Application value of predictive models of contrast-enhanced ultrasound in the evaluation of breast imaging reporting and data system 4 breast lesions

Abstract

Objective To build breast contrast-enhanced ultrasound (CEUS) predictive models of benign and malignant lesions, and evaluate its value in malignant risk assessment. Methods Breast CEUS predictive models of benign and malignant lesions was build depending on the analysis of published papers, conclusion of the previous study results and the clinical experience. With final pathology results as the gold standard, the predictive value was evaluated in totally 235 nodules which were classified as breast imaging reporting and data system (BI-RADS) 4. Results There were 139 benign nodules (59.2%) and 96 malignant nodules (40.8%) in this study. Diagnostic sensitivity, specificity and accuracy of three malignant CEUS models and three benign CEUS models were 87.01%, 86.92% and 86.99%, respectively. Conclusions The contrast-enhanced patterns of benign and malignant breast tumors are different. The breast CEUS models can predict malignant risk of breast lesions more accurately, may lead to the decrease of false-positive biopsy and better BI-RADS used nowadays. Key words: Ultrasonography; Contrast media; Breast diseases