X-ray Interpretation Research Articles

National cross-sectional cluster survey of tuberculosis prevalence in Timor-Leste: a study protocol

IntroductionTimor-Leste has one of the world’s highest estimated tuberculosis (TB) incidences, yet the data which informs this estimate is limited and the true burden of TB disease is not known....

A deep-learning-based framework for identifying and localizing multiple abnormalities and assessing cardiomegaly in chest X-ray

Accurate identification and localization of multiple abnormalities are crucial steps in the interpretation of chest X-rays (CXRs); however, the lack of a large CXR dataset with bounding boxes severely constrains accurate localization research based on deep learning. We created a large CXR dataset named CXR-AL14, containing 165,988 CXRs and 253,844 bounding boxes. On the basis of this dataset, a deep-learning-based framework was developed to identify and localize 14 common abnormalities and calculate the cardiothoracic ratio (CTR) simultaneously. The mean average precision values obtained by the model for 14 abnormalities reached 0.572-0.631 with an intersection-over-union threshold of 0.5, and the intraclass correlation coefficient of the CTR algorithm exceeded 0.95 on the held-out, multicentre and prospective test datasets. This framework shows an excellent performance, good generalization ability and strong clinical applicability, which is superior to senior radiologists and suitable for routine clinical settings.

Evaluation of a short training course of chest X-ray interpretation for the diagnosis of paediatric TB.

Chest X-ray (CXR) interpretation is challenging for the diagnosis of paediatric TB. We assessed the performance of a three half-day CXR training module for healthcare workers (HCWs) at low healthcare levels in six high TB incidence countries. Within the TB-Speed Decentralization Study, we developed a three half-day training course to identify normal CXR, CXR of good quality and identify six TB-suggestive features. We performed a pre-post training assessment on a pre-defined set of 20 CXR readings. We compared the proportion of correctly interpreted CXRs and the median reading score before and after the training using the McNemar test and a linear mixed model. Of 191 HCWs, 43 (23%) were physicians, 103 (54%) nurses, 18 (9.4%) radiology technicians and 12 (6.3%) other professionals. Of 2,840 CXRs with both assessment, respectively 1,843 (64.9%) and 2,277 (80.2%) were correctly interpreted during pre-training and post-training (P < 0.001). The median reading score improved significantly from 13/20 to 16/20 after the training, after adjusting by country, facility and profession (adjusted β = 3.31, 95% CI 2.44-4.47). Despite some limitations of the course assessment that did not include abnormal non-TB suggestive CXR, study findings suggest that a short CXR training course could improve HCWs' interpretation skills in diagnosing paediatric TB.

What is New in the Diagnosis of Childhood Tuberculosis?

The fact that almost half of the 1 million cases of childhood tuberculosis (TB)globally remain undiagnosed jeopardizes the TB elimination goal. Fortunately, there are new advances in this fieldwhich have the potential to bridge this diagnostic gap. Advances in imaging include computer assisted interpretation of chest X-rays (CXRs), point of care ultrasound (POCUS) and faster and superior computed tomography/ magnetic resonance imaging (CT/ MRI) protocols. The urine lipoarabinomannan test has proved to be a good point of care test for diagnosing TB in Human immunodeficiency virus (HIV) infected children. Stool and nasopharyngeal aspirates are emerging as acceptable alternatives for gastric lavage and induced sputum for diagnosing intrathoracic tuberculosis. Xpert MTB/RIF Ultra has improved sensitivitycompared to Xpert MTB/RIF for diagnosing both pulmonary/ extrapulmonary TB. Xpert XDR is another commercially available accurate point of care test for detecting resistance to drugs other than rifampicin in smear positive samples. Other molecular methods including new line probe assays, pyrosequencing, whole genome sequencing, and targeted next generation sequencing are extremely promising but not available commercially at present. The C-Tb skin test isan acceptable alternative to the tuberculin skin test and interferon gamma release assays for diagnosis of latent infection. There is an urgent need to incorporate some of these advances in the existingdiagnostic algorithms of childhood TB.

Clinical Application of Artificial Intelligence-Based Detection Assistance Devices for Chest X-Ray Interpretation: Current Status and Practical Considerations

Artificial intelligence (AI) technology is actively being applied for the interpretation of medical imaging, such as chest X-rays. AI-based software medical devices, which automatically detect various types of abnormal findings in chest X-ray images to assist physicians in their interpretation, are actively being commercialized and clinically implemented in Korea. Several important issues need to be considered for AI-based detection assistant tools to be applied in clinical practice: the evaluation of performance and efficacy prior to implementation; the determination of the target application, range, and method of delivering results; and monitoring after implementation and legal liability issues. Appropriate decision making regarding these devices based on the situation in each institution is necessary. Radiologists must be engaged as medical assessment experts using the software for these devices as well as in medical image interpretation to ensure the safe and efficient implementation and operation of AI-based detection assistant tools.

