Cui F T, Wang Y, Ding X P, Yao Y L, Li B, Shen F H
Occupational Health Care Management Center, Occupational Disease Prevention and Control Institute, Huaibei Mining Co., Ltd., Huaibei 235000, China.
Department of Biomedical Engineering, School of Intelligent Medicine, China Medical University, Shenyang 110122, China.
Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2023 Mar 20;41(3):177-182. doi: 10.3760/cma.j.cn121094-20220111-00011.
To construct and verify a light-weighted convolutional neural network (CNN), and explore its application value for screening the early stage (subcategory 0/1 and stage Ⅰ of pneumoconiosis) of coal workers' pneumoconiosis (CWP) from digital chest radiography (DR) . A total of 1225 DR images of coal workers who were examined at an Occupational Disease Prevention and Control Institute in Anhui Province from October 2018 to March 2021 were retrospectively collected. All DR images were collectively diagnosed by 3 radiologists with diagnostic qualifications and gave diagnostic results. There were 692 DR images with small opacity profusion 0/- or 0/0 and 533 DR images with small opacity profusion 0/1 to stage Ⅲ of pneumoconiosis. The original chest radiographs were preprocessed differently to generate four datasets, namely 16-bit grayscale original image set (Origin16), 8-bit grayscale original image set (Origin 8), 16-bit grayscale histogram equalized image set (HE16) and 8-bit grayscale histogram equalized image set (HE8). The light-weighted CNN, ShuffleNet, was applied to train the generated prediction model on the four datasets separately. The performance of the four models for pneumoconiosis prediction was evaluated on a test set containing 130 DR images using measures such as the receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, and Youden index. The Kappa consistency test was used to compare the agreement between the model predictions and the physician diagnosed pneumoconiosis results. Origin16 model achieved the highest ROC area under the curve (AUC=0.958), accuracy (92.3%), specificity (92.9%), and Youden index (0.8452) for predicting pneumoconiosis, with a sensitivity of 91.7%. And the highest consistency between identification and physician diagnosis was observed for Origin16 model (Kappa value was 0.845, 95%: 0.753-0.937, <0.001). HE16 model had the highest sensitivity (98.3%) . The light-weighted CNN ShuffleNet model can efficiently identify the early stages of CWP, and its application in the early screening of CWP can effectively improve physicians' work efficiency.
构建并验证一个轻量级卷积神经网络（CNN），并探索其在从数字化胸部X线摄影（DR）筛查煤工尘肺（CWP）早期阶段（0/1类及Ⅰ期尘肺）中的应用价值。回顾性收集了2018年10月至2021年3月在安徽省职业病防治院接受检查的煤工的1225张DR图像。所有DR图像由3名具有诊断资质的放射科医生共同诊断并给出诊断结果。其中有692张DR图像的小阴影密集度为0/-或0/0，533张DR图像的小阴影密集度为0/1至Ⅲ期尘肺。对原始胸部X线片进行不同预处理以生成四个数据集，即16位灰度原始图像集（Origin16）、8位灰度原始图像集（Origin 8）、16位灰度直方图均衡化图像集（HE16）和8位灰度直方图均衡化图像集（HE8）。应用轻量级CNN（ShuffleNet）分别在这四个数据集上训练生成预测模型。使用受试者工作特征（ROC）曲线、准确率、灵敏度、特异度和尤登指数等指标，在包含130张DR图像的测试集上评估这四个模型对尘肺预测的性能。采用Kappa一致性检验比较模型预测结果与医生诊断的尘肺结果之间的一致性。Origin16模型在预测尘肺方面达到了最高的曲线下ROC面积（AUC = 0.958）、准确率（92.3%）、特异度（92.9%）和尤登指数（0.8452），灵敏度为91.7%。并且观察到Origin16模型在识别与医生诊断之间具有最高的一致性（Kappa值为0.845，95%：0.753 - 0.937，<0.001）。HE16模型具有最高的灵敏度（98.3%）。轻量级CNN ShuffleNet模型能够有效识别CWP的早期阶段，其在CWP早期筛查中的应用可有效提高医生的工作效率。
