Radiomics Analysis of Contrast-Enhanced CT for Staging Liver Fibrosis: An Update for Image Biomarker

Background: To establish and validate a radiomics-based model for staging liver fibrosis at contrast-enhanced CT images. Materials Methods: This retrospective study developed two models (R-score: radiomics signature; R-fibrosis: integrate radiomic serum variables) in training cohort of 332 patients (median age, 59 years; interquartile range, 51-67 256 men) with biopsy-proven who underwent between January 2017 December 2020. Radiomic features were extracted from non-contrast, arterial portal phase images selected by using the least absolute shrinkage selection operator (LASSO) logistic regression to differentiate stage F3-F4 F0-F2. Optimal cutoffs diagnose significant (stage F2-F4), advanced F3-F4) cirrhosis F4) determined receiver operating characteristic curve analysis. Diagnostic performance was evaluated area under curve, Obuchowski index, calibrations decision An internal validation conducted 111 randomly assigned 58 49-66 89 men). Results: In cohort, R-score R-fibrosis (Obuchowski 0.843 0.846, respectively) significantly outperformed aspartate transaminase-to-platelet ratio (APRI) 0.651; P < .001) fibrosis-4 index (FIB-4) 0.676; fibrosis. By cutoffs, had sensitivity range 70%-87%, specificity 71%-97%, accuracy 82%-86% diagnosing fibrosis, cirrhosis. Conclusion: analysis can reach great diagnostic fibrosis. Funding Information: work supported grants National Natural Science Youth Foundation (Grant Number 81902415 Y.Y.), Jiangsu Province BK20190116 Y.Y.) Key Laboratory Imaging Diagnosis Minimally Invasive Interventional Research Zhejiang (No. YXJR202002, J.W). Declaration Interests: The authors declare that they have no competing interests. Ethics Approval Statement: approved institutional review board Affiliated Drum Tower Hospital Nanjing University Medical School. requirement written informed consent waived due its nature.