| Purpose:The aim of the study was to compare the influence of different reconstruction algorithms on quantitative emphysema analysis in patients using MSCT. To compare lung densitometry with visual assessment of pulmonary emphysema with different degrees measured by MSCT, investigating the value of two methods for evaluating various degrees of emphysema.Material and methods:30patients underwent inspiratory MSCT.The raw data were reconstructed using different algorithms:standard algorithm,lung algorithm and edge algorithm.all the30patients were separated into3groups:standard algorithm group,lung algorithm group and edge algorithm group.CT data were analysed using quantification software post-processing such as reformation,threshold limits segmentation to retain the lungs,and providing emphysema index (EI),mean lung attenuation(MLA) and total lung volume(TLV).The difference correlation between the results of groups were evaluated.a large field water phantom was scanned with the same technical parameters and reconstructed algorithms, and mean, maximum and minimum values and standard deviation were measured.30patients with emphysema who underwent MSCT.According to the results of pulmonary function test, all the30subjects were separated into two groups:the mild emphysema group(group A, RV/TLC>40%and<50%,11males,4females, range53-88,mean age67.27±11.11years) and the moderate and severe emphysema group(group B,RV/TLC>50%,10males,5females,range51-84,mean age68.73±10.17years). The emphysema index(EI), visual score(VS) and the time required for assessment were computed. To compare the inter-observer difference and correlation between the results of group A and group B. To compare the correlation of El and VS between the groups. To compare the average time for lung densitometry and for visual score.Results:The differences in El among the standard algorithm group,the lung algorithm group and the edge algorithm group had statistical significance (P<0.05).There was no statistical significance differences in MLA between the standard algorithm group and the edge algorithm group (P>0.05).There was statistical significance differences in MLA between the other groups (P<0.05).The differences in SDMLA among the groups had statistical significance (P<0.05).The differences in TLV among the groups had statistical significance (P>0.05). The results for the water phantom showed that the mean values of edge algorithm group were the most colse to the water. The mean values of edge algorithm group was close to the standard group.The maximum and minimum values and standard deviation of stand,lung and edge reconstruction algorithms tended to increasing.The inter-operator agreement of lung densitometry was better than visual score(r=0.99～1.00VS r=0.35～0.97, P<0.05).The inter-operator agreement of visual score of group B was better than group A(r=0.73～0.97VS r=0.35～0.77, P<0.05). The differences between agreement of densitometry of the group A and group B were slight(r=0.99～1.00VS r=0.98～0.99, P<0.05).The correlation of El and VS of group B was better than group A(r=0.725VS r=0.317, P<0.05).The difference between the time for densitometry (7.88±0.80min) and the time for visual score (4.42±0.40min) was statistically significant(P<0.05).Conclusion:①Reconstruction algorithm may affect El and MLA,but has small effect on TLV in the quantitative analysis of emphysema.②The standard reconstruction algorithm was more suitable for quantitative emphysema analysis in patients with emphysema using MSCT.③There was a correlation between the results of lung densitometry and visual score of pulmonary emphysema measured by MSCT. Lung densitometry enabled a better inter-operator agreement of the results than visual score in evaluating the extent of emphysema.④Lower agreement was observed between operators of visual score, especially in the mild emphysema group. We should adopt the method of lung densitometry for the measurement and follow-up of mild emphysema. |
PDF Full Text Request