RESUMO
Objective@#To establish a CT image radiomics-based prediction model for the differential diagnosis of silicosis and tuberculosis nodules.@*Methods@#A total of 53 patients with silicosis and 89 patients with tuberculosis who underwent routine CT scans in Suzhou Fifth People's Hospital from January to August, 2018 were enrolled in this study. AK/ITK software was used to segment the images to obtain 139 silicosis lesions and 119 tuberculosis lesions. For each lesion image, 396 features were extracted, and feature dimension reduction was applied to select the most characteristic feature subset. Support vector machine (SVM) , feedforward back propagation neural network (FNN-BP) , and random forest (RF) were implemented using R software (Rstudio V1.1.463) , and the algorithm that achieved the largest area under of the receiver operating characteristic (ROC) curve (AUC) was selected as the final prediction model.@*Results@#RF was the best prediction model for the differential diagnosis of silicosis and tuberculosis nodules, with an accuracy of 83.1%, a sensitivity of 0.76, a specificity of 0.9, and an AUC of 0.917 (95% confidence interval: 0.8431-0.9758) . RF had a significantly larger AUC than SVM and FNN-BP (P<0.05) .@*Conclusion@#CT image-based RF prediction model can be used to differentially diagnose silicosis and tuberculosis nodules.
RESUMO
Objective To evaluate the correlation of CT enhancement parameters with Fuhrman grade in T1 (≤7 cm) clear cell renal carcinoma(ccRCC).Methods From September 2011 to November 2017,102 post-operation cases in our hospital proven to be T1 ccRCC were retrospectively analyzed.There were 71 males and 31 females,with a mean age of (59.1 ± 12.7)years (26 ～79 years),mean body mass index(BMI) of (24.0 ± 2.8)kg/m2 (14.3 ～ 31.6 kg/m2).Tumors of 55 patients were in left kidneys,47 in right kidneys.Fuhrman grade 1 and 2 were defined as low-grade group,meanwhile high-grade group included grade 3 and 4.There were 46 males and 21 females in low-grade group,with a mean age of (59.0 ± 13.2) years,mean BMI of (24.0 ± 2.9) kg/m2.In high-grade group,there were 25 males and 10 females,with a mean age of (58.8 ± 11.8) years,mean BMI of (24.2 ± 2.7) kg/m2.The maximum diameter and tumor enhancement value (TEVX),relative enhancement value (REVX) were measured and calculated.In arterial phase,X =1;in venous phase X =2.The total consumption amount of iodine was recorded.Comparisons of maximum diameter,TEV1,TEV2,REV1,REV2 and the total consumption of iodine between the two different groups were performed.The ROC curves of TEV1,TEV2,REV1,and REV2 were drawn to predict the grade of tumors..Results The TEV1 [(146.1 ± 29.1) HU vs.(100.2 ± 32.1) HU],TEV2 [(98.2 ± 22.9) HU vs.(75.6 ± 25.7) HU],REV1 (1.12 ± 0.24 vs.0.70 ± 0.16),REV2(0.67 ± 0.17 vs.0.54 ± 0.18) between low-grade group and high-grade group had statistical difference (P ＜ 0.05).There was no significant difference in the maximum diameter[(41.8 ± 15.4)mm vs.(45.3 ± 17.0)mm] and the total consumption of iodine [(33.3 ± 5.0)g vs.(34.2 ± 4.4)g] between the two groups (P ＞ 0.05).The area under ROC curve of TEV1,TEV2,REV1 and REV2 were 0.856,0.755,0.901 and 0.728,respectively.REV1 had the highest distinguish efficiency and the best critical value was 0.93.Conclusions The enhancement parameters of T1 ccRCC could contribute to predicting the Fuhrman grade.When the REV1 ≤0.93,the tumor tended to be high-grade tumor(Fuhrman grade 3-4).