DLA-Net: dual lesion attention network for classification of pneumoconiosis using chest X-ray images.

Accurate and early detection of pneumoconiosis using chest X-rays (CXR) is important for preventing the progression of this incurable disease. It is also a challenging task due to large variations in appearance, size and location of lesions in the lung regions as well as inter-class similarity and intra-class variance. Compared to traditional methods, Convolutional Neural Networks-based methods have shown improved results; however, these methods are still not applicable in clinical practice due to limited performance. In some cases, limited computing resources make it impractical to develop a model using whole CXR images. To address this problem, the lung fields are divided into six zones, each zone is classified separately and the zone classification results are then aggregated into an image classification score, based on state-of-the-art. In this study, we propose a dual lesion attention network (DLA-Net) for the classification of pneumoconiosis that can extract features from affected regions in a lung. This network consists of two main components: feature extraction and feature refinement. Feature extraction uses the pre-trained Xception model as the backbone to extract semantic information. To emphasise the lesion regions and improve the feature representation capability, the feature refinement component uses a DLA module that consists of two sub modules: channel attention (CA) and spatial attention (SA). The CA module focuses on the most important channels in the feature maps extracted by the backbone model, and the SA module highlights the spatial details of the affected regions. Thus, both attention modules combine to extract discriminative and rich contextual features to improve classification performance on pneumoconiosis. Experimental results show that the proposed DLA-Net outperforms state-of-the-art methods for pneumoconiosis classification.

Computerized Classification of Pneumoconiosis on Digital Chest Radiography Artificial Neural Network with Three Stages.

It is difficult for radiologists to classify pneumoconiosis from category 0 to category 3 on chest radiographs. Therefore, we have developed a computer-aided diagnosis (CAD) system based on a three-stage artificial neural network (ANN) method for classification based on four texture features. The image database consists of 36 chest radiographs classified as category 0 to category 3. Regions of interest (ROIs) with a matrix size of 32 × 32 were selected from chest radiographs. We obtained a gray-level histogram, histogram of gray-level difference, gray-level run-length matrix (GLRLM) feature image, and gray-level co-occurrence matrix (GLCOM) feature image in each ROI. For ROI-based classification, the first ANN was trained with each texture feature. Next, the second ANN was trained with output patterns obtained from the first ANN. Finally, we obtained a case-based classification for distinguishing among four categories with the third ANN method. We determined the performance of the third ANN by receiver operating characteristic (ROC) analysis. The areas under the ROC curve (AUC) of the highest category (severe pneumoconiosis) case and the lowest category (early pneumoconiosis) case were 0.89 ± 0.09 and 0.84 ± 0.12, respectively. The three-stage ANN with four texture features showed the highest performance for classification among the four categories. Our CAD system would be useful for assisting radiologists in classification of pneumoconiosis from category 0 to category 3.

[Guidelines for the use of the International Classification of Radiographs of Pneumoconioses of the International Labour Office (ILO): Substantial changes in the currrent edition]. / Wytyczne stosowania Miedzynarodowej Klasyfikacji Radiogramów Pylic Pluc Miedzynarodowego Biura Pracy (ILO) ­ istotne zmiany w obecnie obowiazujacej edycji.

The International Classification of Radiographs of Pneumoconioses is the scheme worked out by the International Labour Office in Geneva (ILO), to register radiographic chest abnormalities in a well-ordered, reproducible and comparable way. It is used for diagnosing abnormalities caused by dust exposure. Guidelines for the use of the ILO International Classification of Radiographs of Pneumoconioses contain detailed information and recommendations on how to use the classification, as well as how the chest X-ray examination should be performed and recorded. To facilitate the diagnosis of observed abnormalities the classification is completed by the set of standard radiograms illustrating typical irregularities referring to lungs and pleura, included in the classification. The article presents the key information on classification and the most important amendments adopted in the 2000 and 2011 ILO guidelines revisions. These changes refer to radiographs quality assessment, the way of presenting abnormalities registered in standard radiographs (QUAD set, digital images) and registration of failures not related to dust exposure. Particularly important complements result from the development of radiological imaging techniques. They are concerned about the classification of radiographic images of the chest recorded digitally. Med Pr 2016;67(6):833-837.

