A Nomogram Model For Predicting Hepatocellular Carcinoma With Microvascular Invasion Based On CT Imaging Features

Objective: To access the clinical indicators and CT imaging signs of microvascular invasion（MVI）in hepatocellular carcinoma（HCC）,and to establish a Nomogram model to predict the risk of MVI.Methods: The CT image features and clinical data of 160 patients with postoperatively pathologically confirmed HCC were retrospectively analyzed,and compared with the pathological results,to screen out the predictive factors related to MVI,and the Nomogram method was used to construct a scoring model,and evaluate the predictive ability of the model.Results: Three risk factors including tumor diameter,tumor morphology and Hypodense Halo were screened and the scoring model was constructed by applying the Nomogram.The C-index of the model was 0.703,（95% CI: 0.621～0.784）,and the goodness-of-fit test（χ2=3.433,degrees of freedom 8,P>0.05）.internal validation results of Bootstrap self-sampling method to corrected C-index was 0.683.Conclusion: The Nomogram model constructed by the three CT imaging signs（tumor diameter,tumor morphology and Hypodense Halo）can predict the presence or absence of MVI in HCC patients to a certain extent,and has a certain guiding effect on the clinical treatment decision of HCC patients.