Department of Radiology, Hanyang University Medical Center, Hanyang University College of Medicine, Seoul, Korea.
Department of Radiology, Seoul National University Hospital, Seoul National College of Medicine, Seoul, Korea.
Korean J Radiol. 2021 Mar;22(3):476-488. doi: 10.3348/kjr.2020.0318. Epub 2020 Oct 30.
We aimed to develop a deep neural network for segmenting lung parenchyma with extensive pathological conditions on non-contrast chest computed tomography (CT) images.
Thin-section non-contrast chest CT images from 203 patients (115 males, 88 females; age range, 31-89 years) between January 2017 and May 2017 were included in the study, of which 150 cases had extensive lung parenchymal disease involving more than 40% of the parenchymal area. Parenchymal diseases included interstitial lung disease (ILD), emphysema, nontuberculous mycobacterial lung disease, tuberculous destroyed lung, pneumonia, lung cancer, and other diseases. Five experienced radiologists manually drew the margin of the lungs, slice by slice, on CT images. The dataset used to develop the network consisted of 157 cases for training, 20 cases for development, and 26 cases for internal validation. Two-dimensional (2D) U-Net and three-dimensional (3D) U-Net models were used for the task. The network was trained to segment the lung parenchyma as a whole and segment the right and left lung separately. The University Hospitals of Geneva ILD dataset, which contained high-resolution CT images of ILD, was used for external validation.
The Dice similarity coefficients for internal validation were 99.6 ± 0.3% (2D U-Net whole lung model), 99.5 ± 0.3% (2D U-Net separate lung model), 99.4 ± 0.5% (3D U-Net whole lung model), and 99.4 ± 0.5% (3D U-Net separate lung model). The Dice similarity coefficients for the external validation dataset were 98.4 ± 1.0% (2D U-Net whole lung model) and 98.4 ± 1.0% (2D U-Net separate lung model). In 31 cases, where the extent of ILD was larger than 75% of the lung parenchymal area, the Dice similarity coefficients were 97.9 ± 1.3% (2D U-Net whole lung model) and 98.0 ± 1.2% (2D U-Net separate lung model).
The deep neural network achieved excellent performance in automatically delineating the boundaries of lung parenchyma with extensive pathological conditions on non-contrast chest CT images.
我们旨在开发一种深度神经网络，用于对非对比胸部 CT 图像上具有广泛病理条件的肺实质进行分割。
本研究纳入了 203 例患者（男 115 例，女 88 例；年龄 31-89 岁）的薄层非对比胸部 CT 图像，其中 150 例患者的肺实质疾病广泛，累及实质区域的 40%以上。肺实质疾病包括间质性肺疾病（ILD）、肺气肿、非结核分枝杆菌肺病、结核破坏肺、肺炎、肺癌和其他疾病。5 名经验丰富的放射科医生在 CT 图像上逐片绘制肺的边界。用于开发网络的数据集包括 157 例用于训练，20 例用于开发，26 例用于内部验证。使用二维（2D）U-Net 和三维（3D）U-Net 模型进行任务。该网络经过训练可整体分割肺实质，并分别分割右肺和左肺。使用含有间质性肺疾病高分辨率 CT 图像的日内瓦大学医院ILD 数据集进行外部验证。
内部验证的 Dice 相似系数分别为 99.6±0.3%（2D U-Net 整体肺模型）、99.5±0.3%（2D U-Net 单独肺模型）、99.4±0.5%（3D U-Net 整体肺模型）和 99.4±0.5%（3D U-Net 单独肺模型）。外部验证数据集的 Dice 相似系数分别为 98.4±1.0%（2D U-Net 整体肺模型）和 98.4±1.0%（2D U-Net 单独肺模型）。在 31 例ILD 程度超过肺实质面积 75%的病例中，Dice 相似系数分别为 97.9±1.3%（2D U-Net 整体肺模型）和 98.0±1.2%（2D U-Net 单独肺模型）。
该深度神经网络在自动描绘非对比胸部 CT 图像上具有广泛病理条件的肺实质边界方面表现出优异的性能。
