Establishment of a mathematical model for solitary pulmonary cavity based on 18F-FDG PET/CT
Objective To establish a mathematical model of 18F-FDG PET/CT for the diagnosis of benign and malignant solitary pulmonary cavity by Logistic regression analysis.Methods We analyzed 92 patients with solitary pulmonary cavity who underwent18F-FDG PET/CT examination in the PET/CT Center of China-Japan Union Hospital of Jilin University from December 2020 to February 2023,including 61 males and 31 females.The general data and image features of these patients were analyzed by univariate analysis and multivariate Logistic regression analysis.The independent predictors of malignancy were esti-mated,then the mathematical model was established.The diagnostic capabilities of the model,CT,PET and PET/CT for solitary pulmonary cavity were compared by comparing the area under the receiver operat-ing characteristic curve(AUC)of them.Results 34.78％of the cavities were benign,and 65.22％of the cavities were malignant in 92 patients.Univariate analysis showed that the following factors were statis-tically significant between the benign and malignant groups:SUVmax,SUVmax/SUVmean liver,SUVmax/SUVmean spleen,endocaval tubercle and eccentric cavity.The mathematical model established by Logistic regression analysis was y=exp(x)/[1+exp(x)],x=-2.022+0.419 ×(SUVmax/SUVmean spleen)-[0.462 ×(maximum diameter of lesion)+2.023 × lobulation+1.341 × eccentric cavity.When the cut-off value was set at y=0.51,the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of our model were 91.70％,75.00％,83.35％,78.58％and 90.04％,respectively.The AUC of the model was 0.875[95％confidence interval(CI):0.795 to 0.955],which was greater than that of CT(Z=2.424,P＜0.05)and SUVmax/SUVmean spleen(Z=3.276,P＜0.05),and had no significant difference from that of PET/CT(Z=-0.249,P＞0.05).Conclusion The mathematical model of PET/CT established by Logistic regression analysis has good diagnostic performance in the diagno-sis of benign and malignant solitary pulmonary cavity and high clinical value.
NETL
NSTL
万方数据
