Background At present, the practice of pulmonary rehabilitation for pneumoconiosis in China is in a primary stage. The basis for formulating an individualized comprehensive pulmonary rehabilitation plan is still insufficient, which is one of the factors limiting the development of community-level rehabilitation work. Objective To formulate an exercise prescription based on maximum heart rate measured by cardiopulmonary exercise test (CPET), conduct an individualized comprehensive pulmonary rehabilitation program with the exercise prescription for patients with stable pneumoconiosis, and evaluate its role in improving exercise endurance and quality of life, thus provide a basis for the application and promotion of pulmonary rehabilitation. Methods A total of 68 patients were recruited from the Occupational Disease Prevention Hospital of Jinneng Holding Coal Industry Group Co., Ltd. from April to August 2022 , and were divided into an intervention group and a control group by random number table method, with 34 cases in each group. All the pneumoconiosis patients participated in a baseline test. The control group was given routine drug treatment, while the intervention group received multidisciplinary comprehensive pulmonary rehabilitation treatment on the basis of routine drug treatment, including health education, breathing training, exercise training, nutrition guidance, psychological intervention, and sleep management, whose exercise intensity was determined according to the maximum heart rate provided by CPET. The rehabilitation training lasted for 24 weeks. Patients were evaluated at registration and the end of study respectively. CPET was used to measure peak oxygen uptake per kilogram (pVO₂/kg), anaerobic threshold (AT), carbon dioxide equivalent of ventilation (EqCO₂), maximum metabolic equivalent (METs), and maximum work (W_max). The modified British Medical Research Council Dyspnea Questionnaire (mMRC), Self-rating Anxiety Scale (SAS), Self-rating Depression Scale (SDS), Pittsburgh Sleep Quality Index (PSQI), Chronic Obstructive Pulmonary Disease Assessment Test (CAT), and Short Form of Health Survey (SF-36) were used to evaluate the potential effect of the comprehensive pulmonary rehabilitation program. Results Among the included 68 patients, 63 patients were having complete data, then 31 cases were assigned in the control group and 32 cases in the interventional group. Before the intervention, there was no significant difference in pVO₂/kg, AT, EqCO₂, METs, or W_max between the two groups (P>0.05). At the end of the trail, the indicators like pVO₂/kg [(19.81±2.38) mL·(min·kg)⁻¹], AT [(14.48±2.33) mL·(min·kg)⁻¹], METs (5.64±0.69), and W_max [(85.25±14) W] of patients in the intervention group were all higher than those [(13.90±2.37) mL·(min·kg)⁻¹, (11.70±1.94) mL·(min kg)⁻¹, (3.97±0.70), and (61.77±14.72) W, respectively] in the control group (P<0.001); there was no significant difference in EqCO₂ between the two groups (P=0.083). Before the trial, there was no significant difference in mMRC, SAS, SDS, PSQI, or CAT scores between the two groups (P>0.05). At the end of the trail, the mMRC score (1.16±0.57), SAS score (27.93±2.12), SDS score (26.48±1.44), PSQI score (1.08±0.88), and CAT score (4.34±3.28) of patients in the intervention group were lower than those [(2.03±0.83), (35.87±6.91), (34.23±6.65), (5.37±3.03), and (13.87±7.53), respectively] in the control group (P<0.001). The SF-36 scores of bodily pain (94.13±10.72), general health (87.50±5.68), vitality (95.31±5.53), mental health (99.88±0.71), and health changes (74.22±4.42) in the intervention group were higher than those [(71.87±32.72), (65.81±15.55), (74.52±16.45), (86.97±16.56), and (29.84±13.50), respectively] in the control group (P<0.001), and no significant difference was found in social functioning and role emotional scores (P>0.05). Conclusion Comprehensive pulmonary rehabilitation can increase the oxygen intake and exercise endurance of pneumoconiosis patients, ameliorate dyspnea symptoms, elevate psychological state and sleep quality, and improve the quality of life.

