm areas with or without GGO from the chest phantom image scanned at the different mAs. We then prepared a total of 450 images (5 mAs settings x 30 images with and without GGO x 3 scan levels). The detectability of GGO was estimated with the area under the ROC curves (Az values). Using a 5-point scale, we also visually evaluated the graininess of lung fields, the homogeneity and marginal sharpness of the GGO, and overall image quality.

On ROC analysis there was no statistical difference in the detectability of GGO among images obtained with 22.5, 45, 90 mAs (Az = 0.91, 0.94, and 0.96, respectively), and 180 mAs (Az = 0.98) [Dunnet’s test, p= 0.07, 0.48, and 0.84, respectively]. The mean Az value at 15 mAs (0.68) was significantly lower than at 180 mAs (p0.05 ). The graininess of the lung at 15 and 22.5 mAs was significantly lower than at 180 mAs (p<0.01 and p=0.01)

Except for lung graininess, HRCT scans obtained at 22.5 mAs and 180 mAs were comparable with respect to the detectability of GGO and image quality.

Liu, D, Funama, Y, Awai, K, Hay, O, Nakayama, Y, Yamashita, Y, Sato, N, Hatemura, M, et al, , Detection and Characterization of Ground Glass Opacity in the Lungs by High-Resolution CT: Experimental Comparison of Low- and Standard-Dose Multidetector CT Using a Chest CT Phantom.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4409937.html