3.A systematic review of digital radiography for the screening and recognition of pneumoconiosis.
Chinese Journal of Industrial Hygiene and Occupational Diseases 2014;32(5):327-334
OBJECTIVETo conduct a systematic review of studies reporting the comparison of digital radiography (DR) with conventional film-screen radiograph (FSR) in the screening and recognition of pneumoconiosis worldwide, to evaluate the feasibility of DR in the screening and recognition of pneumoconiosis, to analyze the similarity and difference between DR and FSR, to explore the main challenge to utilize DR in the future.
METHODSThe national and international databases were systematically searched for original articles on DR for screening and recognition of pneumoconiosis published from first Jan 1998 to first Nov 2013, making evaluation and selection of them, and qualitative data and quantitative data were extracted independently from the selected articles and systematically reviewed.
RESULTSFive hundred and twenty articles were found and evaluated and nine of them met the inclusion criteria of systematic review. The research time started from 2002 to 2011 whose objects mainly came from pneumoconiosis cases and dust-exposed workers and control population examined with DR and FSR using the high kV radiography from 120 to 130 kV. The chest radiographs were read at blind and random and standard control method. There were only two papers compare the validity of DR and FSR for recognition and classification of pneumoconiosis using gold standards. There were still some diversity of imaging processing and imaging reading without design and assessment using Standards for Reporting of Diagnostic Accuracy (STARD) in these researches. The evaluation index of the nine articles include detection rate of small opacities, crude agreement, Kappa value of Kappa Consistency Test, Area Under the Curve of ROC, etc. Seven of the nine selected articles estimated DR has generally produced superior image qualities compared to FSR. Four papers had a conclusion that DR could be equivalent to FSR in identification of shapes and profusion of small opacities and in classification of pneumoconiosis. Five papers considered DR had higher presence of pneumoconiosis comparing with FSR especially in recognition the pneumoconiosis of category 1. The variation between different film formats of DR and FSR were smaller than that within and between readers for classification of pneumoconiosis.
CONCLUSIONAlthough there are still some imperfections in the existent researches to solve, DR can be equivalent to FSR in screening and recognition of pneumoconiosis. It is necessary to develop technical specifications of DR and standard digital chest radiographs for pneumoconiosis including both hard copy and soft copy, and develop an evaluation criterion on chest images of DR.
Humans ; Mass Screening ; Pneumoconiosis ; diagnostic imaging ; Radiographic Image Enhancement ; methods
4.Discussion of grading method of small opacity profusion of pneumoconiosis on CT scans and the corresponding reference images.
R C ZHAI ; N C LI ; X D LIU ; S K ZHU ; B F HU ; A N ZHANG ; X TONG ; G D WANG ; Y J WAN ; Y MA
Chinese Journal of Industrial Hygiene and Occupational Diseases 2021;39(6):453-457
5.CT quantitative study of coal miner's pneumoconiosis.
Peicheng LIU ; Dun ZHANG ; Chun WU ; Hanxin SU ; Jingbo CHEN ; Guiping CAI ; Xueru AI ; A WAGULI
Chinese Journal of Industrial Hygiene and Occupational Diseases 2002;20(2):113-115
OBJECTIVETo study the value of CT quantitativeness in the diagnosis of coal miner's pneumoconiosis.
METHODS104 cases were examined by HRCT scan at top of aortic arc, carina of trachea, 3 cm below the bifurcation of bronchi, among them there were 87 patients with different stages of coal miner's pneumoconiosis, 17 cases of normal males as the control group. All images were determined by CT density histogram at specific region (- 1,024-0 HU). Calculated the percentage of each pixel included a varying number of CT value, and the ratio of density values in the specific region.
RESULTSThe ratio of density values in the region of -983 (-) -778 HU was 87.31% in normal control group, and 80.51%, 75.27% and 72.99% respectively in the I, II, III stages of coal miner's pneumoconiosis. There were statistically significant differences among the groups (P < 0.01).
CONCLUSIONCT quantitative histogram information was able to observe the fibrosis and its degree of coal miner's pneumoconiosis. It has a good diagnostic value for its reliability and objectiveness.
Coal Mining ; Humans ; Pneumoconiosis ; diagnostic imaging ; Pulmonary Fibrosis ; diagnostic imaging ; Tomography, X-Ray Computed ; methods
8.Application of a light-weighted convolutional neural network for automatic recognition of coal workers' pneumoconiosis in the early stage.
Feng Tao CUI ; Yan WANG ; Xin Ping DING ; Yu Long YAO ; Bing LI ; Fu Hai SHEN
Chinese Journal of Industrial Hygiene and Occupational Diseases 2023;41(3):177-182
Objective: 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) . Methods: 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. 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%CI: 0.753-0.937, P<0.001). HE16 model had the highest sensitivity (98.3%) . Conclusion: 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.
Humans
;
Retrospective Studies
;
Anthracosis/diagnostic imaging*
;
Pneumoconiosis/diagnostic imaging*
;
Coal Mining
;
Neural Networks, Computer
;
Coal