Decision processes and observer error in the diagnosis of pneumoconiosis by chest roentgenography

Potential of digital chest radiography-based deep learning in screening and diagnosing pneumoconiosis: An observational study

The diagnosis of pneumoconiosis is complex and subjective, leading to inevitable variability in readings. This is especially true for inexperienced doctors. To improve accuracy, a computer-assisted diagnosis system is used for more effective pneumoconiosis diagnoses. Three models (Resnet50, Resnet101, and DenseNet) were used for pneumoconiosis classification based on 1250 chest X-ray images. Three experienced and highly qualified physicians read the collected digital radiography images and classified them from category 0 to category III in a double-blinded manner. The results of the 3 physicians in agreement were considered the relative gold standards. Subsequently, 3 models were used to train and test these images and their performance was evaluated using multi-class classification metrics. We used kappa values and accuracy to evaluate the consistency and reliability of the optimal model with clinical typing. The results showed that ResNet101 was the optimal model among the 3 convolutional neural networks. The AUC of ResNet101 was 1.0, 0.9, 0.89, and 0.94 for detecting pneumoconiosis categories 0, I, II, and III, respectively. The micro-average and macro-average mean AUC values were 0.93 and 0.94, respectively. The accuracy and Kappa values of ResNet101 were 0.72 and 0.7111 for quadruple classification and 0.98 and 0.955 for dichotomous classification, respectively, compared with the relative standard classification of the clinic. This study develops a deep learning based model for screening and staging of pneumoconiosis is using chest radiographs. The ResNet101 model performed relatively better in classifying pneumoconiosis than radiologists. The dichotomous classification displayed outstanding performance, thereby indicating the feasibility of deep learning techniques in pneumoconiosis screening.

A deep learning-based model for screening and staging pneumoconiosis

This study aims to develop an artificial intelligence (AI)-based model to assist radiologists in pneumoconiosis screening and staging using chest radiographs. The model, based on chest radiographs, was developed using a training cohort and validated using an independent test cohort. Every image in the training and test datasets were labeled by experienced radiologists in a double-blinded fashion. The computational model started by segmenting the lung field into six subregions. Then, convolutional neural network classification model was used to predict the opacity level for each subregion respectively. Finally, the diagnosis for each subject (normal, stage I, II, or III pneumoconiosis) was determined by summarizing the subregion-based prediction results. For the independent test cohort, pneumoconiosis screening accuracy was 0.973, with both sensitivity and specificity greater than 0.97. The accuracy for pneumoconiosis staging was 0.927, better than that achieved by two groups of radiologists (0.87 and 0.84, respectively). This study develops a deep learning-based model for screening and staging of pneumoconiosis using man-annotated chest radiographs. The model outperformed two groups of radiologists in the accuracy of pneumoconiosis staging. This pioneer work demonstrates the feasibility and efficiency of AI-assisted radiography screening and diagnosis in occupational lung diseases.

NIOSH's Respiratory Health Division: 50 years of science and service

The year 2017 marked the 50th anniversary of NIOSH's Respiratory Health Division (RHD). RHD began in 1967 as the Appalachian Laboratory for Occupational Respiratory Diseases (ALFORD), with a focus on coal workers' pneumoconiosis. ALFORD became part of NIOSH in 1971 and added activities to address work-related respiratory disease more generally. Health hazard evaluations played an important role in understanding novel respiratory hazards such as nylon flock, diacetyl, and indium-tin oxide. Epidemiologic and laboratory studies addressed many respiratory hazards, including coal mine dust, silica, asbestos, cotton dust, beryllium, diesel exhaust, and dampness and mold. Surveillance activities tracked the burden of diseases and enhanced the quality of spirometry and chest radiography used to screen workers. RHD's efforts to improve scientific understanding, inform strategies for prevention, and disseminate knowledge remain important now and for the future.

Use of chest radiographs in epidemiological investigations of pneumoconioses

The International Labour Organisation (ILO) classification of radiographs of pneumoconioses was developed to limit variation in classification of parenchymal abnormalities. In this study the manner in which chest radiographs were interpreted in 134 investigations reported in four peer reviewed journals during the five year period 1985-90 was examined. The approach for applying the ILO system was poorly described in most studies. For example, of 86 investigations using more than one reader, 66.3% described the method of reconciliation, but methods were not consistent among investigations. Our results indicate a number of potential problems in application of the ILO system, and gaps in existing recommendations that should be considered.

Causes of error in gastrointestinal radiology. I. Upper gastrointestinal examination

This paper explores the causes of errors on upper gastrointestinal studies. Of 129 ulcers and neoplasms seen endoscopically, 14 were unreported radiologically, representing false negative errors. The most common type of error was failure to note a lesion rendered poorly visible by an inadequate examination. Contributing technical failures included insufficient films, poor coating, poor compression films, retained food and secretions, and badly exposed films. Perceptive failures included both misinterpretation and failure to note a lesion. Among 10 false positive errors, barium between distorted folds simulating an ulcer occurred most commonly.

Disagreements in chest roentgen interpretation

Each of 100 chest radiographs selected randomly from a hospital population were read by five experienced radiologists. The films were rich in abnormal findings. Disagreement analysis was performed considering the entire content of the interpretation. The disagreements were graded by a panel as to type (false negative, false positive, and indeterminate) and significance. The overall repeatability of the panel's performance was 75 percent. Forty-one percent of the reports contained potentially significant errors and 56 percent indeterminate disagreements. Among all errors 78 percent were false negatives and 22 percent were false positives. The error rate varied among readers. An average of 25 percent of important findings were omitted by an individual reader. An association between specific abnormalities and the types of disagreement was found.