Radiomics Model Based on Gadoxetic Acid Disodium-Enhanced MR Imaging to Predict Hepatocellular Carcinoma Recurrence After Curative Ablation

• DOI: 10.2147/cmar.s300627
• Publication Date: 2021-03-25T00:41:46Z
• Authors: Ling Zhang; Peiqiang Cai; Jingyu Hou; Ma Luo; Yonggang Li; Xinhua Jiang
• Keywords: Gadoxetic acid

A practical prognostic prediction model is absent for hepatocellular carcinoma (HCC) patients after curative ablation. We aimed to develop a radiomics based on gadoxetic acid disodium-enhanced magnetic resonance (MR) images predict HCC recurrence ablation.We retrospectively enrolled 132 with who underwent Patients were randomly divided into the training (n = 92) and validation 40) cohorts. Radiomic features extracted from MR of liver before ablation, various baseline clinical characteristics collected. Cox regression random survival forests used construct models that incorporated radiomic and/or characteristics. The predictive performance different was compared using concordance index (C-index) decision curves analysis (DCA). cutoff derived combined risk categorization, recurrence-free (RFS) between groups Kaplan-Meier curve analysis.Twenty four identified construction. constructed by tumoral peritumoral had better (C-index 0.698, 95% confidence interval [CI] 0.640-0.755) 0.614, CI 0.499-0.695), while best 0.706, 0.638-0.763). net benefit observed other two according DCA. Distinct RFS distributions when categorized (Log rank test, p 0.007).The which could