自动全乳腺超声与常规手持超声对乳腺肿块的诊断价值研究

Research on the Diagnostic Value of Automated Breast Ultrasound and Hand-Held Ultrasound for Breast Masses

摘要: 目的：比较自动全乳腺超声(automated breast ultrasound, ABUS)与常规手持超声(hand-held ultrasound, HHUS)对乳腺肿块的诊断效能，探讨ABUS与HHUS在乳腺肿块诊断中的应用价值。方法：回顾性收集2023年9月至2024年10月因乳腺疾病来延安大学附属医院甲乳外科门诊就诊患者的临床资料以及超声影像资料，以病理结果为金标准，分别计算HHUS、ABUS及二者联合诊断的灵敏度、特异度、准确度、阳性预测值(positive predictive value, PPV)、阴性预测值(negative predictive value, NPV)，比较三种方法对乳腺肿块的诊断效能，并采用二分类多因素Logistic回归模型筛选诊断乳腺良恶性病变的独立因子。结果：HHUS与ABUS共同检出的170个病灶中有142个进行了穿刺活检或手术切除活检，其中恶性67个，良性75个。HHUS、ABUS及二者联合对诊断乳腺病灶的灵敏度、特异度、准确度、PPV、NPV分别为89.55% vs 88.06% vs 92.53%、90.67% vs 97.33% vs 90.67%、90.14% vs 92.96% vs 91.57%、89.55% vs 96.72% vs 89.86%、90.67% vs 90.12% vs 93.15%，差异无统计学意义(P > 0.05)。乳腺良恶性病变ABUS冠状面超声征象在病灶形态、边缘、方位、微钙化、低回声晕、高回声晕、汇聚征方面差异具有统计学意义(P < 0.05)，在回声、粗大钙化、跳跃征方面差异无统计学意义(P > 0.05)。二分类多因素Logistic回归模型显示低回声晕、汇聚征为诊断乳腺良恶性病变的独立因子。结论：ABUS与HHUS对乳腺肿块的诊断效能无明显差异，二者联合也不能提高诊断能力，但ABUS特有的冠状面视角能够反映出乳腺病灶与周围正常组织的关系，在乳腺病灶的良恶性鉴别诊断中具有重要意义，其中汇聚征与低回声均可作为ABUS评价乳腺良恶性病变的独立因子。

Abstract: Objective: To compare the diagnostic efficacy of automated breast ultrasound (ABUS) and conventional hand-held ultrasound (HHUS) for breast masses, and to explore the application value of ABUS and HHUS in the diagnosis of breast masses. Method: Clinical data and ultrasound imaging data of patients who visited the Department of Breast Surgery at Yan’an University Affiliated Hospital for breast diseases from September 2023 to October 2024 were retrospectively collected. Pathological results were used as the gold standard to calculate the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of HHUS, ABUS, and their combined diagnosis. The diagnostic efficacy of the three methods for breast masses was compared, and a binary multivariate logistic regression model was used to screen for independent factors for the diagnosis of benign and malignant breast lesions. Result: Among the 170 lesions detected by HHUS and ABUS, 142 underwent biopsy or surgical resection, including 67 malignant and 75 benign lesions. The sensitivity, specificity, accuracy, PPV, and NPV of HHUS, ABUS, and their combination in diagnosing breast lesions were 89.55% vs 88.06% vs 92.53%, 90.67% vs 97.33% vs 90.67%, 90.14% vs 92.96% vs 91.57%, 89.55% vs 96.72% vs 89.86%, 90.67% vs 90.12% vs 93.15%, respectively. The difference was not statistically significant (P > 0.05). The ABUS coronal ultrasound features of benign and malignant breast lesions showed statistically significant differences in lesion morphology, edge, orientation, microcalcification, low echo halo, high echo halo, and convergence sign (P < 0.05), while there was no statistically significant difference in echo, coarse calcification, and jumping sign (P > 0.05). The binary multiple factor logistic regression model shows that low echo halo and convergence sign are independent factors for diagnosing benign and malignant breast lesions. Conclusion: There is no significant difference in the diagnostic efficacy of ABUS and HHUS for breast masses, and the combination of the two cannot improve diagnostic ability. However, the unique coronal view of ABUS can reflect the relationship between breast lesions and surrounding normal tissues, which is of great significance in the differential diagnosis of benign and malignant breast lesions. Convergence sign and hypoechoic halo can be used as independent factors for ABUS evaluation of breast benign and malignant lesions.