Chinese Hepatolgy ›› 2026, Vol. 31 ›› Issue (9): 1314-1318.

• Liver Cancer • Previous Articles     Next Articles

Application of multiparametric MRI in differentiating benign and malignant hepatic tumors

Yang Fan1, Cheng Yitao1, Wang Lizhi1, Lin Weixin2   

  1. 1. Department of Imaging, Yingshang County People's Hospital, Fuyang 236200, China;
    2. Department of Radiology, Wujin Hospital of Traditional Chinese Medicine, Changzhou 213000, China
  • Received:2025-03-15 Online:2026-09-30 Published:2026-11-16
  • Contact: Lin Weixin, Email: frankxyx@163.com

Abstract: Objective To investigate the value of multiparametric magnetic resonance imaging (MRI) in differentiating benign and malignant hepatic tumors, and to evaluate the diagnostic performance of conventional MRI features, dynamic contrast-enhanced characteristics, and diffusion-weighted imaging (DWI)/intravoxel incoherent motion (IVIM) parameters. Methods A total of 201 patients from the Yingshang County People's Hospital and Wujin Hospital of Traditional Chinese Medicine with hepatic lesions confirmed by pathology or imaging follow-up were retrospectively included, comprising 44 cases of hepatic hemangioma (HH), 68 cases of focal nodular hyperplasia (FNH), 55 cases of hepatocellular carcinoma (HCC), and 34 cases of liver metastases. Conventional MRI, DWI and/or dynamic contrast-enhanced scanning were performed. Clinical characteristics and MRI features were compared among groups. Apparent diffusion coefficient (ADC) and perfusion fraction (f value) were measured. Logistic regression analysis was performed to identify independent predictors for differentiating benign and malignant tumors. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of individual parameters and the combined model. Results Significant differences were observed among the groups in age, lesion diameter, AFP levels, liver function indices, and MRI features (all P<0.05). The HCC group was significantly older than the HH and FNH groups [(58.9±11.2) years vs. (45.2±10.6) years and (38.7±9.4) years], and AFP levels were markedly elevated [356 (78, 1 250) ng/mL vs. 3.8 (2.2, 6.1) ng/mL and 4.5 (2.8, 6.9) ng/mL, P<0.05]. Conventional MRI showed that benign lesions were more likely to present with well-defined margins and regular morphology, whereas malignant lesions were frequently associated with irregular borders and necrosis. On dynamic contrast-enhanced MRI, HH predominantly exhibited progressive enhancement (90.3%), FNH mainly showed persistent enhancement (88.1%), while HCC and metastatic tumors commonly demonstrated portal venous washout patterns (82.3% and 58.1%, respectively). DWI/IVIM analysis revealed that ADC and f values were significantly higher in benign lesions than in malignant lesions. The highest ADC value was observed in the HH group [(1.98±0.32) ×10-3 mm2/s], whereas the lowest was found in the metastatic tumor group [(0.98±0.18) ×10-3 mm2/s]. The highest f value was observed in the FNH group (35.8±7.2)%, while the lowest was observed in the metastatic tumor group (20.7±4.8)% (all P<0.05). Multivariate logistic regression analysis demonstrated that portal venous washout (OR=8.59, 95%CI: 3.41~21.62), rim enhancement (OR=6.62, 95%CI: 2.73~16.02), and ADC value (OR=0.09, 95%CI: 0.03~0.25) were independent predictors for differentiating benign and malignant hepatic tumors (all P<0.05). ROC analysis showed that the combined multiparametric MRI model achieved the highest diagnostic performance, with an AUC of 0.95, sensitivity of 90.2%, and specificity of 88.6%. Conclusion Multiparametric MRI can comprehensively reflect the biological characteristics of hepatic tumors from morphological, diffusion, and perfusion perspectives, and has high diagnostic value in differentiating benign and malignant lesions. The combined analysis of multiple parameters significantly improves diagnostic performance and provides an important basis for noninvasive clinical diagnosis.

Key words: Multiparametric MRI, Hepatic hemangioma, Focal nodular hyperplasia, Hepatocellular carcinoma