Nomogram Based on CT Radiomics Features Combined With Clinical Factors to Predict Ki-67 Expression in Hepatocellular Carcinoma

• DOI: 10.3389/fonc.2022.943942
• Publication Date: 2022-07-06T17:10:01Z
• Authors: Cuiyun Wu; Junfa Chen; Yuqian Fan; Ming Zhao; Xiaodong He; Yuguo Wei; Weidong Ge; Yang Liu
• Keywords: Nomogram

The study developed and validated a radiomics nomogram based on combination of computed tomography (CT) signature clinical factors explored the ability for individualized prediction Ki-67 expression in hepatocellular carcinoma (HCC).First-order, second-order, high-order features were extracted from preoperative enhanced CT images 172 HCC patients, with predictive value high to construct radiomic model. Based training group, was constructed that showed an independent association expression. area under receiver operating characteristic curve (AUC), calibration curve, decision analysis (DCA) used verify performance nomogram.Sixteen higher-order associated (AUC: 0.854; validation 0.744). In multivariate logistic regression, alfa-fetoprotein (AFP) Edmondson grades identified as predictors Thus, combined AFP 0.884; 0.819). DCA good application nomogram.The this can accurately predict provide guidance treatment monitoring patients.