Objective To discuss the value of clinical radiomic nomogram(CRN)and deep convolutional neural network(DCNN)in distinguishing atypical pulmonary hamartoma(APH)from atypical lung adenocarcinoma(ALA).Methods A total of 307 patients were retrospectively recruited from two institutions.Patients in institu-tion 1 were randomly divided into the training(n=184:APH=97,ALA=87)and internal validation sets(n=79:APH=41,ALA=38)in a ratio of 7∶3,and patients in institution 2 were assigned as the external validation set(n=44:APH=23,ALA=21).A CRN model and a DCNN model were established,respectively,and the performances of two models were compared by delong test and receiver operating characteristic(ROC)curves.A human-machine competition was conducted to evaluate the value of AI in the Lung-RADS classification.Results The areas under the curve(AUCs)of DCNN model were higher than those of CRN model in the training,internal and external validation sets(0.983 vs 0.968,0.973 vs 0.953,and 0.942 vs 0.932,respectively),however,the differences were not statistically significant(p=0.23,0.31 and 0.34,respectively).With a radiologist-AI com-petition experiment,AI tended to downgrade more Lung-RADS categories in APH and affirm more Lung-RADS cat-egories in ALA than radiologists.Conclusion Both DCNN and CRN have higher value in distinguishing APH from ALA,with the former performing better.AI is superior to radiologists in evaluating the Lung-RADS classification of pulmonary nodules.

Objective To investigate the value of radiomics nomogram model based on enhanced CT for predicting postoperative pancreatic fistula(POPF)in patients undergoing pancreaticoduodenectomy(PD).Methods The clinical and abdominal imaging data of 82 patients who underwent PD were analyzed retrospectively.They were divided into training set and validation set at a ratio of 3∶2.In the venous phase of CT,the maximum slice of the pancreatic duct was selected to delineate the region of interest(ROI),and the features were extracted and screened to construct the radiomics model.The independent risk factors associated with POPF were screened by univariate and multivariable logistic regression analyses to construct the clinical model.A nomogram model was created by combining Radiomics score(Radscore)and clinical risk factors.The diagnostic performance of the model was verified by the vali-dation set.Results The nomogram model demonstrated significant predictive power,with an area under the curve(AUC)of 0.862 and 0.806 in the training set and validation set,respectively.Conclusion The nomogram model exhibits excellent predictive performance and outperforms the clinical model and radiomics model in predicting POPF,which can provide important guidance for clinical deci-sion-making.

Objective To investigate the clinical effect of direct-acting antiviral agent (DAA) in the treatment of chronic hepatitis C (CHC) patients with thrombocytopenia and its effect on platelet count (PLT). Methods A retrospective analysis was performed for 83 CHC patients with thrombocytopenia (PLT < 150×10 9 /L) who received the DAA treatment regimen without interferon for 12-24 weeks in Tianjin Third Central Hospital from April 2018 to March 2019, and the changes in virologic response, liver function parameters, PLT, and liver stiffness measurement (LSM) were evaluated at the end of treatment (EOT) and at week 12 after EOT. Quantitative data accord with normal distribution were compared by repeated measures ANOVA. Normal transformation was performed before the comparison between skewed data, then repeated measures ANOVA was carried out. A logistic regression analysis was used to investigate the predictive factors for PLT elevation, and the receiver operating characteristic (ROC) curve was plotted to analyze the value of LSM in predicting PLT elevation after treatment. Results Among the 83 CHC patients with thrombocytopenia, 61.4% had liver cirrhosis, and the rate of sustained virologic response at week 12 after the end of treatment (SVR12) was 98.8%. From baseline to EOT and SVR12, the patients had significant reductions in the serum levels of aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transpeptidase, total bilirubin, and globin, a significant increase in the serum level of albumin, and a significant reduction in LSM (all P < 0.05). For all patients, PLT at EOT and SVR12 was significantly higher than that at baseline [EOT vs baseline: (110.4±44.6)×10 9 /L vs (97.8±33.2)×10 9 /L, P < 0.01; SVR12 vs baseline: (109.0±47.7)×10 9 /L vs (97.8±33.2)×10 9 /L, P < 0.01]. At SVR12, there were significant differences in the proportion of patients with liver cirrhosis, baseline LSM, and baseline white blood cell count between the PLT elevation group and the non-PLT elevation group (all P < 0.05). The multivariate logistic regression analysis showed that LSM was an independent predictive factor for significant PLT elevation after DAA treatment (odds ratio=0.929, 95% confidence interval: 0.864-0.999, P < 0.05). Baseline LSM had an area under the ROC curve of 0.644 in predicting PLT elevation, with a sensitivity of 81.0% and a specificity of 48.6% at a cut-off value of 20.15 kPa. The patients with PLT > 100×10 9 /L at baseline had a greater increase in PLT( P < 0.05). Conclusion CHC patients with thrombocytopenia have significant improvements in liver function and LSM after receiving DAA treatment and obtaining SVR12, and baseline LSM is an independent predictive factor for PLT elevation. There is a significant increase in PLT from baseline to EOT and SVR12.