[The prediction value of enhanced magnetic resonance imaging nomogram model for dual phenotype hepatocellular carcinoma]

Objective: To explore the value of enhanced magnetic resonance imaging nomogram model in the prediction of dual-phenotype hepatocellular carcinoma(DPHCC). Methods: Data of 116 patients of hepatocellular carcinoma (HCC) confirmed by postoperative pathology, who underwent preoperative enhanced MRI between January 2016 and March 2021 in the First Affiliated Hospital of Soochow University were retrospectively evaluated, of these, there are 87 males and 28 females, aged 30-79 (59±10) years, including 31 patients with DPHCC and 85 patients with non-DPHCC. The patients were randomly divided into training set(51 cases of non-DPHCC,19 cases of DPHCC)and validation set(34 cases of non-DPHCC, 12 cases of DPHCC) in a ratio of 6∶4, according to random number table,clinical and imaging characteristics of the two groups were compared. The statistically significant parameters were included in multivariate logistic regression to identify the independent predictors and for the establishment of the nomogram model. The receiver operating characteristic curves were used to evaluate the prediction ability of the models, the corrected curve was used to validate the model. Results: In the training group, the proportions of rim arterial phase hyperenhancement in the DPHCC was significantly higher than that of the non-DPHCC [47.4%(9/19)vs 7.8%(4/51),P<0.001]. Rim arterial phase hyper-enhancement and enhanced capsule were significant predictors for DPHCC[OR=10.17(1.70-60.80),0.17(0.03-0.93),all P<0.05]. In the training group, the area under curve (AUC), sensitivity and specificity of the nomogram were 0.888 (95%CI: 0.806-0.969), 78.9% and 86.3%. In the validation group, the above three indicators were 0.811(95%CI: 0.655-0.968), 75.0% and 82.4%. Conclusion: Enhanced MRI nomogram model has certain value in prediction of DPHCC, with high sensitivity and specificity.