Objective To construct a quantitative diagnostic nomogram model by analyzing the clinical information of patients and the characteristics of multimodal ultrasound images of liver tumors, so as to predict the malignant probability of liver tumors and provide an effective reference for clinical decision-making. Methods Routine ultrasound and ultrasound elastography of 82 patients with liver tumors were retrospectively analyzed. The ultrasonographic features of routine ultrasound were monitored, including tumor echo, shape, size, boundary, capsule and calcification. The maximum elastic value, minimum elastic value and average elastic value of liver tumor were measured by ultrasound elastography. Lasso regression was used to select the variables of the influencing factors, construct the Nomogram model, calculate the C-index, and draw the calibration curve and clinical decision curve of the prediction model. Results The results of Lasso regression analysis showed that the maximum elasticity of tumor, history of viral hepatitis and tumor hypoecho were independent predictors of malignant tumor, and the matrix model was constructed based on the above factors. The C index of the model was 0.98. The calibration curve showed that the model had good consistency, and the clinical decision curve also showed that the model had good clinical applicability. Conclusions The nomogram model based on multimodal ultrasound imaging technology can distinguish benign and malignant liver tumors. The method is simple and effective, and can be applied to the screening of liver tumors in primary hospitals.