Differentiation of small (≤ 3 cm) hepatocellular carcinomas from benign nodules in cirrhotic liver: the added additive value of MRI-based radiomics analysis to LI-RADS version 2018 algorithm

Abstract Background Accurate characterization of small nodules in a cirrhotic liver is challenging. We aimed to determine the additive value MRI-based radiomics analysis Liver Imaging Reporting and Data System version 2018 (LI-RADS v 2018) algorithm differentiating (≤ 3 cm) hepatocellular carcinomas (HCCs) from benign liver. Methods In this retrospective study, 150 cirrhosis patients with histopathologically confirmed (HCC, 112; nodules, 44) were evaluated January 2013 October 2018. Based on LI-RADS algorithm, category was assigned for each lesion. A signature generated based texture features extracted T1-weighted, T2W, apparent diffusion coefficient (ADC) images by using least absolute shrinkage selection operator regression model. nomogram model developed combined diagnosis. Diagnostic performance assessed receiver operating characteristic curve (ROC) analysis. Results consisting eight significantly associated differentiation HCCs nodules. Both (area under ROC [A z ] = 0.898) MRI-Based (A 0.917) demonstrated good discrimination, showed superior classification 0.975). Compared alone, approach improved specificity (97.7% vs 81.8%, p 0.030) positive predictive (99.1% 92.9%, 0.031) afforded comparable sensitivity (97.3% 93.8%, 0.215) negative (93.5% 83.7%, 0.188). Conclusions