Background It is a research hotspot to study the changes of metabolites and metabolic pathways in the process of coal worker's pneumoconiosis (CWP) by metabonomics and to explore its pathogenesis. Objective To study the change of metabolites in bronchoalveolar lavage fluid (BALF) of patients with CWP and explore the metabolic regulation mechanism of the disease. Methods Patients with CWP who met the national diagnostic criteria according to Diagnosis of occupational pneumoconiosis (GBZ 70-2015) and underwent massive whole lung lavage were selected as the case group, and patients with tracheostenosis who underwent bronchoscopy were selected as the control group. BALF samples were collected from the cases and the controls. After filtering out large particles and mucus, the supernatant was stored in a −80 ℃ refrigerator. The samples were detected and analyzed by liquid chromatography-mass spectrometry after adding extraction solution, cold bath ultrasonication, and high-speed centrifugation, and the metabolic profiles and related data of CWP patients were obtained. The differential metabolites related to the occurrence and development of CWP were screened by multiple statistical analysis; furthermore, we searched the Kyoto Encyclopedia of Genes and Genomes (KEGG) database for potential metabolic pathways involved in the progression. Results There was no significant difference in the general conditions of the subjects, such as weight, height, age, and length of service among the stage I group, the stage II group, the stage III group, and the control group (P˃0.05). When comparing the CWP stage I group with the control group, 48 differential metabolites were screened out, among which 14 were up-regulated and 34 were down-regulated. A total of 66 differential metabolites were screened out between the patients with CWP stage II and the controls, 14 up-regulated and 52 down-regulated differential metabolites. Compared with the control group, 63 differential metabolites were screened out in the patients with CWP stage III, including 11 up-regulated and 52 down-regulated differential metabolites. There were 36 differential metabolites that may be related to the occurrence of CWP, among which 11 differential metabolites were up-regulated, and 25 were down-regulated. Four significant differential metabolic pathways were identified through KEGG database query: linoleic acid metabolic pathway, alanine metabolic pathway, sphingolipid metabolic pathway, and glycerophospholipid metabolic pathway. Conclusion The metabolomic study of BALF show that there are 36 different metabolites in the occurrence and development of CWP, mainly associating with linoleic acid metabolism, alanine metabolism, sphingolipid metabolism, and glycerophospholipid metabolism pathways.

Background Pneumoconiosis is a widespread occupational disease in China at present. As a type of lung diseases, its pathological damage is mainly irreversible fibrotic changes in the lungs. Several studies have shown that the occurrence and development of lung diseases such as coal workers' pneumoconiosis are closely related to intestinal flora. Objective To observe intestinal flora of coal workers' pneumoconiosis patients based on the results of 16SrDNA high-throughput sequencing and evaluate the changes of intestinal flora after treatment with tetrandrine tablets. Methods A total of 80 patients with coal workers' pneumoconiosis attending the outpatient clinic of the Department of Occupational Diseases of the Emergency General Hospital from April to July 2022 were enrolled. All patients were treated with tetrandrine tablets for 4 weeks, with group A before the treatment of tetrandrine tablets and group B after the treatment. In the same period, 24 healthy controls (group C) were set up. Stool samples were collected before and after the treatment. Using 16SrDNA high-throughput sequencing, gene V3-V4 sequencing technology, and bioinformatic analysis platform, we evaluated the intestinal flora after treatment by groups. Results The dominant flora at the phylum level and genus level were the same across three groups. The relative abundances of phylum Bacteroidetes, Bifidobacterium, Bacteroides, and Facealibacterium in groups B and C were higher than those in group A, and the relative abundances of phy-lum Actinobacteria, genus Blautia, and genus Romboutsia in groups B and C were lower than those in group A (P＜0.05). The relative abundances of genus Clostridium, genus Megamonas, and genus Lactobacillus in group C was lower than that in groups A and B (P＜0.05). The alpha diversity analysis showed that the Chao1 index was higher in group A than in group C (P＜0.01). Compared with group A, the Shannon index was higher in group B, and the increases of Simpson index were all statistically significant in stage I patients (P＜0.05), but the differences in Chao1 index were not statistically significant (P＞0.05). The differences in the values of Chao1 index, Shannon index, and Simpson index in stage Ⅱ and stage III patients were not statistically significant (P＞0.05). The beta diversity analysis showed that the difference in flora structure between group A and group C was statistically significant (P＜0.05); the differences in flora structure before and after treatment in the same stage patients were statistically significant (P＜0.05). The partial least squares discriminant analysis (PLS-DA) showed that there were significant differences between group A and group C, and between group A and group B. The LEfSe analysis showed that the significant markers contributing to the differences were basically the same in stage I, stage Ⅱ, and stage Ⅲ after treatment, which were mainly phylum Bacteroidetes and its subordinate groups, class Negativicutes, or-der Selenomonas, and genus Facealibacterium. Conclusion There are differences in the distribution of flora between coal workers' pneumoconiosis patients and healthy individuals, and the structure and relative abundance of intestinal flora are changed and the number of beneficial flora is increased after treatment with tetrandrine tablets.

