Diagnostic accuracy of chest X-ray and CT using artificial intelligence for osteoporosis: systematic review and meta-analysis.

INTRODUCTION: Artificial intelligence (AI)-based systems using chest images are potentially reliable for diagnosing osteoporosis. METHODS: We performed a systematic review and meta-analysis to assess the diagnostic accuracy of chest X-ray and computed tomography (CT) scans using AI for osteoporosis in accordance with the diagnostic test accuracy guidelines. We included any type of study investigating the diagnostic accuracy of index test for osteoporosis. We searched MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials, and IEEE Xplore Digital Library on November 8, 2023. The main outcome measures were the sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) for osteoporosis and osteopenia. We described forest plots for sensitivity, specificity, and AUC. The summary points were estimated from the bivariate random-effects models. We summarized the overall quality of evidence using the Grades of Recommendation, Assessment, Development, and Evaluation approach. RESULTS: Nine studies with 11,369 participants were included in this review. The pooled sensitivity, specificity, and AUC of chest X-rays for the diagnosis of osteoporosis were 0.83 (95% confidence interval [CI] 0.75, 0.89), 0.76 (95% CI 0.71, 0.80), and 0.86 (95% CI 0.83, 0.89), respectively (certainty of the evidence, low). The pooled sensitivity and specificity of chest CT for the diagnosis of osteoporosis and osteopenia were 0.83 (95% CI 0.69, 0.92) and 0.70 (95% CI 0.61, 0.77), respectively (certainty of the evidence, low and very low). CONCLUSIONS: This review suggests that chest X-ray with AI has a high sensitivity for the diagnosis of osteoporosis, highlighting its potential for opportunistic screening. However, the risk of bias of patient selection in most studies were high. More research with adequate participants' selection criteria for screening tool will be needed in the future.

Hallazgos imagenológicos torácicos en seguimiento de pacientes con síndrome PostCovid-19 / Thoracic imagenological findings in follow-up of patients with Post-Covid-19 syndrome

Objetivo: Describir los hallazgos imagenológicos en radiografías de tórax y ecografías pulmonares de pacientes con síndrome post-COVID-19. Métodos: estudio descriptivo, prospectivo y transversal que incluyó pacientes con síndrome post-COVID-19, sometidos a radiografías de tórax y ecografías pulmonares en el Servicio de Neumonología Clínica del Hospital Dr. José Ignacio Baldo, entre enero y octubre de 2022, con la finalidad de establecer su evolución imagenológica pulmonar. Se utilizó estadística descriptiva, chi-cuadrado de Pearson y prueba kappa de concordancia, considerando significativo un valor de p < 0,05. Resultados: La muestra consistió en 58 pacientes con una edad media de 55 ± 13 años, predominando el sexo femenino (58,6%). El 60,3% mostró alteraciones en la radiografía de tórax; un 74,3% con patrón intersticial bilateral y un 25,7% con patrón intersticial unilateral. La ecografía reveló patrón intersticial en el 43,1% de los casos y se observaron dos microconsolidaciones subpleurales. Conclusiones: Las radiografías de tórax y las ecografías pulmonares son herramientas imagenológicas eficaces, accesibles y económicas para detectar alteraciones en pacientes con síndrome post-COVID-19. (AU)

Objective: To describe imaging findings in chest radiographs and lung ultrasounds of patients with post-COVID-19 syndrome. Methods: A descriptive, prospective, and cross-sectional study was carried out that included patients with post-COVID-19 syndrome, who underwent chest radiographs and lung ultrasounds at the Clinical Pneumonology Service of Dr. José Ignacio Baldo Hospital, between January and October 2022. Descriptive statistics, Pearson's chi-square, and kappa concordance test were used, considering a p-value < 0.05 significant. Results: The sample consisted of 58 patients with an average age of 55 ± 13 years, with a predominance of females (58.6%). 60.3% showed alterations in the chest radiograph; 74.3% with a bilateral interstitial pattern and 25.7% with a unilateral interstitial pattern. The ultrasound revealed an interstitial pattern in 43.1% of the cases and two subpleural microconsolidations were observed. Conclusions: Chest radiographs and lung ultrasounds are effective, accessible, and economical imaging tools to detect alterations in patients with post-COVID-19 syndrome. (AU)

Early Detection of Heart Failure with Autonomous AI-Based Model Using Chest Radiographs: A Multicenter Study.