Patrón de lectura radiológica digital para la clasificación de imágenes compatibles con silicosis / Digital reading radiation pattern for classification of image about silicosis (P0140-2013)

La silicosis es una enfermedad sin tratamiento. El programa de vigilancia de trabajadores expuestos a sílice está diseñado para detectarla lo más precozmente posible mediante una radiografía de tórax. A pesar de su rol fundamental, esta adolece de una gran variabilidad técnica y de interpretación. Para disminuir estos fenómenos, la OIT creó un sistema de clasificación ampliamente usado en Chile y el mundo. Este se basó en la radiografía análoga, la que ha sido reemplazada por el mismo set análogo, digitalizado. Si bien la OIT permite el uso de este nuevo tipo de imagen, indica que el patrón de comparación debe seguir siendo el análogo. Muchos estudios han demostrado que es mejor utilizar patrones de comparación digitales. El objetivo de este trabajo fue generar un set de radiografías digitales que representen las profusiones centrales de opacidades redondas del sistema de clasificación OIT. De un universo de 32.826 radiografías de pacientes expuestos a sílice, se obtuvo 252 radiografías, de las cuales 54 fueron informadas por 3 lectores acreditados, seleccionándose 8 casos que contaban con opacidades redondas puras de diferentes profusiones, las que se proponen como patrón digital para Chile.

Silicosis is a disease without treatment. The surveillance program for workers exposed to silica is designed to detect as early as possible by a chest radiograph. Despite its fundamental role, it suffers from a great technique and interpretation variability. To reduce these phenomena, the ILO created a classification system widely used in Chile and the world. This was based on analog radiography, which has been replaced by the same analog set, digitized. While the ILO permits the use of this new type of image it indicates that the pattern comparison must remain analog. Many studies have shown that it is best to use digital comparison patterns. The objective of this work was to generate a set of digital x-rays representing the central profusion of round opacities of ILO classification system. From a universe of 32,826 x-rays of patients exposed to silica, 252 films, of which 54 were reported for 3 accredited readers, selected 8 cases that had pure round opacities of different profusions was obtained, which are proposed as digital pattern for Chile.

From the definition of silicosis at the 1930 Johannesburg conference to the blurred boundaries between pneumoconioses, sarcoidosis, and pulmonary alveolar proteinosis (PAP).

The 1930 International Labour Office Conference on silicosis in Johannesburg identified silicosis by setting a medicolegal framework to its nosology: as with other occupational illnesses, its medical content was fixed under economic pressure. This article follows a reading of all the proceedings of this conference (debates and reports of experts) to examine their potential impact on the etiology and nosology of other diseases, specifically sarcoidosis and pulmonary alveolar proteinosis (PAP), "idiopathic" diseases in which inorganic particles may be involved. We propose renewed study of the role of inorganic particles in these diseases. To do this, we propose to mobilize detection means such as mineralogical analysis and electron microscopy and in depth interviewing that are currently seldom used in France, in order to establish diagnosis and the potential occupational and environmental origin of these diseases.

Relationships (II) of International Classification of High-resolution Computed Tomography for Occupational and Environmental Respiratory Diseases with ventilatory functions indices for parenchymal abnormalities.

The International Classification of High-Resolution Computed Tomography (HRCT) for Occupational and Environmental Respiratory Diseases (ICOERD) is used to screen and diagnose respiratory illnesses. Using univariate and multivariate analysis, we investigated the relationship between subject characteristics and parenchymal abnormalities according to ICOERD, and the results of ventilatory function tests (VFT). Thirty-five patients with and 27 controls without mineral-dust exposure underwent VFT and HRCT. We recorded all subjects' occupational history for mineral dust exposure and smoking history. Experts independently assessed HRCT using the ICOERD parenchymal abnormalities (Items) grades for well-defined rounded opacities (RO), linear and/or irregular opacities (IR), and emphysema (EM). High-resolution computed tomography showed that 11 patients had RO; 15 patients, IR; and 19 patients, EM. According to the multiple regression model, age and height had significant associations with many indices ventilatory functions such as vital capacity, forced vital capacity, and forced expiratory volume in 1 s (FEV1). The EM summed grades on the upper, middle, and lower zones of the right and left lungs also had significant associations with FEV1 and the maximum mid-expiratory flow rate. The results suggest the ICOERD notation is adequate based on the good and significant multiple regression modeling of ventilatory function with the EM summed grades.

Relationships (I) of International Classification of High-resolution Computed Tomography for Occupational and Environmental Respiratory Diseases with the ILO International Classification of Radiographs of Pneumoconioses for parenchymal abnormalities.