Validation of a Deep Learning Chest X-ray Interpretation Model: Integrating Large-Scale AI and Large Language Models for Comparative Analysis with ChatGPT.

This study evaluates the diagnostic accuracy and clinical utility of two artificial intelligence (AI) techniques: Kakao Brain Artificial Neural Network for Chest X-ray Reading (KARA-CXR), an assistive technology developed using large-scale AI and large language models (LLMs), and ChatGPT, a well-known LLM. The study was conducted to validate the performance of the two technologies in chest X-ray reading and explore their potential applications in the medical imaging diagnosis domain. The study methodology consisted of randomly selecting 2000 chest X-ray images from a single institution's patient database, and two radiologists evaluated the readings provided by KARA-CXR and ChatGPT. The study used five qualitative factors to evaluate the readings generated by each model: accuracy, false findings, location inaccuracies, count inaccuracies, and hallucinations. Statistical analysis showed that KARA-CXR achieved significantly higher diagnostic accuracy compared to ChatGPT. In the 'Acceptable' accuracy category, KARA-CXR was rated at 70.50% and 68.00% by two observers, while ChatGPT achieved 40.50% and 47.00%. Interobserver agreement was moderate for both systems, with KARA at 0.74 and GPT4 at 0.73. For 'False Findings', KARA-CXR scored 68.00% and 68.50%, while ChatGPT scored 37.00% for both observers, with high interobserver agreements of 0.96 for KARA and 0.97 for GPT4. In 'Location Inaccuracy' and 'Hallucinations', KARA-CXR outperformed ChatGPT with significant margins. KARA-CXR demonstrated a non-hallucination rate of 75%, which is significantly higher than ChatGPT's 38%. The interobserver agreement was high for KARA (0.91) and moderate to high for GPT4 (0.85) in the hallucination category. In conclusion, this study demonstrates the potential of AI and large-scale language models in medical imaging and diagnostics. It also shows that in the chest X-ray domain, KARA-CXR has relatively higher accuracy than ChatGPT.

Utility of Magnetic Resonance Imaging for Ligamentous Injury in Cervical Spine Trauma: A 2-Year Consecutive Case Cohort

Magnetic resonance imaging (MRI) is increasingly used as an adjunct to spinal soft tissue evaluation in cervical spine (C-spine) trauma; however, the utility of this information remains controversial. In this consecutive observational study, we reviewed the utility of MRI in patients with C-spine trauma. We identified patients in real time over a 2-year period as they presented to our level 1 trauma center for C-spine computed tomography (CT) scan followed by MRI. MRI was obtained by the trauma team prior to the spine service consultation if (1) they were unable to clear the C-spine according to protocol or (2) if the on-call radiologist reported a concern for ligamentous integrity from the CT findings. Thirty-three patients, including 19 males (58%) and 14 females, with a mean age of 54 years, were referred to the spine service for concerns of ligamentous instability. The most common mechanisms of injury were motor vehicle accidents (n= 13) and falls (n= 11). MRI demonstrated ligamentous signal change identified by the radiologist as potentially unstable in all patients. Fifteen patients (45%) had multiple C-spine ligaments affected. The interspinous ligament was involved most frequently (28%), followed by the ligamentum flavum (21%) and supraspinous ligament (15%). All patients underwent dynamic upright C-spine X-rays that were interpreted by both the ordering surgeon and radiologist. There was no evidence of instability in any patient; concurrence between X-ray interpretation was 100%. The cervical collar was successfully removed in all cases. No patients required late surgical intervention, and there were no return visits to the emergency department of a spinal nature. MRI signal change within the ligaments of the C-spine should be interpreted with caution in the setting of trauma. To physicians less familiar with spinal biomechanics, MRI findings may be perceived in an inadvertently alarming manner. Bony alignment and, when indicated, dynamic upright X-rays remain the gold standard for evaluating the ligamentous integrity of the C-spine.

