Radiology Department, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
Radiology Department, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands.
Br J Radiol. 2022 May 1;95(1133):20210637. doi: 10.1259/bjr.20210637. Epub 2022 Feb 16.
Chest CT can display the main pathogenic factors of chronic obstructive pulmonary disease (COPD), emphysema and airway wall remodeling. This study aims to establish deep convolutional neural network (CNN) models using these two imaging markers to diagnose and grade COPD.
Subjects who underwent chest CT and pulmonary function test (PFT) from one hospital ( = 373) were retrospectively included as the training cohort, and subjects from another hospital ( = 226) were used as the external test cohort. According to the PFT results, all subjects were labeled as Global Initiative for Chronic Obstructive Lung Disease (GOLD) Grade 1, 2, 3, 4 or normal. Two DenseNet-201 CNNs were trained using CT images of lung parenchyma and bronchial wall to generate two corresponding confidence levels to indicate the possibility of COPD, then combined with logistic regression analysis. Quantitative CT was used for comparison.
In the test cohort, CNN achieved an area under the curve of 0.899 (95%CI: 0.853-0.935) to determine the existence of COPD, and an accuracy of 81.7% (76.2-86.7%), which was significantly higher than the accuracy 68.1% (61.6%-74.2%) using quantitative CT method ( < 0.05). For three-way (normal, GOLD 1-2, and GOLD 3-4) and five-way (normal, GOLD 1, 2, 3, and 4) classifications, CNN reached accuracies of 77.4 and 67.9%, respectively.
CNN can identify emphysema and airway wall remodeling on CT images to infer lung function and determine the existence and severity of COPD. It provides an alternative way to detect COPD using the extensively available chest CT.
CNN can identify the main pathological changes of COPD (emphysema and airway wall remodeling) based on CT images, to infer lung function and determine the existence and severity of COPD. CNN reached an area under the curve of 0.853 to determine the existence of COPD in the external test cohort. The CNN approach provides an alternative and effective way for early detection of COPD using extensively used chest CT, as an important alternative to pulmonary function test.
胸部 CT 可显示慢性阻塞性肺疾病（COPD）的主要致病因素，包括肺气肿和气道壁重塑。本研究旨在使用这两种影像学标志物建立深度学习卷积神经网络（CNN）模型，以诊断和分级 COPD。
回顾性纳入一家医院（n=373）行胸部 CT 和肺功能检查（PFT）的受试者作为训练队列，另一家医院（n=226）的受试者作为外部测试队列。根据 PFT 结果，所有受试者均被标记为全球倡议慢性阻塞性肺疾病（GOLD）1 级、2 级、3 级、4 级或正常。使用肺实质和支气管壁 CT 图像训练两个 DenseNet-201 CNN，生成两个对应的置信度水平来指示 COPD 的可能性，然后结合逻辑回归分析。与定量 CT 进行比较。
在测试队列中，CNN 用于确定 COPD 存在的曲线下面积为 0.899（95%CI：0.853-0.935），准确率为 81.7%（76.2%-86.7%），明显高于定量 CT 方法的准确率 68.1%（61.6%-74.2%）（<0.05）。对于三分类（正常、GOLD 1-2 和 GOLD 3-4）和五分类（正常、GOLD 1、2、3 和 4），CNN 分别达到了 77.4%和 67.9%的准确率。
CNN 可识别 CT 图像上的肺气肿和气道壁重塑，以推断肺功能并确定 COPD 的存在和严重程度。它为使用广泛应用的胸部 CT 检测 COPD 提供了一种替代方法。
CNN 可以根据 CT 图像识别 COPD 的主要病理变化（肺气肿和气道壁重塑），以推断肺功能并确定 COPD 的存在和严重程度。在外部测试队列中，CNN 用于确定 COPD 存在的曲线下面积达到 0.853。CNN 方法为使用广泛应用的胸部 CT 进行 COPD 的早期检测提供了一种替代和有效的方法，是肺功能检查的重要替代方法。