ObjectiveTo explore the distribution of pharyngeal microbiota in coal miners exposed to dust. Methods Eight coal miners who had been engaged in occupational dust exposure for more than 20 years were selected as the dust-exposed group, and four coal miners who were not exposed to dust at work were selected as the control group using the judgment sampling method. Pharyngeal secretions of the coal miners were collected with throat swabs, and its pharyngeal microbiota was analyzed. The diversity, abundance and evenness of the microbiota were analyzed by gene sequencing using the 16sRNA gene high-throughput sequencing technology. Results A total of 254 operational taxonomic units of pharyngeal microbiota were detected in the coal miners in the control group, which was 210 more than that in the dust-exposed group. The Chao1 index, Shannon index, PD-tree index and Pielou index of pharyngeal microbiota in the dust-exposed group decreased compared with the control group (all P<0.01). The abundance of Bacteroidetes and Clostridum, at the phylum level, in the pharynx of coal miners in the dust-exposed group was higher than that in the control group (all P<0.05). The abundance of Prevotella, Neisseria, and Monas, at the genus level, in the pharynx of coal miners in the dust-exposed group was higher than that in the control group(all P<0.05), while the abundance of Lactobacillus decreased (P<0.05). The analysis results of the receiver operating characteristic curve showed that Lactobacillus, Fusobacterium and Rothia may play a role for pharyngeal microbiota imbalance prediction in dust-exposed workers, and the area under the curves were all 1.00±0.00. Conclusion The species diversity and evenness of pharyngeal microbiota in coal miners exposed to dust are decreased, which may be related to the continuous inhalation of coal dust that disrupts the microbial environment of the throat.

Background Multi-slice spiral computerized tomography (MSCT) can be used as an auxiliary diagnosis of chest radiography in diagnosis of pneumoconiosis, but there are few studies on the correlations between interstitial images and stage classification of coal workers' pneumoconiosis in the existing literature. Objective To present MSCT imaging manifestations and distribution characteristics of coal workers' pneumoconiosis and complications, evaluate correlations between coal workers' pneumoconiosis stages and pulmonary interstitial lesions, and provide a reliable imaging diagnosis basis for pneumoconiosis interstitial lesions. Methods From June 2022 to June 2023, a total of 1002 patients with coal workers' pneumoconiosis confirmed by the pneumoconiosis diagnostic and identification group in the Department of Occupational Diseases of the Emergency General Hospital were enrolled. MSCT was used to observe the abnormal imaging manifestations of the lungs of coal workers' pneumoconiosis patients and the diseases of pulmonary fibrosis related to their own diseases (thickening of the interlobular septum, bronchial perivascular interstitial mass thickening, parenchymal banding, subpleural line, intralobular interstitial thickening, honeycomb, and subpleural interstitial thickening), the occurrence of coal workers' pneumoconiosis and complications (old tuberculosis, active tuberculosis, pneumonia, atelectasis, lung cancer, bronchiectasis), and the density, size, and location of pneumoconiosis nodules. Imaging data were analyzed and statistically processed. Results All 1002 patients were male, with an average age of (60.71±6.87) years and an average dust exposure time of (23.01±7.80) years. Among them, there were 470 patients with stage I, 422 patients with stage II, and 110 patients with stage III. There were significant differences in the distribution of thickening of the interlobular septum, bronchial perivascular interstitial mass thickening, parenchymal banding, intralobular interstitial thickening, subpleural interstitial thickening, and honeycomb across different stages (P<0.05). Statistically significant differences in p, q, and r subsets of round nodules were found in patients with pneumoconiosis at different stages (P<0.05). Observed nodule types included solid nodules, pure ground-glass shadow nodules, and partial solid nodules. There were statistically significant differences in pulmonary tuberculosis and bronchiectasis among different stages of coal workers' pneumoconiosis (P<0.05). There were statistically significant differences in interstitial shadows and patches combined with interstitial shadows among different stages of pneumoconiosis complicated with pneumonia (P<0.05). Conclusion MSCT provides images of the progression of coal workers' pneumoconiosis and have a certain relationship with the stages of coal workers' pneumoconiosis, which is conducive to the formulation of reasonable treatment plans in the early clinical stage. Therefore, in the diagnosis and treatment of pneumoconiosis, a great attention should be paid to the imaging technology of chest computerized tomography, especially the use of MSCT examination.