Medical Interns' Skill Levels in Emergency Chest X-ray Interpretation

Objectives: The research aims to evaluate the proficiency of medical interns in identifying critical findings on chest X-rays, such as pneumothorax, pleural effusion, pulmonary edema, pneumonia, and other urgent conditions. It also seeks to determine the interns' capability to differentiate between normal and abnormal findings, as well as their ability to prioritize and communicate these findings to senior medical staff promptly. Materials and Methods: A cross-sectional online questionnaire-based form was used to collect data from medical interns across different medical schools around Saudi Arabia. Pearson correlation was carried out to see the relationship between x-ray interpretation scores and degree of confidence. A Fischer exact test was conducted to find an association between the demographic characteristics and diagnostic accuracy of X-rays. Results: A total of 545 medical interns from different universities participated in this study. The lowest diagnosis accuracy (27%) was when identifying a case of left lower lobe pneumonia. On the other hand, the most identified case (60.9%) was COVID-19 pneumonia. In cases of life-threatening conditions, only 13.1% of them have high confidence in their diagnosis of pneumoperitoneum, and 23.8% have high confidence in their diagnosis of pneumothorax. Normal CXR was answered incorrectly by 47.7% of the participants. The performance among recently started interns for some conditions is better than that of recently finished interns. Moreover, there are no significant differences in performance between those who are interested in radiology and those who are not. More than half of the participants advise doing extracurricular courses, practical case-based training in college, and volunteer work in radiology to improve their CXR interpretation skills. Conclusion: The competence of medical interns in interpreting chest X-rays varied across different medical conditions. Only 27.0% of interns correctly identified left lower lobe pneumonia, while 60.9% of them were able to identify COVID-19 pneumonia. Medical interns' competencies have limitations across different medical conditions; therefore, increasing exposure to radiological cases during internships and adequate training in college are recommended.

Deploying artificial intelligence software in an NHS trust: a how-to guide for clinicians.

Over the past 10 years, artificial intelligence (AI) has become one of the fastest-growing sectors in healthcare. There are now numerous new technologies designed to cut costs and improve diagnoses and treatment pathways. However, there is significant scepticism amongst National Health Service (NHS) staff regarding the usefulness of AI and it's cost to the NHS. This has likely resulted in underuse and slow adoption of software that may revolutionize our healthcare system and ensure its continued survival and effectiveness. Several governing bodies have put forward guidance on the safe and effective adoption of AI tools, but this rarely covers the reality of selecting and deploying new software. This article set out clear guidance on the practicalities and pitfalls of deploying digital solutions in healthcare, using the example of a deep learning algorithm designed to improve the accuracy of chest X-ray (CXR) interpretation in the emergency department.

Development and validation of open-source deep neural networks for comprehensive chest x-ray reading: a retrospective, multicentre study

Artificial intelligence (AI) systems for automated chest x-ray interpretation hold promise for standardising reporting and reducing delays in health systems with shortages of trained radiologists. Yet, there are few freely accessible AI systems trained on large datasets for practitioners to use with their own data with a view to accelerating clinical deployment of AI systems in radiology. We aimed to contribute an AI system for comprehensive chest x-ray abnormality detection. In this retrospective cohort study, we developed open-source neural networks, X-Raydar and X-Raydar-NLP, for classifying common chest x-ray findings from images and their free-text reports. Our networks were developed using data from six UK hospitals from three National Health Service (NHS) Trusts (University Hospitals Coventry and Warwickshire NHS Trust, University Hospitals Birmingham NHS Foundation Trust, and University Hospitals Leicester NHS Trust) collectively contributing 2 513 546 chest x-ray studies taken from a 13-year period (2006-19), which yielded 1 940 508 usable free-text radiological reports written by the contemporary assessing radiologist (collectively referred to as the "historic reporters") and 1 896 034 frontal images. Chest x-rays were labelled using a taxonomy of 37 findings by a custom-trained natural language processing (NLP) algorithm, X-Raydar-NLP, from the original free-text reports. X-Raydar-NLP was trained on 23 230 manually annotated reports and tested on 4551 reports from all hospitals. 1 694 921 labelled images from the training set and 89 238 from the validation set were then used to train a multi-label image classifier. Our algorithms were evaluated on three retrospective datasets: a set of exams sampled randomly from the full NHS dataset reported during clinical practice and annotated using NLP (n=103 328); a consensus set sampled from all six hospitals annotated by three expert radiologists (two independent annotators for each image and a third consultant to facilitate disagreement resolution) under research conditions (n=1427); and an independent dataset, MIMIC-CXR, consisting of NLP-annotated exams (n=252 374). X-Raydar achieved a mean AUC of 0·919 (SD 0·039) on the auto-labelled set, 0·864 (0·102) on the consensus set, and 0·842 (0·074) on the MIMIC-CXR test, demonstrating similar performance to the historic clinical radiologist reporters, as assessed on the consensus set, for multiple clinically important findings, including pneumothorax, parenchymal opacification, and parenchymal mass or nodules. On the consensus set, X-Raydar outperformed historical reporter balanced accuracy with significance on 27 of 37 findings, was non-inferior on nine, and inferior on one finding, resulting in an average improvement of 13·3% (SD 13·1) to 0·763 (0·110), including a mean 5·6% (13·2) improvement in critical findings to 0·826 (0·119). Our study shows that automated classification of chest x-rays under a comprehensive taxonomy can achieve performance levels similar to those of historical reporters and exhibit robust generalisation to external data. The open-sourced neural networks can serve as foundation models for further research and are freely available to the research community. Wellcome Trust.

