PURPOSE: To determine whether extraction of pulmonary vessels from com
puted tomographic (CT) images with automated segmentation improves the
detection of pulmonary nodules. MATERIALS AND METHODS: Simulated nodu
les were superimposed on normal spiral CT images. Eight patients refer
red for CT assessment of pulmonary nodules were selected for clinical
evaluation Vessels were extracted from both the simulation and clinica
l study with a three-dimensional seeded region-growing algorithm. Thre
e experienced radiologists were asked to locate the nodules and assign
a level of confidence to their findings. Sensitivity and proportion o
f false-positive results per case (FPC) were calculated. Observer perf
ormance was evaluated by alternate free-response receiver operating ch
aracteristic analysis. RESULTS: Extraction of vascular structures from
CT scans improved sensitivity from 63% to 84% in the simulation study
and from 58% to 78% in the clinical study. The proportion of FPC decr
eased from 52% to 24% and from 55% to 12% respectively. Radiologists p
erformed Consistently better with the segmented images than with the o
riginal images in both the simulation (P = .006) and the clinical (P =
.0013) study. CONCLUSION: Automated vessel subtraction and extraction
improves detection of pulmonary nodules.