Differential diagnosis model of hepatocellular carcinoma and intrahepatic cholangiocarcinoma based on serological indexes
Objective To develop a differential model between hepatocellular carcinoma(HCC)and intrahepatic cholangiocarcinoma(ICC)using routine serological markers and to evaluate its accuracy.Methods To retrospectively analyze clinical data from 380 patients diag-nosed with either HCC or ICC in the First Affiliated Hospital of Zhengzhou University from June 2021 to June 2023.Comprehensive clinical data were collected,including hepatitis virus history,liver function,tumor markers,and data from percutaneous or post-operative pathology.Key variables for differentiating HCC and ICC were identified by using lasso regression and stepwise regression.Based on this,we created a differen-tial diagnosis model,drew the nomogram and performed comprehensive model evaluation by RSTUDIO.Results The variables with VIF＞5 were screened out by collinear diagnosis,and seven key variables including Gender,HV,AFP,CA19-9,GGT,ALP,and TBIL were screened by Lasso regression.And then determined five core discriminative variables(Gender,HV,AFP,CA19-9,and ALP)by Stepwise regression.The ROC curve for the constructed logistic regression model had an AUC value of 0.986,with a sensitivity of 0.942 and specificity of 0.958.The calibration curve shows that the prediction probability of the model fits well with the actual observation,and the decision curve a-nalysis shows that the model has excellent clinical benefit.The differential model based on routine serological indexes constructed in this study is excellent in the differential diagnosis of HCC and ICC,and has broad clinical application potential.Conclusion Our differential model based on routine serological markers exhibited outstanding performance in distinguishing between HCC and ICC and had significant clinical ap-plicability potential.
Hepatocellular carcinomaIntrahepatic cholangiocarcinomaNomogramDiagnostic model
万方数据
维普
