Xiao Mengsu, Zhao Chenyang, Li Jianchu, Zhang Jing, Liu He, Wang Ming, Ouyang Yunshu, Zhang Yixiu, Jiang Yuxin, Zhu Qingli
Department of Ultrasound, Chinese Academy of Medical Sciences and Peking Union Medical College Hospital, Beijing, China.
Front Oncol. 2020 Jul 7;10:1070. doi: 10.3389/fonc.2020.01070. eCollection 2020.
The purpose of this study was to compare the diagnostic performance of breast lesions between deep learning-based computer-aided diagnosis (deep learning-based CAD) system and experienced radiologists and to compare the performance between symptomatic and asymptomatic patients. From January to December 2018, a total of 451 breast lesions in 389 consecutive patients were examined (mean age 46.86 ± 13.03 years, range 19-84 years) by both ultrasound and deep learning-based CAD system, all of which were biopsied, and the pathological results were obtained. The lesions were diagnosed by two experienced radiologists according to the fifth edition Breast Imaging Reporting and Data System (BI-RADS). The final deep learning-based CAD assessments were dichotomized as possibly benign or possibly malignant. The diagnostic performances of the radiologists and deep learning-based CAD were calculated and compared for asymptomatic patients and symptomatic patients. There were 206 asymptomatic screening patients with 235 lesions (mean age 45.06 ± 10.90 years, range 21-73 years) and 183 symptomatic patients with 216 lesions (mean age 50.03 ± 14.97 years, range 19-84 years). The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy and area under the receiver operating characteristic curve (AUC) of the deep learning-based CAD in asymptomatic patients were 93.8, 83.9, 75.0, 96.3, 87.2, and 0.89%, respectively. In asymptomatic patients, the specificity (83.9 vs. 66.5%, < 0.001), PPV (75.0 vs. 59.4%, = 0.013), accuracy (87.2 vs. 76.2%, = 0.002) and AUC (0.89 to 0.81, = 0.0013) of CAD were all significantly higher than those of the experienced radiologists. The sensitivity (93.8 vs. 80.0%), specificity (83.9 vs. 61.8%,), accuracy (87.2 vs. 73.6%) and AUC (0.89 vs. 0.71) of CAD were all higher for asymptomatic patients than for symptomatic patients. If the BI-RADS 4a lesions diagnosed by the radiologists in asymptomatic patients were downgraded to BI-RADS 3 according to the CAD, then 54.8% (23/42) of the lesions would avoid biopsy without missing the malignancy. The deep learning-based CAD system had better performance in asymptomatic patients than in symptomatic patients and could be a promising complementary tool to ultrasound for increasing diagnostic specificity and avoiding unnecessary biopsies in asymptomatic screening patients.
本研究的目的是比较基于深度学习的计算机辅助诊断（深度学习CAD）系统与经验丰富的放射科医生对乳腺病变的诊断性能，并比较有症状和无症状患者之间的性能。2018年1月至12月，对389例连续患者的451个乳腺病变（平均年龄46.86±13.03岁，范围19 - 84岁）进行了超声和基于深度学习的CAD系统检查，所有病变均进行了活检并获得了病理结果。两位经验丰富的放射科医生根据第五版乳腺影像报告和数据系统（BI-RADS）对病变进行诊断。基于深度学习的CAD最终评估结果分为可能良性或可能恶性。计算并比较了无症状患者和有症状患者中放射科医生和基于深度学习的CAD的诊断性能。有206例无症状筛查患者，共235个病变（平均年龄45.06±10.90岁，范围21 - 73岁），以及183例有症状患者，共216个病变（平均年龄50.03±14.97岁，范围19 - 84岁）。基于深度学习的CAD在无症状患者中的敏感性、特异性、阳性预测值（PPV）、阴性预测值（NPV）、准确性和受试者操作特征曲线下面积（AUC）分别为93.8%、83.9%、75.0%、96.3%、87.2%和0.89。在无症状患者中，CAD的特异性（83.9%对66.5%，<0.001）、PPV（75.0%对59.4%，=0.013）、准确性（87.2%对76.2%，=0.002）和AUC（0.89对0.81，=0.0013）均显著高于经验丰富的放射科医生。CAD在无症状患者中的敏感性（93.8%对80.0%）、特异性（83.9%对61.8%）、准确性（87.2%对73.6%）和AUC（0.89对0.71）均高于有症状患者。如果将放射科医生在无症状患者中诊断为BI-RADS 4a的病变根据CAD降级为BI-RADS 3，那么54.8%（23/42）的病变将避免活检且不会漏诊恶性肿瘤。基于深度学习的CAD系统在无症状患者中的表现优于有症状患者，并且可能成为超声的一个有前景的辅助工具，用于提高诊断特异性并避免无症状筛查患者进行不必要的活检。
