Radiomic Analysis of Intrahepatic Cholangiocarcinoma: Non-Invasive Prediction of Pathology Data: A Multicenter Study to Develop a Clinical–Radiomic Model

Standard imaging cannot assess the pathology details of intrahepatic cholangiocarcinoma (ICC). We investigated whether CT-based radiomics may improve prediction tumor characteristics. All consecutive patients undergoing liver resection for ICC (2009-2019) in six high-volume centers were evaluated inclusion. On preoperative CT, we segmented (Tumor-VOI, i.e., volume-of-interest) and a 5-mm parenchyma rim around (Margin-VOI). considered two types data: grading (G) microvascular invasion (MVI). The predictive models internally validated. Overall, 244 analyzed: 82 (34%) had G3 tumors 139 (57%) MVI. For prediction, clinical model an AUC = 0.69 Accuracy 0.68 at internal cross-validation. addition radiomic features extracted from portal phase CT improved performance (Clinical data+Tumor-VOI: 0.73/Accuracy 0.72; +Tumor-/Margin-VOI: 0.77/Accuracy 0.77). Also MVI 0.75/Accuracy 0.70; +Tumor-VOI: 0.82/Accuracy 0.73; 0.75). permutation tests confirmed that combined clinical–radiomic outperforms purely one (p < 0.05). textural arterial no impact. In conclusion, peritumoral tissue