The opportunistic use of radiological examinations for disease detection can potentially enable timely management. We assessed if an index created by an AI software to quantify chest radiography (CXR) findings associated with heart failure (HF) could distinguish between patients who would develop HF or not within a year of the examination. Our multicenter retrospective study included patients who underwent CXR without an HF diagnosis. We included 1117 patients (age 67.6 ± 13 years; m:f 487:630) that underwent CXR. A total of 413 patients had the CXR image taken within one year of their HF diagnosis. The rest (n = 704) were patients without an HF diagnosis after the examination date. All CXR images were processed with the model (qXR-HF, Qure.AI) to obtain information on cardiac silhouette, pleural effusion, and the index. We calculated the accuracy, sensitivity, specificity, and area under the curve (AUC) of the index to distinguish patients who developed HF within a year of the CXR and those who did not. We report an AUC of 0.798 (95%CI 0.77-0.82), accuracy of 0.73, sensitivity of 0.81, and specificity of 0.68 for the overall AI performance. AI AUCs by lead time to diagnosis (<3 months: 0.85; 4-6 months: 0.82; 7-9 months: 0.75; 10-12 months: 0.71), accuracy (0.68-0.72), and specificity (0.68) remained stable. Our results support the ongoing investigation efforts for opportunistic screening in radiology.

Deep learning for tubes and lines detection in critical illness: Generalizability and comparison with residents.

Background: Artificial intelligence (AI) has been proven useful for the assessment of tubes and lines on chest radiographs of general patients. However, validation on intensive care unit (ICU) patients remains imperative. Methods: This retrospective case-control study evaluated the performance of deep learning (DL) models for tubes and lines classification on both an external public dataset and a local dataset comprising 303 films randomly sampled from the ICU database. The endotracheal tubes (ETTs), central venous catheters (CVCs), and nasogastric tubes (NGTs) were classified into "Normal," "Abnormal," or "Borderline" positions by DL models with and without rule-based modification. Their performance was evaluated using an experienced radiologist as the standard reference. Results: The algorithm showed decreased performance on the local ICU dataset, compared to that of the external dataset, decreasing from the Area Under the Curve of Receiver (AUC) of 0.967 (95 % CI 0.965-0.973) to the AUC of 0.70 (95 % CI 0.68-0.77). Significant improvement in the ETT classification task was observed after modifications were made to the model to allow the use of the spatial relationship between line tips and reference anatomy with the improvement of the AUC, increasing from 0.71 (95 % CI 0.70 - 0.75) to 0.86 (95 % CI 0.83 - 0.94). Conclusions: The externally trained model exhibited limited generalizability on the local ICU dataset. Therefore, evaluating the performance of externally trained AI before integrating it into critical care routine is crucial. Rule-based algorithm may be used in combination with DL to improve results.

Chest X-Ray-Based Telemedicine Platform for Pediatric Tuberculosis Diagnosis in Low-Resource Settings: Development and Validation Study.

