Quantitative analysis on the CT diagnosis value of peripheral solitary pulmonary nodule with nodule ;notch due to pleural indentation
Wufei Chen,Yibo Sun,Fang Lu,Dingbiao Mao,Wangernuo,Yanqing Hua
DOI: https://doi.org/10.3760/cma.j.issn.2095-7041.2015.03.007
2015-01-01
Abstract:Objective To evaluate diagnostic value of SPN with NNPI by computed tomography through the method of quantitative analysis. Methods CT findings of 90 cases of SPN with NNPI proved pathologically were analyzed retrospectively. There were 42 males and 48 females with an average age of 58 years(31-76). GE 4. 4 Workstation was used to measure the volume (V) of SPN, the maximum cross-sectional area ( Smax) of NNPI and the length ( d) from the center of the Smax to the vertice of NNPI. Pearson correlation analysis was used to analyze the correlation between the three values, Drag coefficient ( K) of benign and malignant SPN was calculated respectively according to the relevant parameter measured above. Differences between benign and malignant K value were analyzed by Rank-sum test. Receiver operating characteristic ( ROC ) curves were used to analyze the ratios. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy were compared with the morphology method. Results The maximum cross-sectional area ( Smax ) were moderately positive related with volume ( V ) (r=0. 65,P<0. 05), but highly negative correlated with length (d) (r= -0. 83,P<0. 05). Statistical difference of K value was significant between benign and malignant SPN (P<0. 05). However, there was no statistical difference between different types of benign SPN (P>0. 05). On the basis of the ROC curves, the cut off value for K value was 0. 052, AUC was 0. 852, when the ratio was greater than 0. 052, sensitivity, specificity, positive predictive value, negative predictive value and accuracy was 78. 0%, 92. 5%, 92. 8%, 77. 1%, 84. 4%, respectively, which represented an improvement over values obtained by conventional morphology criteria ( sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 64. 0%, 82. 5%, 82. 1%, 64. 7%, 72. 2%, respectively). Conclusions A higher performance level is achieved by quantitative analysis than conventional morphology criteria. Calculation of K value is a simple technique for evaluating diagnosis value of SPN with NNPI.