The International Classification of High-resolution Computed Tomography (HRCT) for Occupational and Environmental Respiratory Diseases (ICOERD) has been developed for the screening, diagnosis, and epidemiological reporting of respiratory diseases caused by occupational hazards. This study aimed to establish a correlation between readings of HRCT (according to the ICOERD) and those of chest radiography (CXR) pneumoconiotic parenchymal opacities (according to the International Labor Organization Classification/International Classification of Radiographs of Pneumoconioses [ILO/ICRP]). Forty-six patients with and 28 controls without mineral dust exposure underwent posterior-anterior CXR and HRCT. We recorded all subjects' exposure and smoking history. Experts independently read CXRs (using ILO/ICRP). Experts independently assessed HRCT using the ICOERD parenchymal abnormalities grades for well-defined rounded opacities (RO), linear and/or irregular opacities (IR), and emphysema (EM). The correlation between the ICOERD summed grades and ILO/ICRP profusions was evaluated using Spearman's rank-order correlation. Twenty-three patients had small opacities on CXR. HRCT showed that 21 patients had RO; 20 patients, IR opacities; and 23 patients, EM. The correlation between ILO/ICRP profusions and the ICOERD grades was 0.844 for rounded opacities (p<0.01). ICOERD readings from HRCT scans correlated well with previously validated ILO/ICRP criteria. The ICOERD adequately detects pneumoconiotic micronodules and can be used for the interpretation of pneumoconiosis.

[Recognize and classify pneumoconiosis]. / Pneumokoniosen erkennen und klassifizieren.

BACKGROUND: In the year 2012, out of the 10 most frequently recognized occupational diseases 6 were forms of pneumoconiosis. With respect to healthcare and economic aspects, silicosis and asbestos-associated diseases are of foremost importance. The latter are to be found everywhere and are not restricted to large industrial areas. PROBLEM: Radiology has a central role in the diagnosis and evaluation of occupational lung disorders. In cases of known exposure mainly to asbestos and quartz, the diagnosis of pneumoconiosis, with few exceptions will be established primarily by the radiological findings. As these disorders are asymptomatic for a long time they are quite often detected as incidental findings in examinations for other reasons. Therefore, radiologists have to be familiar with the pattern of findings of the most frequent forms of pneumoconiosis and the differential diagnoses. STANDARDIZED PROCEDURE IN EXAMINATIONS: For reasons of equal treatment of the insured a quality-based, standardized performance, documentation and evaluation of radiological examinations is required in preventive procedures and evaluations. Above all, a standardized low-dose protocol has to be used in computed tomography (CT) examinations, although individualized concerning the dose, in order to keep radiation exposure as low as possible for the patient. STANDARDIZED EVALUATION: The International Labour Office (ILO) classification for the coding of chest X-rays and the international classification of occupational and environmental respiratory diseases (ICOERD) classification used since 2004 for CT examinations meet the requirements of the insured and the occupational insurance associations as a means of reproducible and comparable data for decision-making.

[The problems of classification of the occupational interstitial lung diseases, approaches to solving].

The analysis of the applied in Russia classifications of pneumoconiosis and occupational hypersensitivity pneumonitis in relation to the international classifications of occupational lung diseases and the classification of interstitial lung diseases (ILD) was performed. The necessity of a new approach to the classification of occupational ILD was proved and solutions of the problem were offered.

Utility of digital radiography for the screening of pneumoconiosis as compared to analog radiography: radiation dose, image quality, and pneumoconiosis classification.

The purpose of this study was to compare digital radiography (DR) and analog radiography (AR) for the screening of pneumoconiosis with respect to radiation dose, image quality, and pneumoconiosis classification. DR was performed on 50 subjects who were enrolled for an examination of pneumoconiosis (Digital Diagnost™, Philips, Netherlands), and AR (MXO-15B, Toshiba, Japan) was performed the same day after the study was approved by the Institutional Review Board and written informed consent was obtained from all subjects. Entrance surface doses (ESDs) of DR and AR were measured using a glass dosimeter attached to a Rando human phantom (Alderson Co., U.S.) under exposure conditions commonly used in clinical practice in Korea. Visibilities on all images were evaluated using a 5-point scale by four chest radiologists using a modified form of the European Chest Guidelines (EUR 16260). All the images were classified using the ILO's guidelines by referencing standard analog radiographs. ESDs of DR were significantly lower than those of AR (0.15 mGy vs. 0.21 mGy, p < 0.05). All anatomic structures were significantly more visible by DR images (p < 0.0001), especially the left main bronchus, ribs, and thoracic spine. Body mass index did not correlate with anatomic structure visibility by DR (r = -0.029, p = 0.842) or AR images (r = -0.076, p = 0.602). Overall intra- and inter-reader agreements for DR images were significantly higher than for AR images. DR offers improved image quality with a significant reduction of up to 23.6% in radiation dose and more accurate pneumoconiosis classification than AR.