Objective:To explore the active ingredients of shengxian and jinshuiliujun decoction with the method of network pharmacology, and to verify the experimental mechanism of its treatment of silicosis.Methods:In May 2023, the active ingredients and targets of drugs in shengxian and jinshuiliujun decoction were obtained through the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) database. The target of silicosis disease was screened by databases such as Genecards, Disease Gene Network (DisGeNET), Comparative Toxicogenomics Database (CTD), etc. The screened drug targets and disease targets were intersected to obtain the target set of shengxian and jinshuiliujun decoction for the treatment of silicosis. Protein-protein interaction (PPI) network analysis was performed on the target set through STRING database, and core target genes were screened. GO enrichment analysis and KEGG pathway analysis of intersection genes were performed based on Metascape database, and molecular docking verification of key components and targets of shengxian and jinshuiliujun decoction was carried out. Twenty-four adult male SD rats with SPF grade were randomly divided into control group, model group and TCM intervention group, with 8 rats in each group. The dust-stained rat model was prepared by non-tracheal exposure of 1 ml silica suspension (50 mg/ml) in one go, and TCM intervention group was given shengxian and jinshuiliujun decoction[6 g/ (kg·d) ] on the second day. The CT of the lungs of each group was observed 28 days after the dust-stained rat model. Paraffin sections of rat lung tissues were prepared and stained with Hematoxylin-Eosin (HE) and Masson. Western blot was used to verify the expression of core target-related proteins in rat lung tissues after the intervention of shengxian and jinshuiliujun decoction for 28 days, and the differences in protein expression between groups were compared by one-way analysis of variance.Results:A total of 205 active ingredients and 3345 active compounds were selected from shengxian and jinshuiliujun decoction, corresponding to 281 targets, among which 240 targets were related to silicosis. Serine/threonine kinase 1 (AKT1), tumor protein p53 (TP53), tumor necrosis factor (TNF) and interleukin (IL) 6 may be the key targets of shengxian and jinshuiliujun decoction in the treatment of silicosis. Through enrichment analysis, 30 GO entries and 20 potential signaling pathways were screened according to P-value, including nuclear factor κB (NF-κB), mitogen-activated protein kinase (MAPK) and cancer signaling pathways. Molecular docking showed that the active compounds of shengxian and jinshuiliujun decoction had good binding with the core target proteins, and the strongest binding properties were beta-sitosterol and TNF-α (-10.45 kcal/mol). In animal experiments, the inflammatory infiltration and fibrosis of lung tissue of rats in TCM intervention group were significantly improved. Compared with control group, the levels of TNF-α, IL-1β, IL-6 and NF-κB in lung tissue of model group were significantly increased ( P<0.05). Compared with model group, the lung injury of rats in TCM intervention group was significantly improved, and the expressions of TNF-α, IL-1β, IL-6 and NF-κB were significantly decreased ( P<0.05) . Conclusion:Shengxian and jinshuiliujun decoction in the treatment of silicosis may play an anti-fibrosis role by inhibiting the NF-κB signal transduction pathway mediated by inflammatory factors such as TNF-α and IL-1β, which provides a reference for further exploring the material basis and mechanism of its action.

