Abstract:

Objective: To study the efficacy of smart three-dimensional superb microvascular imaging (Smart 3D SMI) on displaying blood flow in breast lesions of breast imaging reporting and data system(BI-RADS）4, and analyze the diagnostic value of smart 3D SMI in BI-RADS 4 breast lesions. Methods: A total of 120 breast lesions of BI-RADS 4 confirmed by post-operative pathology were enrolled, and the lesions were divided into benign group (32 lesions）and malignant one (88 lesions). The power Doppler flow imaging (PDFI), advanced dramatic flow imaging (ADFI） and Smart 3D SMI were performed to detect blood flow in the lesions. Referring to the semi quantitative grading method, the blood flow levels in the lesions were divided into two groups: the group lack of blood flow and the group rich of blood flow. The difference in blood flow displayed by PDFI, ADFI and Smart 3D SMI was compared between the benign and malignant breast lesions. Furthermore, the results of Smart 3D SMI were used to correct the BI-RADS classification. The area under receiver operator characteristic curve(ROC curve) for performance of corrected BI-RADS to diagnose malignant breast lesions was compared before and after correction. Results: ALL these three methods showed that the blood flow in malignant lesions were more abundant than that in benign group (P<0.001, P<0.001, P=0.004). The distribution of blood flow levels detected by different detection methods was different. The proportion of patients with abundant blood flow detected by Smart 3D SMI method was 75.0% (benign 18 cases, malignant 72 cases), while it was 50.0% (5 cases of benign and 55 cases of malignant) by PDFI and 56.7% (7 cases of benign and 61 cases of malignant) by ADFI, respectively. There was no significant difference in distribution of blood flow levels between PDFI and ADFI (P=0.301). The percentage of patients with abundant blood flow detected by Smart 3D SMI was significantly higher than that by PDFI (50.0%, P<0.001) and ADFI (56.7%, P=0.004). Corrected with Smart 3D SMI, the BI-RADS classification of 52 cases with malignant lesions (59.1%) up-graded and 14 cases with benign lesion (43.8%) up-graded. The area under ROC curve of corrected BI-RADS for diagnosing malignant lesions was 0.86, which was higher than that of BI-RADS without correction (area under curve 0.80) (P=0.011). Conclusions: The Smart 3D SMI is a new method and have more advantage than traditional methods in detecting blood flow of BI-RADS 4 breast lesions, especially in the diagnosis of malignant tumors.

Key words: Smart three-dimensional superb microvascular imaging, Breast imaging reporting and data system4, Breast lesion