We aimed to evaluate the performance of convolutional neural networks (CNNs) in the classification of coronavirus disease 2019 (COVID-19) disease using normal, pneumonia, and COVID-19 chest radiographs (CXRs). First, we collected 9194 CXRs from open datasets and 58 from the Korea University Anam Hospital (KUAH). The number of normal, pneumonia, and COVID-19 CXRs were 4580, 3884, and 730, respectively. The CXRs obtained from the open dataset were randomly assigned to the training, tuning, and test sets in a 70:10:20 ratio. For external validation, the KUAH (20 normal, 20 pneumonia, and 18 COVID-19) dataset, verified by radiologists using computed tomography, was used. Subsequently, transfer learning was conducted using DenseNet169, InceptionResNetV2, and Xception to identify COVID-19 using open datasets (internal) and the KUAH dataset (external) with histogram matching. Gradient-weighted class activation mapping was used for the visualization of abnormal patterns in CXRs. The average AUC and accuracy of the multiscale and mixed-COVID-19Net using three CNNs over five folds were (0.99 ± 0.01 and 92.94% ± 0.45%), (0.99 ± 0.01 and 93.12% ± 0.23%), and (0.99 ± 0.01 and 93.57% ± 0.29%), respectively, using the open datasets (internal). Furthermore, these values were (0.75 and 74.14%), (0.72 and 68.97%), and (0.77 and 68.97%), respectively, for the best model among the fivefold cross-validation with the KUAH dataset (external) using domain adaptation. The various state-of-the-art models trained on open datasets show satisfactory performance for clinical interpretation. Furthermore, the domain adaptation for external datasets was found to be important for detecting COVID-19 as well as other diseases.

中文翻译：

---

深度卷积神经网络在胸片上区分 COVID-19 和其他肺部异常：使用内部和外部数据集进行评估

我们旨在使用正常、肺炎和 COVID-19 胸片 (CXR) 评估卷积神经网络 (CNN) 在 2019 年冠状病毒病 (COVID-19) 疾病分类中的性能。首先，我们从开放数据集中收集了 9194 个 CXR，从韩国大学安南医院 (KUAH) 收集了 58 个。正常、肺炎和 COVID-19 CXR 的数量分别为 4580、3884 和 730。从开放数据集中获得的 CXR 以 70:10:20 的比例随机分配到训练、调整和测试集。对于外部验证，使用了由放射科医生使用计算机断层扫描验证的 KUAH（20 例正常、20 例肺炎和 18 例 COVID-19）数据集。随后，使用 DenseNet169、InceptionResNetV2、和 Xception 使用开放数据集（内部）和带有直方图匹配的 KUAH 数据集（外部）识别 COVID-19。梯度加权类激活映射用于可视化 CXR 中的异常模式。使用三个 CNN 超过五倍的多尺度和混合 COVID-19Net 的平均 AUC 和准确度分别为（0.99 ± 0.01 和 92.94% ± 0.45%）、（0.99 ± 0.01 和 93.12% ± 0.23%）和（0.99 ± 0.01）和 93.57% ± 0.29%），分别使用开放数据集（内部）。此外，这些值分别为（0.75 和 74.14%）、（0.72 和 68.97%）和（0.77 和 68.97%），这是使用域适应与 KUAH 数据集（外部）进行五重交叉验证中的最佳模型。在开放数据集上训练的各种最先进的模型显示出令人满意的临床解释性能。此外，