摘要
This study examined the value of volume rendering （VR） interpretation in assessing the growth of pulmonary nodular ground-glass opacity （nGGO）. A total of 47 nGGOs （average size, 9.5 mm; range, 5.7-20.6 mm） were observed by CT scanning at different time under identical parameter settings. The growth of nGGO was analyzed by three radiologists by comparing the thin slice （TS） CT images of initial and repeat scans with side-by-side cine mode. One week later synchronized VR images of the two scans were compared by side-by-side cine mode to evaluate the nGGO growth. The nodule growth was rated on a 5-degree scale： notable growth, slight growth, dubious growth, stagnant growth, shrinkage. Growth standard was defined as： Density increase 〉 30 HU and （or） diameter increase （by 20% in nodules 〉_10 mm, 30% in nodules of 5-9 mm）. Receiver operating characteristic （ROC） was performed. The results showed that 32 nGGOs met the growth criteria （29 nGGOs showed an increase in density; 1 nGGO showed an increase in diameter; 2 nGGOs showed an increase in both diameter and density）. Area under ROC curve revealed that the performance with VR interpretation was better than that with TS interpretation （P〈0.01, P〈0.05 and P〈0.05 for observers A, B and C respectively）. Consistency between different observers was excellent with both VR interpretation （κ=0.89 for observers A＆C A＆B, B＆C） and TS interpretation （κ=0.71 for A＆B, κ=0.68 for A＆C, κ=0.74 for B＆C）, but time spending was less with VR interpretation than with TS interpretation （P〈0.0001, P〈0.0001 and P〈0.05 for observers A, B and C, respectively）. It was concluded that VR is a useful technique for evaluating the growth of nGGO.
This study examined the value of volume rendering （VR） interpretation in assessing the growth of pulmonary nodular ground-glass opacity （nGGO）. A total of 47 nGGOs （average size, 9.5 mm; range, 5.7-20.6 mm） were observed by CT scanning at different time under identical parameter settings. The growth of nGGO was analyzed by three radiologists by comparing the thin slice （TS） CT images of initial and repeat scans with side-by-side cine mode. One week later synchronized VR images of the two scans were compared by side-by-side cine mode to evaluate the nGGO growth. The nodule growth was rated on a 5-degree scale： notable growth, slight growth, dubious growth, stagnant growth, shrinkage. Growth standard was defined as： Density increase 〉 30 HU and （or） diameter increase （by 20% in nodules 〉_10 mm, 30% in nodules of 5-9 mm）. Receiver operating characteristic （ROC） was performed. The results showed that 32 nGGOs met the growth criteria （29 nGGOs showed an increase in density; 1 nGGO showed an increase in diameter; 2 nGGOs showed an increase in both diameter and density）. Area under ROC curve revealed that the performance with VR interpretation was better than that with TS interpretation （P〈0.01, P〈0.05 and P〈0.05 for observers A, B and C respectively）. Consistency between different observers was excellent with both VR interpretation （κ=0.89 for observers A＆C A＆B, B＆C） and TS interpretation （κ=0.71 for A＆B, κ=0.68 for A＆C, κ=0.74 for B＆C）, but time spending was less with VR interpretation than with TS interpretation （P〈0.0001, P〈0.0001 and P〈0.05 for observers A, B and C, respectively）. It was concluded that VR is a useful technique for evaluating the growth of nGGO.
基金
supported by a grant from the Science and Technology Program of Guangdong Province of China(No.2009B030801120)