Objective:To explore the active ingredients of shengxian and jinshuiliujun decoction with the method of network pharmacology, and to verify the experimental mechanism of its treatment of silicosis.Methods:In May 2023, the active ingredients and targets of drugs in shengxian and jinshuiliujun decoction were obtained through the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) database. The target of silicosis disease was screened by databases such as Genecards, Disease Gene Network (DisGeNET), Comparative Toxicogenomics Database (CTD), etc. The screened drug targets and disease targets were intersected to obtain the target set of shengxian and jinshuiliujun decoction for the treatment of silicosis. Protein-protein interaction (PPI) network analysis was performed on the target set through STRING database, and core target genes were screened. GO enrichment analysis and KEGG pathway analysis of intersection genes were performed based on Metascape database, and molecular docking verification of key components and targets of shengxian and jinshuiliujun decoction was carried out. Twenty-four adult male SD rats with SPF grade were randomly divided into control group, model group and TCM intervention group, with 8 rats in each group. The dust-stained rat model was prepared by non-tracheal exposure of 1 ml silica suspension (50 mg/ml) in one go, and TCM intervention group was given shengxian and jinshuiliujun decoction[6 g/ (kg·d) ] on the second day. The CT of the lungs of each group was observed 28 days after the dust-stained rat model. Paraffin sections of rat lung tissues were prepared and stained with Hematoxylin-Eosin (HE) and Masson. Western blot was used to verify the expression of core target-related proteins in rat lung tissues after the intervention of shengxian and jinshuiliujun decoction for 28 days, and the differences in protein expression between groups were compared by one-way analysis of variance.Results:A total of 205 active ingredients and 3345 active compounds were selected from shengxian and jinshuiliujun decoction, corresponding to 281 targets, among which 240 targets were related to silicosis. Serine/threonine kinase 1 (AKT1), tumor protein p53 (TP53), tumor necrosis factor (TNF) and interleukin (IL) 6 may be the key targets of shengxian and jinshuiliujun decoction in the treatment of silicosis. Through enrichment analysis, 30 GO entries and 20 potential signaling pathways were screened according to P-value, including nuclear factor κB (NF-κB), mitogen-activated protein kinase (MAPK) and cancer signaling pathways. Molecular docking showed that the active compounds of shengxian and jinshuiliujun decoction had good binding with the core target proteins, and the strongest binding properties were beta-sitosterol and TNF-α (-10.45 kcal/mol). In animal experiments, the inflammatory infiltration and fibrosis of lung tissue of rats in TCM intervention group were significantly improved. Compared with control group, the levels of TNF-α, IL-1β, IL-6 and NF-κB in lung tissue of model group were significantly increased ( P<0.05). Compared with model group, the lung injury of rats in TCM intervention group was significantly improved, and the expressions of TNF-α, IL-1β, IL-6 and NF-κB were significantly decreased ( P<0.05) . Conclusion:Shengxian and jinshuiliujun decoction in the treatment of silicosis may play an anti-fibrosis role by inhibiting the NF-κB signal transduction pathway mediated by inflammatory factors such as TNF-α and IL-1β, which provides a reference for further exploring the material basis and mechanism of its action.

Objective Based on the multi-software visual analysis of the literature on the effect of Janus kinase(JAK)/signal transducer and activator of transcription(STAT)signaling pathway on rheumatoid arthritis in the past decade,the development trend and research hotspot in this field are summarized.To provide researchers with new directions and ideas to promote the innovative development of the field.Methods The literatures related to JAK/STAT signaling pathway in rheumatoid arthritis were collected from the Web of Science Core Collection database from 2013 to 2023.CiteSpace and VOSviewer software were used to analyze the number of publications,countries,authors and keywords of 354 articles retrieved.Results The number of published papers in this field continues to increase.According to the author's research direction,the presentation of high-frequency words,and the attention to the preface and hot topics,it is suggested that this field focuses on gene expression,immune mechanism,inflammatory mechanism,pathway inhibitors,drug therapy,etc.Future research will focus on the safety,mechanism and controlled trials of pathway inhibitors and antirheumatic drugs.Conclusion The effect of JAK/STAT signaling pathway on rheumatoid arthritis has attracted much attention in the past,present and future.There are differences in the research of different teams in this field,and the regional development is unbalanced,suggesting that we should strengthen cooperation and exchanges,focus on the international frontier,and carry out more high-quality research to promote the development and progress of this field,and provide clinical basis.

