Zhang Xianyu, Li Hui, Wang Chaoyun, Cheng Wen, Zhu Yuntao, Li Dapeng, Jing Hui, Li Shu, Hou Jiahui, Li Jiaying, Li Yingpu, Zhao Yashuang, Mo Hongwei, Pang Da
Department of Breast Surgery, Harbin Medical University Cancer Hospital, Harbin, China.
Harbin Engineering University Automation College, Harbin, China.
Front Oncol. 2021 Mar 5;11:623506. doi: 10.3389/fonc.2021.623506. eCollection 2021.
Breast ultrasound is the first choice for breast tumor diagnosis in China, but the Breast Imaging Reporting and Data System (BI-RADS) categorization routinely used in the clinic often leads to unnecessary biopsy. Radiologists have no ability to predict molecular subtypes with important pathological information that can guide clinical treatment. This retrospective study collected breast ultrasound images from two hospitals and formed training, test and external test sets after strict selection, which included 2,822, 707, and 210 ultrasound images, respectively. An optimized deep learning model (DLM) was constructed with the training set, and the performance was verified in both the test set and the external test set. Diagnostic results were compared with the BI-RADS categorization determined by radiologists. We divided breast cancer into different molecular subtypes according to hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) expression. The ability to predict molecular subtypes using the DLM was confirmed in the test set. In the test set, with pathological results as the gold standard, the accuracy, sensitivity and specificity were 85.6, 98.7, and 63.1%, respectively, according to the BI-RADS categorization. The same set achieved an accuracy, sensitivity, and specificity of 89.7, 91.3, and 86.9%, respectively, when using the DLM. For the test set, the area under the curve (AUC) was 0.96. For the external test set, the AUC was 0.90. The diagnostic accuracy was 92.86% with the DLM in BI-RADS 4a patients. Approximately 70.76% of the cases were judged as benign tumors. Unnecessary biopsy was theoretically reduced by 67.86%. However, the false negative rate was 10.4%. A good prediction effect was shown for the molecular subtypes of breast cancer with the DLM. The AUC were 0.864, 0.811, and 0.837 for the triple-negative subtype, HER2 (+) subtype and HR (+) subtype predictions, respectively. This study showed that the DLM was highly accurate in recognizing breast tumors from ultrasound images. Thus, the DLM can greatly reduce the incidence of unnecessary biopsy, especially for patients with BI-RADS 4a. In addition, the predictive ability of this model for molecular subtypes was satisfactory,which has specific clinical application value.
乳腺超声是我国乳腺肿瘤诊断的首选方法，但临床常规使用的乳腺影像报告和数据系统（BI-RADS）分类常常导致不必要的活检。放射科医生无法通过重要的病理信息预测分子亚型，而这些信息可以指导临床治疗。本回顾性研究收集了两家医院的乳腺超声图像，经过严格筛选后形成训练集、测试集和外部测试集，分别包含2822、707和210幅超声图像。使用训练集构建了优化的深度学习模型（DLM），并在测试集和外部测试集中验证其性能。将诊断结果与放射科医生确定的BI-RADS分类进行比较。我们根据激素受体（HR）和人表皮生长因子受体2（HER2）表达将乳腺癌分为不同的分子亚型。在测试集中证实了使用DLM预测分子亚型的能力。在测试集中，以病理结果为金标准，根据BI-RADS分类，准确率、敏感性和特异性分别为85.6%、98.7%和63.1%。使用DLM时，同一组的准确率、敏感性和特异性分别达到89.7%、91.3%和86.9%。对于测试集，曲线下面积（AUC）为0.96。对于外部测试集（AUC）为0.90。在BI-RADS 4a患者中，DLM的诊断准确率为92.86%。约70.76%的病例被判定为良性肿瘤。理论上不必要活检的发生率降低了67.86%。然而，假阴性率为10.4%。DLM对乳腺癌分子亚型显示出良好的预测效果。三阴型、HER2（+）型和HR（+）型预测的AUC分别为0.864、0.811和0.837。本研究表明，DLM在从超声图像识别乳腺肿瘤方面具有很高的准确性。因此，DLM可以大大降低不必要活检的发生率，尤其是对于BI-RADS 4a患者。此外，该模型对分子亚型的预测能力令人满意，具有特定的临床应用价值。