BACKGROUND: Tuberculosis (TB) remains a major cause of morbidity and death worldwide, with a significant impact on children, especially those under the age of 5 years. The complex diagnosis of pediatric TB, compounded by limited access to more accurate diagnostic tests, underscores the need for improved tools to enhance diagnosis and care in resource-limited settings. OBJECTIVE: This study aims to present a telemedicine web platform, BITScreen PTB (Biomedical Image Technologies Screen for Pediatric Tuberculosis), aimed at improving the evaluation of pulmonary TB in children based on digital chest x-ray (CXR) imaging and clinical information in resource-limited settings. METHODS: The platform was evaluated by 3 independent expert readers through a retrospective assessment of a data set with 218 imaging examinations of children under 3 years of age, selected from a previous study performed in Mozambique. The key aspects assessed were the usability through a standardized questionnaire, the time needed to complete the assessment through the platform, the performance of the readers to identify TB cases based on the CXR, the association between the TB features identified in the CXRs and the initial diagnostic classification, and the interreader agreement of the global assessment and the radiological findings. RESULTS: The platform's usability and user satisfaction were evaluated using a questionnaire, which received an average rating of 4.4 (SD 0.59) out of 5. The average examination completion time ranged from 35 to 110 seconds. In addition, the study on CXR showed low sensitivity (16.3%-28.2%) but high specificity (91.1%-98.2%) in the assessment of the consensus case definition of pediatric TB using the platform. The CXR finding having a stronger association with the initial diagnostic classification was air space opacification (χ21>20.38, P<.001). The study found varying levels of interreader agreement, with moderate/substantial agreement for air space opacification (κ=0.54-0.67) and pleural effusion (κ=0.43-0.72). CONCLUSIONS: Our findings support the promising role of telemedicine platforms such as BITScreen PTB in enhancing pediatric TB diagnosis access, particularly in resource-limited settings. Additionally, these platforms could facilitate the multireader and systematic assessment of CXR in pediatric TB clinical studies.

Imaging the Lung in ARDS: A Primer.

Despite periodic changes in the clinical definition of ARDS, imaging of the lung remains a central component of its diagnostic identification. Several imaging modalities are available to the clinician to establish a diagnosis of the syndrome, monitor its clinical course, or assess the impact of treatment and management strategies. Each imaging modality provides unique insight into ARDS from structural and/or functional perspectives. This review will highlight several methods for lung imaging in ARDS, emphasizing basic operational and physical principles for the respiratory therapist. Advantages and disadvantages of each modality will be discussed in the context of their utility for clinical management and decision-making.

Deep Learning in Cardiothoracic Ratio Calculation and Cardiomegaly Detection.

Objectives: The purpose of this study is to evaluate the performance of our deep learning algorithm in calculating cardiothoracic ratio (CTR) and thus in the assessment of cardiomegaly or pericardial effusion occurrences on chest radiography (CXR). Methods: From a database of 8000 CXRs, 13 folders with a comparable number of images were created. Then, 1020 images were chosen randomly, in proportion to the number of images in each folder. Afterward, CTR was calculated using RadiAnt Digital Imaging and Communications in Medicine (DICOM) Viewer software (2023.1). Next, heart and lung anatomical areas were marked in 3D Slicer. From these data, we trained an AI model which segmented heart and lung anatomy and determined the CTR value. Results: Our model achieved an Intersection over Union metric of 88.28% for the augmented training subset and 83.06% for the validation subset. F1-score for subsets were accordingly 90.22% and 90.67%. In the comparative analysis of artificial intelligence (AI) vs. humans, significantly lower transverse thoracic diameter (TTD) (p < 0.001), transverse cardiac diameter (TCD) (p < 0.001), and CTR (p < 0.001) values obtained using the neural network were observed. Conclusions: Results confirm that there is a significant correlation between the measurements made by human observers and the neural network. After validation in clinical conditions, our method may be used as a screening test or advisory tool when a specialist is not available, especially on Intensive Care Units (ICUs) or Emergency Departments (ERs) where time plays a key role.

Detection of pulmonary nodules in chest radiographs: novel cost function for effective network training with purely synthesized datasets.