Objective To observe the effect of Danshen injection (DSI) on pulmonary fibrosis in silicosis mice, and to analyze the differential metabolic pathway on pulmonary fibrosis in silicosis using DSI by urine metabolomics. Methods The specific pathogen free C57BL/6J mice were randomly divided into control group, silicosis model group, DSI prevention group and DSI treatment group. The mice in the last three groups were given 1 mL silica suspension with a mass concentration of 50 g/L by the one-time non-exposed tracheal method, and the mice in the control group were not given any treatment. Subsequently, mice in the DSI prevention group and the DSI treatment group were given intraperitoneal injection of DSI with a dose of 5 mL/kg body weight from 24 hours after exposure to dust and from the 29th day after exposure to dust, respectively, once per day until the 56th day after exposure. Mice in the other two groups were not treated. After DSI intervention, the lung histopathological changes of mice in all groups were evaluated. The components of mouse urine metabolites were analyzed using ultra-high performance liquid chromatography-quadrupole-time-of-fight mass spectrometry method. Human Metabolome Database was used to screen the potential differential metabolites (DMs). The related metabolic pathways were analyzed using MetaboAnanlyst 5.0 Web analytics platform. Results The result of hematoxylin-eosin staining and Van Gieson staining of mouse lung tissues showed that the pulmonary alveolar structure destroyed, typical fibrotic nodules appeared, collagen fiber deposition increased, and clumpy accumulation in the silicosis model group, compared with the control group. Compared with the silicosis model group, the degree of pulmonary alveolar inflammation and fibrosis in the lung tissues of mice in the DSI prevention group was obviously reduced to close to the control group, while pulmonary alveolar inflammation and fibrosis in the lung tissues of mice in the DSI treatment group were also reduced, although the outcome was not as good as that in the DSI prevention group. The result of urine metabolomics analysis identified four DMs in the model group and control group, seven DMs were identified in the DSI prevention group and silicosis model group, seven DMs were identified in the DSI treatment group and silicosis model group. A total of three DMs pathways related to pulmonary fibrosis in silicosis model group and the protective effect of DSI prevention group were identified, including D-arginine and D-ornithine metabolism, folic acid biosynthesis and metabolism, pantothenate and succinyl coenzyme A biosynthesis pathways (all P<0.01). Conclusion DSI treatment in any time point can interfere the process of pulmonary fibrosis in the silicosis mice, while the interference is more effective in the DSI group treated right after dust-exposure. DSI interferes with the urinary metabolism pathway of silicosis mice, and the D-arginine and D-ornithine metabolism, folic acid biosynthesis and metabolism, pantothenate and succinyl coenzyme A biosynthesis pathways may participate in the inhibiting process of early pulmonary fibrosis in silicosis mice by DSI.

Background Diagnosis of pneumoconiosis by radiologist reading chest X-ray images is affected by many factors and is prone to misdiagnosis/missed diagnosis. With the rapid development of artificial intelligence in the field of medical imaging, whether artificial intelligence can be used to read images of pneumoconiosis deserves consideration. Objective Three deep learning models for identifying presence of pneumoconiosis were constructed based on deep convolutional neural network. An optimal model was selected by comparing diagnostic efficiency of the three models. Methods Digital radiography (DR) chest images were collected between June 2017 and December 2020 from 7 hospitals and standard radiograph quality control protocol was also followed. The DR chest images with positive results were classified into the positive group, while those without pneumoconiosis were classified into the negative group. The collected chest radiographs were labeled by experts who had passed the assessment of reading radiographs,and the experts were constantly assessed for consistency in the labeling process based on an expectation-maximization algorithm. The labeled data were cleaned, archived, and preprocessed, and then were grouped into a training set and a verification set. Three deep convolutional neural network models TMNet, ResNet-50, and ResNeXt-50 were constructed and trained by ten-fold cross-validation method to obtain an optimal model. Five hundred cases of DR chest radiographs that were not included in the training set and the validation set were collected, and identified by five senior experts as the gold standard, named the test set. The accuracy rate, sensitivity, specificity, area under curve (AUC), and other indexes of the three models were derived after testing, and the efficiency of the three models was evaluated and compared. Results A total of 24867 DR chest radiographs of the training set and the validation set were collected in this study, including 6978 images in the positive group and 17889 images in the negative group. There were 312 cases of pulmonary abnormalities such as pneumothorax and pulmonary tuberculosis. A total of nine experts labeled the chest radiographs, the labeling consistency rate of pneumoconiosis (non-staging) was above 88%, and the labeling consistency rate of pneumoconiosis staging ranged from 84.68% to 93.66%. The diagnostic accuracy, sensitivity, specificity, and AUC of TMNet were 95.20%, 99.66%, 88.61%, and 0.987, respectively. The indicators of ResNeXt were 87.00%, 89.93%, 82.67%, and 0.911, respectively. Those of ResNet were 84.00%, 85.91%, 81.19%, and 0.912, respectively. All these indexes of TMNet were higher than those of ResNeXt-50 and ResNet-50 models. The AUC differences between TMNet and the other two models were both statistically significant (P<0.001). Conclusion All the three convolutional neural network models can effectively diagnose the presence of pneumoconiosis, among which TMNet provides the best efficiency.