Proficiency in reading pneumoconiosis radiographs examined by the 60-film set with 4-factor structuring 8-index.

29 physicians (A1-Group) and 24 physicians (A2-Group) attending the 1st and 2nd "Asian Intensive Reader of Pneumoconiosis" (AIR Pneumo) training course, respectively, and 22 physicians (B-Group) attending the Brazilian training course took the examination of reading the 60-film set. The objective of the study was firstly to investigate the factor structure of physicians' proficiency of reading pneumoconiosis chest X-ray, and secondly to examine differences in factor scores between groups. Reading results in terms of the 8-index of all examinees (Examinee Group) were subjected to the exploratory factor analysis. A 4-factor was analyzed to structure the 8-index: the specificity for pneumoconiosis, specificity for large opacities, specificity for pleural plaque and shape differentiation for small opacities loaded on the Factor 1; the sensitivity for pneumoconiosis and sensitivity for large opacities loaded on the Factor 2; the sensitivity for pleural plaque loaded on the Factor 3; the profusion increment consistency loaded on the Factor 4. 4-Factor scores were compared between each other of the three groups. The Factor 2 scores in A1 and A2 groups were significantly higher than in B-Group. Four factors could reflect four aspects of reading proficiency of pneumoconiosis X-ray, and it was suggested that 4-factor scores could also assess the attained skills appropriately.

Comparison of digital with film radiographs for the classification of pneumoconiotic pleural abnormalities.

RATIONALE AND OBJECTIVES: Analog film radiographs are typically used to classify pneumoconiosis to allow comparison with standard film radiographs. The aim of this study was to determine if digital radiography is comparable to film for the purpose of classifying pneumoconiotic pleural abnormalities. MATERIALS AND METHODS: Subjects were 200 asbestos-exposed patients, from whom digital and film chest radiographs were obtained along with chest high-resolution computed tomographic scans. Using a crossover design, radiographs were independently read on two occasions by seven readers, using conventional International Labour Organization standards for film and digitized standards for digital. High-resolution computed tomographic scans were read independently by three readers. Areas under the receiver-operating characteristic curves were calculated using high-resolution computed tomographic ratings as the gold standard for disease status. Mixed linear models were fit to estimate the effects of order of presentation, occasion, and modality, treating the seven readers as a random effect. Comparing digital and film radiography for each reader and occasion, crude agreement and agreement beyond chance (κ) were also calculated. RESULTS: The linear models showed no statistically significant sequence effect for order of presentation (P = .73) or occasion (P = .28). Most important, the difference between modalities was not statistically significant (digital vs film, P = .54). The mean area under the curve for film was 0.736 and increased slightly to 0.741 for digital. Mean crude agreement for the presence of pleural abnormalities consistent with pneumoconiosis across all readers and occasions was 78.3%, while the mean κ value was 0.49. CONCLUSIONS: These results indicate that digital radiography is not statistically different from analog film for the purpose of classifying pneumoconiotic pleural abnormalities, when appropriate standards are used.

[Computerized classification of pneumoconiosis radiographs based on grey level co-occurrence matrices].

Pneumoconiosis is diagnosed as categories 0-4 according to the Pneumoconiosis Law. Physicians have difficulty precisely categorizing many chest images. Therefore, we have developed a computerized method for automatically categorizing pneumoconiosis from chest radiographs. First, we extracted the rib edge regions from lung ROIs. Second, texture features were extracted using a dot enhancement filter, line enhancement filter, and grey level co-occurrence matrix. Third, the rib edge regions were removed from these processed images. Finally, we used a support vector machine for feature analysis. In a consistency test, 56 cases (69.7%) were classified correctly, and 45 cases (61.8%) were classified correctly in a validation test. These results show that the proposed features and removal of the rib edge are effective in classifying the profusion of opacities that indicate pneumoconiosis.

[On parity in development of obstructive and restrictive changes in miners].

Spirometric studies in 397 miners having pneumoconiosis and chronic dust bronchitis demonstrated 80% occurrence of respiratory failure with equally marked obstructive and restrictive disorders. Therefore, bronchopulmonary diseases in miners do not match the criteria of chronic obstrictive lung disease.

[Lung ventilation parameters in various pneumoconiosis forms during follow-up in miners of Kouzbass].

Studies of spirometric parameters in 120 miners who live in Kemerovo region and suffer from diagnosed pneumoconiosis demonstrated disordered pulmonary ventilation in 53.3% of the examinees. The highest velocity parameters in spirometry are seen in nodular form, the lowest ones--in mixed form of pneumoconiosis. During 2 years of follow-up forced expiratory volume over the first second decreased. Velocity and volume spirometric parameters in pneumoconiosis among coal miners of Kemerovo region under 2 years of follow-up demonstrate invert correlation with initial age of the patients when pneumoconiosis was diagnosed.