PURPOSE: Many large radiographic datasets of lung nodules are available, but the small and hard-to-detect nodules are rarely validated by computed tomography. Such difficult nodules are crucial for training nodule detection methods. This lack of difficult nodules for training can be addressed by artificial nodule synthesis algorithms, which can create artificially embedded nodules. This study aimed to develop and evaluate a novel cost function for training networks to detect such lesions. Embedding artificial lesions in healthy medical images is effective when positive cases are insufficient for network training. Although this approach provides both positive (lesion-embedded) images and the corresponding negative (lesion-free) images, no known methods effectively use these pairs for training. This paper presents a novel cost function for segmentation-based detection networks when positive-negative pairs are available. METHODS: Based on the classic U-Net, new terms were added to the original Dice loss for reducing false positives and the contrastive learning of diseased regions in the image pairs. The experimental network was trained and evaluated, respectively, on 131,072 fully synthesized pairs of images simulating lung cancer and real chest X-ray images from the Japanese Society of Radiological Technology dataset. RESULTS: The proposed method outperformed RetinaNet and a single-shot multibox detector. The sensitivities were 0.688 and 0.507 when the number of false positives per image was 0.2, respectively, with and without fine-tuning under the leave-one-case-out setting. CONCLUSION: To our knowledge, this is the first study in which a method for detecting pulmonary nodules in chest X-ray images was evaluated on a real clinical dataset after being trained on fully synthesized images. The synthesized dataset is available at https://zenodo.org/records/10648433 .

Deep Learning-Based Localization and Detection of Malpositioned Nasogastric Tubes on Portable Supine Chest X-Rays in Intensive Care and Emergency Medicine: A Multi-center Retrospective Study.

Malposition of a nasogastric tube (NGT) can lead to severe complications. We aimed to develop a computer-aided detection (CAD) system to localize NGTs and detect NGT malposition on portable chest X-rays (CXRs). A total of 7378 portable CXRs were retrospectively retrieved from two hospitals between 2015 and 2020. All CXRs were annotated with pixel-level labels for NGT localization and image-level labels for NGT presence and malposition. In the CAD system, DeepLabv3 + with backbone ResNeSt50 and DenseNet121 served as the model architecture for segmentation and classification models, respectively. The CAD system was tested on images from chronologically different datasets (National Taiwan University Hospital (National Taiwan University Hospital)-20), geographically different datasets (National Taiwan University Hospital-Yunlin Branch (YB)), and the public CLiP dataset. For the segmentation model, the Dice coefficients indicated accurate delineation of the NGT course (National Taiwan University Hospital-20: 0.665, 95% confidence interval (CI) 0.630-0.696; National Taiwan University Hospital-Yunlin Branch: 0.646, 95% CI 0.614-0.678). The distance between the predicted and ground-truth NGT tips suggested accurate tip localization (National Taiwan University Hospital-20: 1.64 cm, 95% CI 0.99-2.41; National Taiwan University Hospital-Yunlin Branch: 2.83 cm, 95% CI 1.94-3.76). For the classification model, NGT presence was detected with high accuracy (area under the receiver operating characteristic curve (AUC): National Taiwan University Hospital-20: 0.998, 95% CI 0.995-1.000; National Taiwan University Hospital-Yunlin Branch: 0.998, 95% CI 0.995-1.000; CLiP dataset: 0.991, 95% CI 0.990-0.992). The CAD system also detected NGT malposition with high accuracy (AUC: National Taiwan University Hospital-20: 0.964, 95% CI 0.917-1.000; National Taiwan University Hospital-Yunlin Branch: 0.991, 95% CI 0.970-1.000) and detected abnormal nasoenteric tube positions with favorable performance (AUC: 0.839, 95% CI 0.807-0.869). The CAD system accurately localized NGTs and detected NGT malposition, demonstrating excellent potential for external generalizability.

Development and Validation of an Artificial Intelligence Model for Detecting Rib Fractures on Chest Radiographs.

