Rationale and objectives: Several studies have evaluated the usefulness of deep learning for lung segmentation using chest x-ray (CXR) images with small- or medium-sized abnormal findings. Here, we built a database including both CXR images with severe abnormalities and experts' lung segmentation results, and aimed to evaluate our network's efficacy in lung segmentation from these images. Materials and Methods: For lung segmentation, CXR images from the Japanese Society of Radiological Technology (JSRT, N = 247) and Montgomery databases (N = 138), were included, and 65 additional images depicting severe abnormalities from a public database were evaluated and annotated by a radiologist, thereby adding lung segmentation results to these images. Baseline U-net was used to segment the lungs in images from the three databases. Subsequently, the U-net network architecture was automatically optimized for lung segmentation from CXR images using Bayesian optimization. Dice similarity coefficient (DSC) was calculated to confirm segmentation. Results: Our results demonstrated that using baseline U-net yielded poorer lung segmentation results in our database than those in the JSRT and Montgomery databases, implying that robust segmentation of lungs may be difficult because of severe abnormalities. The DSC values with baseline U-net for the JSRT, Montgomery and our databases were 0.979, 0.941, and 0.889, respectively, and with optimized U-net, 0.976, 0.973, and 0.932, respectively. Conclusion: For robust lung segmentation, the U-net architecture was optimized via Bayesian optimization, and our results demonstrate that the optimized U-net was more robust than baseline U-net in lung segmentation from CXR images with large-sized abnormalities.

中文翻译：

---

使用深度学习对有严重异常发现的患者的胸部 X 线图像进行肺分割

基本原理和目标：多项研究评估了深度学习对使用具有中小型异常发现的胸部 X 射线 (CXR) 图像进行肺分割的有用性。在这里，我们建立了一个数据库，包括严重异常的 CXR 图像和专家的肺分割结果，旨在评估我们的网络从这些图像中进行肺分割的有效性。材料和方法：对于肺分割，包括来自日本放射技术学会（JSRT，N = 247）和蒙哥马利数据库（N = 138）的 CXR 图像，并评估了来自公共数据库的 65 幅描述严重异常的附加图像，并由放射科医生注释，从而将肺分割结果添加到这些图像中。基线 U-net 用于分割来自三个数据库的图像中的肺。随后，U-net 网络架构使用贝叶斯优化从 CXR 图像自动优化肺分割。计算骰子相似系数（DSC）以确认分割。结果：我们的结果表明，在我们的数据库中使用基线 U-net 产生的肺分割结果比 JSRT 和 Montgomery 数据库中的要差，这意味着由于严重的异常，肺的稳健分割可能很困难。JSRT、Montgomery 和我们的数据库的基线 U-net DSC 值分别为 0.979、0.941 和 0.889，优化 U-net 的 DSC 值分别为 0.976、0.973 和 0.932。结论：对于稳健的肺分割，U-net 架构通过贝叶斯优化进行了优化，