Original Article
Comparing attenuations of malignant and benign solitary pulmonary nodule using semi-automated region of interest selection on contrast-enhanced CT
Abstract
Background: The purpose of this study was to determine whether semi-automated region of interest (ROI) measurement of CT attenuations of solitary pulmonary nodule (SPN) is an accurate approach in differentiating malignant from benign SPN.
Methods: Ninety cases of pathologically proven SPN were retrospectively reviewed. CT attenuations of SPN before and after contrast injection were measured using semi-automated ROI selection method. Attenuations within a range of −100 to 200 Hounsfield units (HU) as soft tissue density range were set. The ROI included the entire SPN regardless of its internal soft tissue contents after automatic elimination of airs, calcific, or bony densities.
Results: There were 42 (46.7%) malignant SPN and 48 (53.3%) benign SPN, which were grouped into A (18 tuberculoma, 13 fungus), B (5 focal organizing pneumonia, 5 abscess), and C (7 other benign tumors). The malignant SPN showed significantly higher mean attenuations of enhancement and net-enhancement than all benign SPN (P<0.001). Using the area under the receiver operating characteristic curve (AUC), the cut-off net-enhancement of 15 HU gave 83% sensitivity, 65% specificity and 73% accuracy for predicting malignancy. Malignant SPN (mean 67.9 HU) had significantly higher enhancement than group A (mean 52.6 HU, P<0.001, 95% CI: 8.73, 21.81) and group B (mean 57.0 HU, P=0.025, 95% CI: −1.43, 20.34) while group C showed no significant difference (mean 68.1 HU, P=0.97). Net enhancements were higher in group B (mean 18.8 HU) than in group A (mean 8.8 HU) (P<0.001, 95% CI: 11.8, 23.18).
Conclusions: The semi-automated ROI measurement of SPN’s attenuations on CT is an accurate approach in distinguishing indeterminate SPN.
Methods: Ninety cases of pathologically proven SPN were retrospectively reviewed. CT attenuations of SPN before and after contrast injection were measured using semi-automated ROI selection method. Attenuations within a range of −100 to 200 Hounsfield units (HU) as soft tissue density range were set. The ROI included the entire SPN regardless of its internal soft tissue contents after automatic elimination of airs, calcific, or bony densities.
Results: There were 42 (46.7%) malignant SPN and 48 (53.3%) benign SPN, which were grouped into A (18 tuberculoma, 13 fungus), B (5 focal organizing pneumonia, 5 abscess), and C (7 other benign tumors). The malignant SPN showed significantly higher mean attenuations of enhancement and net-enhancement than all benign SPN (P<0.001). Using the area under the receiver operating characteristic curve (AUC), the cut-off net-enhancement of 15 HU gave 83% sensitivity, 65% specificity and 73% accuracy for predicting malignancy. Malignant SPN (mean 67.9 HU) had significantly higher enhancement than group A (mean 52.6 HU, P<0.001, 95% CI: 8.73, 21.81) and group B (mean 57.0 HU, P=0.025, 95% CI: −1.43, 20.34) while group C showed no significant difference (mean 68.1 HU, P=0.97). Net enhancements were higher in group B (mean 18.8 HU) than in group A (mean 8.8 HU) (P<0.001, 95% CI: 11.8, 23.18).
Conclusions: The semi-automated ROI measurement of SPN’s attenuations on CT is an accurate approach in distinguishing indeterminate SPN.
