Department of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43 Gil, Songpa-gu, Seoul, 138-736, South Korea.
Department of Radiology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 300 Gumi-dong, Bundang-gu, Seongnam-si, Gyeonggi-do, 13620, South Korea.
Eur Radiol. 2020 Mar;30(3):1359-1368. doi: 10.1007/s00330-019-06532-x. Epub 2019 Nov 20.
To investigate the feasibility of a deep learning-based detection (DLD) system for multiclass lesions on chest radiograph, in comparison with observers.
A total of 15,809 chest radiographs were collected from two tertiary hospitals (7204 normal and 8605 abnormal with nodule/mass, interstitial opacity, pleural effusion, or pneumothorax). Except for the test set (100 normal and 100 abnormal (nodule/mass, 70; interstitial opacity, 10; pleural effusion, 10; pneumothorax, 10)), radiographs were used to develop a DLD system for detecting multiclass lesions. The diagnostic performance of the developed model and that of nine observers with varying experiences were evaluated and compared using area under the receiver operating characteristic curve (AUROC), on a per-image basis, and jackknife alternative free-response receiver operating characteristic figure of merit (FOM) on a per-lesion basis. The false-positive fraction was also calculated.
Compared with the group-averaged observations, the DLD system demonstrated significantly higher performances on image-wise normal/abnormal classification and lesion-wise detection with pattern classification (AUROC, 0.985 vs. 0.958; p = 0.001; FOM, 0.962 vs. 0.886; p < 0.001). In lesion-wise detection, the DLD system outperformed all nine observers. In the subgroup analysis, the DLD system exhibited consistently better performance for both nodule/mass (FOM, 0.913 vs. 0.847; p < 0.001) and the other three abnormal classes (FOM, 0.995 vs. 0.843; p < 0.001). The false-positive fraction of all abnormalities was 0.11 for the DLD system and 0.19 for the observers.
The DLD system showed the potential for detection of lesions and pattern classification on chest radiographs, performing normal/abnormal classifications and achieving high diagnostic performance.
• The DLD system was feasible for detection with pattern classification of multiclass lesions on chest radiograph. • The DLD system had high performance of image-wise classification as normal or abnormal chest radiographs (AUROC, 0.985) and showed especially high specificity (99.0%). • In lesion-wise detection of multiclass lesions, the DLD system outperformed all 9 observers (FOM, 0.962 vs. 0.886; p < 0.001).
与观察者相比，探究基于深度学习的检测（DLD）系统用于胸部 X 线片中多类病变的可行性。
从两家三级医院共收集了 15809 张胸部 X 光片（7204 张正常和 8605 张异常，包括结节/肿块、间质混浊、胸腔积液或气胸）。除了测试集（100 张正常和 100 张异常（结节/肿块，70 张；间质混浊，10 张；胸腔积液，10 张；气胸，10 张））外，还使用 DLD 系统来检测多类病变，开发了一种用于检测多类病变的 DLD 系统。使用基于图像的接收者操作特征曲线下面积（AUROC）和基于病变的替代无响应接收者操作特征的质量指标（FOM）评估和比较了开发的模型和 9 名具有不同经验的观察者的诊断性能。还计算了假阳性分数。
与组平均观察结果相比，DLD 系统在基于图像的正常/异常分类和基于模式分类的病变检测方面表现出显著更高的性能（AUROC，0.985 与 0.958；p = 0.001；FOM，0.962 与 0.886；p < 0.001）。在病变检测方面，DLD 系统优于所有 9 名观察者。在亚组分析中，DLD 系统对结节/肿块（FOM，0.913 与 0.847；p < 0.001）和其他三类异常（FOM，0.995 与 0.843；p < 0.001）的性能均有显著提高。DLD 系统检测所有异常的假阳性率为 0.11，而观察者的假阳性率为 0.19。
DLD 系统显示出在胸部 X 光片上检测病变和模式分类的潜力，能够进行正常/异常分类，并达到较高的诊断性能。
• DLD 系统可用于胸部 X 光片中多类病变的模式分类检测。• DLD 系统在作为正常或异常胸部 X 光片的图像分类方面表现出色（AUROC，0.985），且特异性很高（99.0%）。• 在多类病变的病变检测方面，DLD 系统优于所有 9 名观察者（FOM，0.962 与 0.886；p < 0.001）。
