Objective To investigate the risk model based on preoperative dynamic contrast enhancement MRI (DCE-MRI) characteristics to predict the presence of microvascular invasion in hepatocellular carcinoma at an early stage.Methods A total of 242 HCC patients undergoing surgical treatment admitted to our hospital from January 2020 to April 2022 were included. Clinical laboratory data of patients were collected. DCE-MRI scans were performed on all patients 2 weeks before surgery. The quantitative and qualitative indicators of dynamic contrast-enhanced MRI were analyzed. HCC was divided into MVI group and control group according to the presence or absence of microvascular invasion. The clinical laboratory data of the two groups were compared. The factors influencing MVI in HCC were analyzed. All patients were divided into the training set (170 cases) and the test set (72 cases) according to the ratio of 7〖DK〗∶3. Based on the data of training set, R software was used to build a nomogram model, and the effectiveness of the model was verified in the test set.Results A total of 70 MVI cases occurred in 170 patients in the training set. After grouping and comparing the clinical data of the two groups, the indexes of Ktrans,Kep, Ve and Vp in MVsI group were significantly higher than those in control group. The proportion of irregular tumor margin and low tumor signal in hepatobiliary stage in MVI group was higher than that in control group (P<0.05). Multi-factor Logistics regression analysis showed that Ktrans,Ve,Kep, Vp, irregular tumor margin and low peritumbal signal in hepatobiliary stage were independent correlated factors with MVI in HCC (P<0.05). By constructing a nomogram prediction model, the results showed that the consistency index (C-index) of the model in predicting the risk of MIV in HCC was 0.948. External validation showed that the sensitivity was 88.00% and the specificity was 96.36%.Conclusion Ktrans, Ve, Kep, Vp, irregular tumor margins, and low perihepatobiliary tumor signal are independent factors related to the presence of MVI in HCC. The risk model based on preoperative DCE-MRI characteristics has good predictive efficacy in predicting MVI in HCC at an early stage.