Background: Chest radiography is the standard method for detecting rib fractures. Our study aims to develop an artificial intelligence (AI) model that, with only a relatively small amount of training data, can identify rib fractures on chest radiographs and accurately mark their precise locations, thereby achieving a diagnostic accuracy comparable to that of medical professionals. Methods: For this retrospective study, we developed an AI model using 540 chest radiographs (270 normal and 270 with rib fractures) labeled for use with Detectron2 which incorporates a faster region-based convolutional neural network (R-CNN) enhanced with a feature pyramid network (FPN). The model's ability to classify radiographs and detect rib fractures was assessed. Furthermore, we compared the model's performance to that of 12 physicians, including six board-certified anesthesiologists and six residents, through an observer performance test. Results: Regarding the radiographic classification performance of the AI model, the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were 0.87, 0.83, and 0.89, respectively. In terms of rib fracture detection performance, the sensitivity, false-positive rate, and free-response receiver operating characteristic (JAFROC) figure of merit (FOM) were 0.62, 0.3, and 0.76, respectively. The AI model showed no statistically significant difference in the observer performance test compared to 11 of 12 and 10 of 12 physicians, respectively. Conclusions: We developed an AI model trained on a limited dataset that demonstrated a rib fracture classification and detection performance comparable to that of an experienced physician.

Unusual Air Crescents on Chest Radiographs.

The air crescent (AC) is a common radiological sign. Even if its commonest aetiology remains pulmonary aspergillosis, various other causes have been described. In this study, we report four rare causes of ACs seen on chest radiographs that haven't been described in the literature. Teaching point: The differential diagnosis of an air crescent sign on chest radiographs includes oesophageal bezoar, interstitial lung emphysema, central bronchial stenosis and perforated emphysematous cholecystitis.

Low-Level Respirable Crystalline Silica and Silicosis: Long-Term Follow-Up of Vermont Granite Workers.

The lifetime risk of silicosis associated with low-level occupational exposure to respirable crystalline silica remains unclear because most previous radiographic studies included workers with varying exposure concentrations and durations. This study assessed the prevalence of silicosis after lengthy exposure to respirable crystalline silica at levels ≤ 0.10 mg/m3. Vermont granite workers employed any time during 1979-1987 were traced and chest radiographs were obtained for 356 who were alive in 2017 and residing in Vermont. Work history, smoking habits and respiratory symptoms were obtained by interview, and exposure was estimated using a previously developed job-exposure matrix. Associations between radiographic findings, exposure, and respiratory symptoms were assessed by ANOVA, chi-square tests and binary regression. Fourteen workers (3.9%) had radiographic evidence of silicosis, and all had been employed ≥30 years. They were more likely to have been stone cutters or carvers and their average exposure concentrations and cumulative exposures to respirable crystalline silica were significantly higher than workers with similar durations of employment and no classifiable parenchymal abnormalities. This provides direct evidence that workers with long-term exposure to low-level respirable crystalline silica (≤0.10 mg/m3) are at risk of developing silicosis.

The chest X-ray score baseline in predicting continuous oxygen therapy failure in low-risk aged patients after thoracic surgery.

Background: Radiographic severity assessment can be instrumental in diagnosing postoperative pulmonary complications (PPCs) and guiding oxygen therapy. The radiographic assessment of lung edema (RALE) and Brixia scores correlate with disease severity, but research on low-risk elderly patients is lacking. This study aimed to assess the efficacy of two chest X-ray scores in predicting continuous oxygen therapy (COT) treatment failure in patients over 70 years of age after thoracic surgery. Methods: From January 2019 to December 2021, we searched for patients aged 70 years and above who underwent thoracic surgery and received COT treatment, with a focus on those at low risk of respiratory complications. Bedside chest X-rays, RALE, Brixia scores, and patient data were collected. Univariate, multivariate analyses, and 1:2 matching identified risk factors. Receiver operating characteristic (ROC) curves determined score sensitivity, specificity, and predictive values. Results: Among the 242 patients surviving to discharge, 19 (7.9%) patients experienced COT failure. COT failure correlated with esophageal cancer surgeries, thoracotomies (36.8% vs. 9%, P=0.003; 26.3% vs. 9.4%, P=0.004), and longer operation time (3.4 vs. 2.8 h, P=0.003). Surgical approach and RALE score were independent risk factors. The prediction model had an area under the curve (AUC) of 0.839 [95% confidence interval (CI), 0.740-0.938]. Brixia and RALE scores predicted COT failure with AUCs of 0.764 (95% CI, 0.650-0.878) with a cut-off value of 6.027 and 0.710 (95% CI, 0.588-0.832) with a cut-off value of 17.134, respectively, after 1:2 matching. Conclusions: The RALE score predict the risk of COT failure in elderly, low-risk thoracic patients better than the Brixia score. This simple, cheap, and noninvasive method helps evaluate postoperative lung damage, monitor treatment response, and provide early warning for oxygen therapy escalation. Further studies are required to confirm the validity and applicability of this model in different settings and populations.

