目的： 分析超微血管成像（superb microvascular imaging,SMI）、超声弹性成像（ultrasounic elastography,UE）联合高频超声检查在微小乳腺癌诊断中的价值,并筛选相关的高危超声特征。方法： 选取乳腺影像报告和数据系统（Breast Imaging Reporting and Data System,BI-RADS）分类为4类的乳腺微小病灶（最大径≤1 cm）87个,病灶来自85例患者。根据术后病理结果将其分为良性组和恶性组,分析2组间病灶的灰阶超声特征、彩色多普勒血流图(color Doppler flow imaging,CDFI）特征、SMI分级及弹性评分,筛选微小乳腺癌的高危超声特征。结果： 单因素分析显示,乳腺微小病灶良性组与恶性组间的病灶形态、边缘、钙化情况、弹性评分、SMI及CDFI分级方面差异有统计学意义（P<0.05）;多因素Logistic回归分析显示,病灶边缘不光整、弹性评分≥4分及SMI的Adler分级≥Ⅱ级是预测微小乳腺癌的独立影响因素,而三者联合诊断微小乳腺癌准确率最高（81.61%）。结论： SMI检查显示Adler分级≥Ⅱ级及弹性评分≥4分是诊断微小乳腺癌的高危超声特征,两者与病灶边缘不光整的超声特征联合时诊断的效能较高,可为临床提供重要的影像学依据。

Objective: To analyze the diagnostic value of superb microvascular imaging (SMI) in combination with ultrasonic elastography (UE) and high-frequency ultrasonography on micro breast cancer and screening for high-risk features on ultrasound. Methods: A total of 87 breast lesions from 85 patients were collected. The maximum diameters of the lesions were no more than 1 cm,and all lesions were classified as BI-RADS （Breast Imaging Reporting and Data System） 4. The lesions were grouped into benign and malignant groups based on results of postoperative pathology. The characteristics of grey-scale ultrasound and color Doppler flow imaging(CDFI), SMI and elasticity scores of the lesions were compared between 2 groups and high-risk features on ultrasound for micro breast cancer were screened. Results: Univariate analysis showed that lesion morphology, margin, calcification status, elasticity score, SMI and CDFI classification were statistically different between benign and malignant groups (P<0.05). Multivariate Logistic regression analysis revealed that lesions with ill-defined margin, elasticity score ≥4 or Adler scale of SMI ≥Ⅱ were independently predicting factors for micro breast cancer. The accuracy of combining the three indices for diagnosing micro breast cancer was 81.61%, higher than any single one. Conclusions: High-risk features of micro breast cancer on ultrasound include Adler scale SMI≥Ⅱ and elasticity score ≥4. Both features in combination with ill-defined margin show good diagnostic efficiency for micro breast cancer, which may provide an important reference for clinical diagnosis.