Department of Breast Surgery, National Hospital Organization Sendai Medical Center, 2-11-12 Miyaginohara, Miyagino-ku, Sendai, Miyagi, 983-8520, Japan.
Division of Biostatistics, Tohoku University Graduate School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi, 980-8574, Japan.
J Med Ultrason (2001). 2021 Jan;48(1):71-81. doi: 10.1007/s10396-020-01072-0. Epub 2021 Jan 3.
Breast ultrasound has been widely used as an essential examination for diagnosing breast cancer. However, standardized diagnostic criteria are as yet lacking. This study aimed to develop a simple diagnostic flowchart for beginners learning breast ultrasonography. The diagnostic flowchart was developed based on the recall criteria widely used in Japan.
We conducted a multicenter study to examine recall criteria usefulness in the diagnostic phase of breast disease. Women with ultrasound-visible breast masses who underwent B-mode breast ultrasound examination were recruited from 22 hospitals in Japan between September 2009 and January 2010. B-mode images were evaluated by members of the centralized image interpretation committee. We developed the new diagnostic flowchart based on the results. The usefulness of the diagnostic flowchart was assessed by employing datasets from the current study and another study which we conducted (BC-04 study).
We evaluated 1045 solid masses (malignant: 495, benign: 550). Multivariate analysis showed that shape, margin, echogenic halo, interruption of the mammary gland interface, and depth width ratio were significant findings for distinguishing between benign and malignant masses. We modified the recall criteria and developed our novel diagnostic flowchart using these findings. The sensitivity and specificity of the new flowchart (current study: 0.97, 0.45; BC-04 study dataset: 0.95, 0.45) were similar to those of experts (current study: 0.96, 0.54; BC-04 study dataset: 0.98, 0.38).
We developed a simple diagnostic flowchart for breast ultrasound. This flowchart is anticipated to be applicable to educating beginners learning breast ultrasound.
乳腺超声已广泛应用于乳腺癌的诊断。然而，目前尚缺乏标准化的诊断标准。本研究旨在为初学者学习乳腺超声检查制定一个简单的诊断流程图。该诊断流程图是基于在日本广泛使用的召回标准制定的。
我们进行了一项多中心研究，以检验在乳腺疾病的诊断阶段召回标准的实用性。本研究纳入了 2009 年 9 月至 2010 年 1 月期间，来自日本 22 家医院的经超声可见乳腺肿块的女性。由集中的图像解读委员会成员评估 B 型乳腺超声图像。我们根据研究结果制定了新的诊断流程图。我们通过使用当前研究和我们进行的另一项研究（BC-04 研究）的数据集评估了诊断流程图的实用性。
我们评估了 1045 个实性肿块（恶性：495 个，良性：550 个）。多变量分析显示，形状、边界、回声晕、乳腺界面中断和深度宽度比是区分良性和恶性肿块的重要发现。我们修改了召回标准，并使用这些发现制定了我们的新诊断流程图。新流程图的敏感性和特异性（当前研究：0.97，0.45；BC-04 研究数据集：0.95，0.45）与专家相似（当前研究：0.96，0.54；BC-04 研究数据集：0.98，0.38）。
我们为乳腺超声制定了一个简单的诊断流程图。该流程图有望应用于乳腺超声初学者的教育。