Association of cytokine gene polymorphisms in CWP and its severity in Turkish coal workers.

BACKGROUND: Cytokines appear to play a key role in some inflammatory reactions affecting the interactions among pro- and anti-inflammatory mechanisms that result in several diseases such as coal workers' pneumoconiosis (CWP). In this study, to determine the cytokine gene profiles of Turkish coal miners, we performed genotyping analysis to investigate the polymorphisms of CWP-related pro-inflammatory (TNFA, IL1A, IL1B, and IL6) and anti-inflammatory cytokines (IL-1RN and TGFB1). An additional goal was to observe whether these cytokine gene polymorphisms influence the development risk and severity of. METHODS: Genotyping was carried out by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) technique. RESULTS: TNFA (-238) gene polymorphism principally affected CWP development and severity (OR = 3.47: 95% CI, 1.12-10.77 and OR = 4.30: 95% CI, 1.25-14.74, respectively) and also risk of CWP (OR = 3.79: 95% CI, 1.37-10.46). The TNFA (-308) variant was associated with a risk for the CWP severity (OR = 2.84: 95% CI, 1.08-7.39). A protective effect of IL6 was found on the development (OR = 0.48: 95% CI, 0.21-0.93) and severity of CWP (OR = 0.37: 95% CI, 0.15-0.91). CONCLUSIONS: We suggest that TNFA (-238) variant may be a risk factor in both development and the severity of CWP, while TNFA (-308) variant seems to be important only in disease severity. On the other hand, IL6 variant may have a protective effect on the development and disease severity.

Outras pneumoconioses / Other pneumoconioses

As pneumoconioses mais prevalentes são a silicose, a asbestose e a pneumoconiose do trabalhador do carvão. Outras pneumoconioses com repercussões clínicas, funcionais e estruturais distintas são causadas pela inalação de poeiras metálicas a partir de fumos metálicos e sais orgânicos. A distinção quanto à forma química do composto inalado tem relação com a reação tecidual e o prognóstico. São apresentadas de forma sucinta a pneumoconiose simples, a siderose, a pneumoconiose por rocha fosfática, e a doença pulmonar crônica pelo berílio e por exposição a metais duros. Uma anamnese ocupacional como instrumento de busca etiológica dessas pneumoconioses é essencial.

The most prevalent pneumoconioses are silicosis, asbestosis and coal worker's pneumoconiosis. Other pneumoconioses that have distinct clinical, functional and structural repercussions are caused by inhalation of metal powder in fumes from metals or organic salts. The distinction in terms of the chemical form of the inhaled compound is related to the tissue reaction and to the prognosis. Simple pneumoconiosis, siderosis, berylliosis and phosphate rock-related pneumoconiosis, as well as chronic obstructive pulmonary disease caused by exposure to heavy metals, are succinctly discussed. As an instrument of etiologic investigation of these pneumoconioses, the taking of occupational histories is essential.

Selection of reference films based on reliability assessment of a classification of high-resolution computed tomography for pneumoconioses.

OBJECTIVE: Worldwide demand has increased for the development of a computed tomography (CT) classification system that supplements the ILO classification of radiographs for pneumoconioses. The authors aimed to show preliminary reliability test results for selected referent films for the CT classification system developed through an international effort by researchers from seven countries. METHODS: Reading trials by eight physicians who have considerable experience in pneumoconioses using a total of 114 lung zones consisting of 6 lung zones of 19 CT films of dust-exposed workers were performed to assess reliability of the classification system by weighted kappa. The results were also utilized for selecting reference films. RESULTS: A good agreement was observed for both first and second reading trials for rounded opacities (weighted kappa=0.76, 0.74, first and second trial results, respectively), irregular opacities (0.60, 0.48), emphysema (0.56, 0.70) and honeycombing (0.72, 0.79). Ground glass opacities, on the other hand, showed moderate agreement (0.43, 0.38). Intra-reader agreements among eight readers were shown in the same table as the mean and standard deviation of weighted kappa statistics. The inter-reader agreement for pleural thickening was not as good as for parenchymal lesions. DISCUSSION: The CT classification development may pioneer noble and sensitive medical screening for dust-exposed workers in selected settings. This system may be applied to radiographic borderline cases of profusion 0/1 and 1/0 by the ILO classification, in a setting that assures the occupational safety and health of workers exposed to some newly developed chemical compounds.