Doctor's Orders-Why Radiologists Should Consider Adjusting Commercial Machine Learning Applications in Chest Radiography to Fit Their Specific Needs.

This retrospective study evaluated a commercial deep learning (DL) software for chest radiographs and explored its performance in different scenarios. A total of 477 patients (284 male, 193 female, mean age 61.4 (44.7-78.1) years) were included. For the reference standard, two radiologists performed independent readings on seven diseases, thus reporting 226 findings in 167 patients. An autonomous DL reading was performed separately and evaluated against the gold standard regarding accuracy, sensitivity and specificity using ROC analysis. The overall average AUC was 0.84 (95%-CI 0.76-0.92) with an optimized DL sensitivity of 85% and specificity of 75.4%. The best results were seen in pleural effusion with an AUC of 0.92 (0.885-0.955) and sensitivity and specificity of each 86.4%. The data also showed a significant influence of sex, age, and comorbidity on the level of agreement between gold standard and DL reading. About 40% of cases could be ruled out correctly when screening for only one specific disease with a sensitivity above 95% in the exploratory analysis. For the combined reading of all abnormalities at once, only marginal workload reduction could be achieved due to insufficient specificity. DL applications like this one bear the prospect of autonomous comprehensive reporting on chest radiographs but for now require human supervision. Radiologists need to consider possible bias in certain patient groups, e.g., elderly and women. By adjusting their threshold values, commercial DL applications could already be deployed for a variety of tasks, e.g., ruling out certain conditions in screening scenarios and offering high potential for workload reduction.

Respiratory symptoms due to a twisted nasogastric tube: A case report.

Insertion of a nasogastric tube (NGT) is generally considered safe; however, it is not without risk, and in cases of misplacement, complications and even death may occur. In this article, we reported a case of NGT misplacement in a 75-year-old male, which resulted in aspiration pneumonia. We also reviewed published cases of NGT misplacement. Clinicians should pay enough attention to the confirmation of the proper placement of an NGT. A systematic approach for NGT insertion and confirmation is required to prevent misplacement.

Utility of artificial intelligence for detection of pneumothorax on chest radiopgraphs done after transthoracic percutaneous transthoracic biopsy guided by computed tomography.

OBJETIVE: To assess the ability of an artificial intelligence software to detect pneumothorax in chest radiographs done after percutaneous transthoracic biopsy. MATERIAL AND METHODS: We included retrospectively in our study adult patients who underwent CT-guided percutaneous transthoracic biopsies from lung, pleural or mediastinal lesions from June 2019 to June 2020, and who had a follow-up chest radiograph after the procedure. These chest radiographs were read to search the presence of pneumothorax independently by an expert thoracic radiologist and a radiodiagnosis resident, whose unified lecture was defined as the gold standard, and the result of each radiograph after interpretation by the artificial intelligence software was documented for posterior comparison with the gold standard. RESULTS: A total of 284 chest radiographs were included in the study and the incidence of pneumothorax was 14.4%. There were no discrepancies between the two readers' interpretation of any of the postbiopsy chest radiographs. The artificial intelligence software was able to detect 41/41 of the present pneumothorax, implying a sensitivity of 100% and a negative predictive value of 100%, with a specificity of 79.4% and a positive predictive value of 45%. The accuracy was 82.4%, indicating that there is a high probability that an individual will be adequately classified by the software. It has also been documented that the presence of Port-a-cath is the cause of 8 of the 50 of false positives by the software. CONCLUSIONS: The software has detected 100% of cases of pneumothorax in the postbiopsy chest radiographs. A potential use of this software could be as a prioritisation tool, allowing radiologists not to read immediately (or even not to read) chest radiographs classified as non-pathological by the software, with the confidence that there are no pathological cases.

