摘要
Abstract
Objective To investigate the feasibility and clinical application value of further subclassifying MRI Breast Imaging Reporting and Data System(BI-RADS)category 4 lesions into 4A,4B,and 4C,and to compare its diagnostic performance with ultrasound(US)and mammography(MG).Methods A total of 220 patients with breast lesions were retrospectively included.All patients underwent US,MG,and breast MRI examinations and obtained pathological results by surgery or biopsy.A total of 225 lesions were classified as MRI BI-RADS category 4.According to a series of reports from the Society of Breast Imaging(SBI)and the Radiological Society of North America(RSNA),the lesions were further subclassified into MRI BI-RADS 4A,4B,and 4C.Using pathological results as the gold standard,the malignancy rates of each subcategory were calculated and compared.Receiver operating characteristic(ROC)curve analysis was used to evaluate the diagnostic performance of US BI-RADS,MG BI-RADS,and MRI BI-RADS classifications for breast lesions,and the area under the curve(AUC),accuracy,sensitivity,and specificity were calculated.Differences in AUC values were compared using the DeLong test.Results There were 42,53,and 130 lesions classified as MRI BI-RADS 4A,4B,and 4C,respectively,among which 5,41,and 127 lesions were pathologically diagnosed as malignant.The corresponding malignancy rates were 11.9%,77.4%,and 97.7%,respectively.Pairwise comparisons among the subcategories showed statistically significant differences(P<0.016).The malignancy rate was lowest in category 4A and highest in category 4C.MRI BI-RADS classification had the highest accuracy and sensitivity for differentiating benign from malignant lesions(91.11%and 97.11%,respectively),whereas MG BI-RADS classification had the highest specificity(86.54%).The AUC values of MRI BI-RADS,US BI-RADS,and MG BI-RADS were 0.910(95%CI:0.867-0.953),0.871(95%CI:0.824-0.917),and 0.806(95%CI:0.748-0.863),respectively.MRI BI-RADSshowed higher diagnostic performance than MG BI-RADS(Z=3.407,P=0.001),while its performance was comparable to that ofUS BI-RADS(Z=1.402,P=0.161).Conclusion Subcategorization of MRI BI-RADS category 4 lesions is feasible and has important clinical application value.It can provide useful diagnostic evidence for differentiating benign from malignant breast lesions classified as BI-RADS category 4.关键词
磁共振成像/乳腺病变/乳腺影像报告与数据系统分类/亚分类Key words
Magnetic resonance imaging/Breast lesion/Breast imaging reporting and data system category/Subcategorization分类
医药卫生
