The Differential Diagnosis Of BI-RADS Classification Joint Real-time Shear-wave Elastography Imaging In Breast 3, 4 Class Nodules

Background and ObjectiveAround the world, the incidence and mortality of breast cancer are sustained, rapid growth, seriously threat to people’s health and life safety. Breast cancer is one of the most common malignancies of women. It has the highest incidence rates among malignant tumors of women. There are about 130 million people worldwide are diagnosed with breast cancer each year. It is also gradually becoming younger patient population. Breast cancer seriously harm to the global women’s health, therefore, it is crucial to early diagnosis and treatment of breast cancer. Ultrasonography has become a routine screening method currently in the differential diagnosis of breast nodules. In order to unify and standardize ultrasound breast nodules report describes, the American College of Radiology(American College of Radiology, ACR) published the Breast Imaging Reporting Data System(Breast Imaging-Reporting and Data System, BI-RADS) ultrasound part in 2003 to further guide treatment clinic. BI-RADS is divided into 0-6 categories, including category of 3 as possible benign(malignancy rate of <2%); category of 4 as suspicious abnormality, is divided into: 4a, 4b, 4c(malignancy may 2-95%). It is visible that category of 3 and 4 may span a large malignant nodules, thus the sonographer has some difficulties to accurately classify them. Shear wave elastography(SWE) technology can obtain quantitative breast tissue hardness. Based on BI-RADS classification, the diagnostic performance can be improved. Most existing research on its application for Q-Box was either cover the lesions or placed in the hardest possible coverage area, obtained benign and malignant diagnostic cutoffs and applied this cutoff value for the differential diagnosis of breast nodules. We innovatively use quantitative SWE measurement average of the Young’s modulus of the internal and surrounding area of breast nodules, and further combined with the BI-RADS classification category 3 and 4 diagnosis of breast nodules, in order to improve their diagnostic performance. Materials and MethodsPart I: collecting 100 patients with 126 breast nodules from October 2013 to April 2014 in our hospital ultrasound examination that has pathology results. All patients were female, BI-RADS classification was 3 or 4 classes. With pathological findings as "gold standard" draw ROC curve, use the area under the curve(AUC) to evaluate the diagnostic value of BI-RADS classification of breast nodules identified.Part II: 100 patients with 126 breast nodules underwent shear wave elastography real-time check, we need to place two Q-Box, the first possibly coverage(range on grayscale images nodules) all of the lesions, the second take on 1.37 times of gray-scale image area of the nodule, measuring the average value of Young’s modulus(Emean1, Emean2) inside and around the nodules. With pathological findings as "gold standard" draw ROC curve, use area under the curve(AUC) to evaluate the diagnostic value and get the best diagnostic cutoff value(C1, C2). Evaluate the consistency of intraobserver and interobserver with intraclass correlation coefficient(ICC).Part III: According to the best diagnostic cutoff value obtained by the part II of C1, C2, to conduct new classification of the included category 3,4 nodules: When Emean1> C1 or Emean2> C2, category 3 patients promoted to 4a, 4a promoted to 4b; when Emean1≤C1 and Emean2≤C2, category 4b dropped to 4a; other BI-RADS classification remain unchanged. BI-RADS category 3 and 4a classified as benign, 4b and 4c as malignant. With pathological findings as "gold standard" draw ROC curve, comparison new BI-RADS classification joint Emean1 Emean2 and original BI-RADS classification of the diagnostic value of breast nodules. Results1.Among the 126 breast nodules, 65 benign and 61 malignant. Area under the curve(AUC1) of BI-RADS classification is 0.883, sensitivity was 82.0%, specificity was 87.7%.2.AUC of Emean1 was 0.794, and compared with AUC = 0.5, P <0.05. The best diagnostic cutoff value for C1 was 38.10 k Pa; AUC of Emean2 was 0.891, and compared with AUC = 0.5, P <0.05. The best diagnostic cutoff value for C2 was 39.10 k Pa. Reliability of intraobserver and interobserver repeated measures analysis of consistency: ICC values were 0.95, 0.92, 0.88, 0.79, respectively, 95% confidence interval: 0.89 ~ 0.97, 0.85 ~ 0.96, 0.77 ~ 0.94, 0.62 ~ 0.89(P <0.01), indicating that SWE has very good repeatability intra observer and observer.3.Area under the curve of the new BI-RADS classification that joint Emean1 and Emean2(AUC2) was 0.952, sensitivity was 93.4%, specificity was 95.4%. The difference between AUC1 and AUC2 was statistically significant by Z test(P <0.05). ConclusionBI-RADS classification has a higher value in the differential diagnosis of breast nodules. It can guide clinical practice well.The average value of Young’s modulus by Shear wave elastography inside and around breast nodules can guide the differential diagnosis of breast nodules, and the intraobserver and interobserver reproducibility are very good.BI-RADS classification joint the average value of Young’s modulus inside and around nodules by shear wave elastography as a new method has a higher value for the differential diagnosis of breast category 3 and 4 nodules.