基于多模式磁共振检查特征的浸润性乳腺癌伴腋窝淋巴结转移风险预测模型构建OA

Objective:To construct the prediction model for axillary lymph node metastasis risk in patients with invasive breast cancer based on multimodal magnetic resonance imaging features and evaluate its prediction performance,aiming to provide evidence for early and accurate identification of patients requiring intraoperative lymph node biopsy and guidance for subsequent diagnosis and treatment planning.Methods:A retrospective study included 141 patients with invasive breast cancer confirmed by surgical pathology.Patients were grouped according to the presence of axillary lymph node metastasis after surgery.Univariate and multivariate analyses were conducted to identify independent risk factors for axillary lymph node metastasis,and a regression prediction model was constructed.Results:Univariate analysis revealed that vascular nerve invasion,Ki-67 expression level,lesion edge features,internal enhancement pattern,and apparent diffusion coefficient were associated with axillary lymph node metastasis(P<0.05).Multivariate Logistic regression analysis showed that early rapid enhancement,uneven or circular enhancement,and lower apparent diffusion coefficient were independent risk factors(P<0.05).ROC curve analysis demonstrated that the AUC of the regression model prediction probability was 0.90,higher than that of single indicators.Conclusion:There is a close correlation between multimodal MRI features,including early enhancement,internal enhancement,and apparent diffusion coefficient,and axillary lymph node metastasis in patients with invasive breast cancer.The prediction model based on these features exhibits superior performance in predicting axillary lymph node metastasis risk.