Assessment of the comparative agreement between chest radiographs and CT scans in intensive care units.

PURPOSE: Chest radiographs in critically ill patients can be difficult to interpret due to technical and clinical factors. We sought to determine the agreement of chest radiographs and CT scans, and the inter-observer variation of chest radiograph interpretation, in intensive care units (ICUs). METHODS: Chest radiographs and corresponding thoracic computerised tomography (CT) scans (as reference standard) were collected from 45 ICU patients. All radiographs were analysed by 20 doctors (radiology consultants, radiology trainees, ICU consultants, ICU trainees) from 4 different centres, blinded to CT results. Specificity/sensitivity were determined for pleural effusion, lobar collapse and consolidation/atelectasis. Separately, Fleiss' kappa for multiple raters was used to determine inter-observer variation for chest radiographs. RESULTS: The median sensitivity and specificity of chest radiographs for detecting abnormalities seen on CTs scans were 43.2% and 85.9% respectively. Diagnostic sensitivity for pleural effusion was significantly higher among radiology consultants but no specialty/experience distinctions were observed for specificity. Median inter-observer kappa coefficient among assessors was 0.295 ("fair"). CONCLUSIONS: Chest radiographs commonly miss important radiological features in critically ill patients. Inter-observer agreement in chest radiograph interpretation is only "fair". Consultant radiologists are least likely to miss thoracic radiological abnormalities. The consequences of misdiagnosis by chest radiographs remain to be determined.

Deep convolutional network-based chest radiographs screening model for pneumoconiosis.

Background: Pneumoconiosis is the most important occupational disease all over the world, with high prevalence and mortality. At present, the monitoring of workers exposed to dust and the diagnosis of pneumoconiosis rely on manual interpretation of chest radiographs, which is subjective and low efficiency. With the development of artificial intelligence technology, a more objective and efficient computer aided system for pneumoconiosis diagnosis can be realized. Therefore, the present study reported a novel deep learning (DL) artificial intelligence (AI) system for detecting pneumoconiosis in digital frontal chest radiographs, based on which we aimed to provide references for radiologists. Methods: We annotated 49,872 chest radiographs from patients with pneumoconiosis and workers exposed to dust using a self-developed tool. Next, we used the labeled images to train a convolutional neural network (CNN) algorithm developed for pneumoconiosis screening. Finally, the performance of the trained pneumoconiosis screening model was validated using a validation set containing 495 chest radiographs. Results: Approximately, 51% (25,435/49,872) of the chest radiographs were labeled as normal. Pneumoconiosis was detected in 49% (24,437/49,872) of the labeled radiographs, among which category-1, category-2, and category-3 pneumoconiosis accounted for 53.1% (12,967/24,437), 20.4% (4,987/24,437), and 26.5% (6,483/24,437) of the patients, respectively. The CNN DL algorithm was trained using these data. The validation set of 495 digital radiography chest radiographs included 261 cases of pneumoconiosis and 234 cases of non-pneumoconiosis. As a result, the accuracy of the AI system for pneumoconiosis identification was 95%, the area under the curve was 94.7%, and the sensitivity was 100%. Conclusion: DL algorithm based on CNN helped screen pneumoconiosis in the chest radiographs with high performance; thus, it could be suitable for diagnosing pneumoconiosis automatically and improve the efficiency of radiologists.