Objective: To investigate the risk factors of breast cancer lymph node metastasis, and to construct a nomogram model of breast cancer lymph node metastasis. Methods: Clinical data of 159 patients with breast cancer admitted from April 2021 to August 2023 were retrospectively analyzed, and 159 patients were divided into metastatic group and non-metastatic group according to lymph node metastasis. Multivariate Logistic regression was used to screen the risk factors of lymph node metastasis in breast cancer. In addition, ultrasound was used to collect and extract the imaging features of breast cancer patients, and LASSO was used to analyze and screen the imaging features of breast cancer patients, and the imaging scores were calculated. The nomogram prediction model of breast cancer lymph node metastasis was established by using R language. Finally, area under receiver Operating characteristic (ROC) curve (AUC), calibration curve, and policy curve analysis (DCA) were used to evaluate the differentiation, calibration and clinical impact of the model. Results: Among 159 patients with breast cancer, 66 patients had lymph node metastasis, the incidence of lymph node metastasis was 41.51%. Age, education level, place of residence, diabetes mellitus, family history of tumor, course of disease, tumor echo, tumor margin, posterior echo, calcification, molecular typing, PR, ER and other data were compared between metastatic and non-metastatic groups (P>0.05). Tissue grade, tumor size, vascular invasion, tumor shape, blood flow signal grade, HER-2, Ki-67 and other data were compared (P<0.05). Logistic regression analysis showed that tissue grade, tumor size, vascular invasion, tumor shape, blood flow signal grade and HER-2 were independent risk factors for lymph node metastasis of breast cancer (P<0.05). Four imaging features were obtained by Lasso regression analysis. Based on the results of the above factors and the selection results of ultrasound imaging features of breast nodules, three prediction models were constructed, including clinical model, ultrasound imaging of breast nodules and combined model, with AUC values of 0.822, 0.785 and 0.914, respectively. The area under ROC curve of the nomogram model of breast cancer lymph node metastasis was 0.925, and the actual value of the correction curve was basically consistent with the predicted value. When the decision curve showed that the threshold probability was 2% to 94%, the net benefit value of the nomogram was higher for the prediction of breast cancer lymph node metastasis. Conclusion: Tissue grade, tumor size, vascular invasion, tumor shape, blood flow signal grade, and HER-2 are all factors affecting the occurrence of lymph node metastasis in breast cancer. In this study, the nomogram constructed based on ultrasound imaging of breast cancer nodules is helpful for medical personnel to identify high-risk patients with lymph node metastasis in early stage, so as to facilitate early