Radiomics Analysis on Noncontrast CT for Distinguishing Hepatic Hemangioma (HH) and Hepatocellular Carcinoma (HCC)

Figure 2

Main working procedures in this research. (a) Volume of interest was individually depicted on CT images at each transverse. (b) Radiomic features include first-order statistics, textural ones, and wavelet transformations. (c) Intraobserver and interobserver reproducibility analysis and LASSO regression were applied in feature selection. (d) Radiomics-based models were established on the basis of selected features. (e) Calibration curves and decision curve analysis were used to evaluate diagnostic performance of two models.