A Predictive Model for the Malignancy of Breast Tumors Based on Ultrasound Elastography and MRI Features was Constructed

LIAO Huixian; WEI Lingfeng; WEI Taoyi; LU Xiaoxiong; LU Songbai; NONG Yulin; YANG Huining

doi:10.16458/j.cnki.1007-0893.2025.11.004

Abstract：Objective To explore the construction and validation of a nomogram model for predicting the malignancy of breast tumors based on ultrasound elastography(UE)and magnetic resonance imaging(MRI)features.Methods A total of 171 female patients who underwent ultrasound examination and were found to have solid breast masses at Wuming Hospital Affiliated to Guangxi Medical University since 2021 were selected as the research subjects.According to the postoperative pathological results,the patients were divided into the malignant group(89 cases)and the benign group(82 cases).General data,ultrasound and MRI imaging data of the two groups were collected and compared.Multivariate analysis was used to identify the independent risk factors for breast tumor malignancy,and a visual nomogram model was established.The receiver operating characteristic(ROC)curve and calibration curve were drawn to verify the performance of the model.Results Multivariate analysis showed that advanced age,ultrasound BI-RADS classification≥4b,ultrasound elastography score>3,MRI BI-RADS classification≥4b,combined ultrasound and MRI diagnosis classification≥4b,and increased ultrasound elastography strain ratio were independent risk factors for breast tumor malignancy,and the differences were statistically significant(P<0.05).The nomogram model predicted the probability of breast tumor malignancy ranging from 0.1 to 0.9.The area under the ROC curve(AUC)for predicting breast tumor malignancy was 0.883.The calibration curve and Hosmer-Lemeshow test showed that the predicted values of the nomogram were highly consistent with the actual observed values,with an average absolute error of 0.032 and good calibration.Conclusion Advanced age,ultrasound BI-RADS classification≥4b,ultrasound elastography score>3,MRI BI-RADS classification≥4b,combined ultrasound and MRI diagnosis≥4b,and increased ultrasound elastography strain ratio are independent risk factors for breast tumor malignancy.The nomogram model established based on these factors shows good discrimination and calibration performance.