Reliability of the CURL classification for proximal ulna fracture-dislocations of the elbow

BackgroundProximal ulna fracture-dislocations are challenging injuries with a myriad of existing classification systems. The Coronoid, proximal Ulna, Radius and Ligaments (CURL) classification is a simple framework designed to focus attention on the key components affecting outcome and guide surgical management. This study evaluates inter- and intra-observer reliability of this new classification. MethodsFour observers independently reviewed plain radiographs and CT scans of patients with proximal ulna fracture-dislocations. Each observer scored the Coronoid (C), proximal Ulna (U) and Radius (R) components for each fracture on two occasions. The osseous components were sub-classified as ‘intact’, ‘simple’ or ‘complex’. The Ligament component (L) was not rated as this requires intra-operative classification. Inter- and intra-observer reliability was calculated using Cohen’s weighted kappa coefficients. X-ray and CT were compared for patients with both imaging modalities. The Landis and Koch criteria were used to interpret the strength of the kappa statistics. Results177 patients had plain X rays, 58 patients had both X ray and CT scans. Overall, in the X ray only cohort, there was ‘almost perfect’ inter-observer reliability for the radial head (k=0.94) and coronoid (k=0.83), and ‘substantial’ reliability (k=0.68) for the proximal ulna. For the X ray and CT cohort, inter-observer reliability was ‘almost perfect’ across both modalities for the radial head (k=0.88 and k=0.93, respectively) and ‘moderate’ for the proximal ulna (k=0.48 and k=0.52, respectively). For the coronoid, inter-observer reliability for X-ray interpretation was ‘substantial’ (k=0.74) and for CT was ‘almost perfect’ (k=0.89). Intra-observer reliability was ‘almost perfect’ for all components, other than CT assessment of the proximal ulna which demonstrated ‘substantial’ reliability (k=0.74). ConclusionThe CURL classification demonstrates strong inter- and intra-observer reliability, supporting use of the classification for proximal ulna fracture-dislocations. CT is recommended for improved characterization of any fracture with a coronoid component.

Evaluation of lung ultrasound in the diagnosis of respiratory distress in neonates

Background: Respiratory distress (RD) is a leading cause of morbidity and mortality among neonates in neonatal intensive care units (NICUs). Traditionally, clinical criteria combined with chest X-ray have been considered the gold standard for diagnosing respiratory distress. This study aims to assess the efficiency of lung ultrasound in diagnosing respiratory distress, with a focus on preventing neonates from unnecessary radiation exposure. Methods: A multicentric prospective study was performed where neonates with gestational age 28-40 weeks were included based on presence of clinical signs of respiratory distress. Neonates with congenital anomalies or those who received surfactant therapy were excluded. Within 6 hrs of birth, these neonates underwent X ray chest and ultrasound. Final interpretation of chest x-ray was done by the blinded radiologist. A trained neonatologist used transthoracic view in ultrasound for the initial interpretation. Results: A total of 80 neonates were enrolled in the study, among whom 49 were diagnosed with respiratory distress syndrome, 22 with transient tachypnoea of the newborn, 4 with pneumonia, and 5 classified as normal. Lung ultrasound exhibited a sensitivity of 95.9% and a specificity of 90.3% for the diagnosis of respiratory distress, with a positive predictive value of 94% and a negative predictive value of 93.3%. Conclusions: Lung ultrasound proves to be a highly efficient diagnostic tool for respiratory distress in neonates, offering the added advantage of avoiding harmful radiation exposure associated with chest X-rays. Early detection of respiratory distress can facilitate timely management and significantly improve neonatal outcomes. Further adoption and validation of point of care lung ultrasound in clinical practice hold the potential to enhance neonatal care.

Acceptability of decentralizing childhood tuberculosis diagnosis in low-income countries with high tuberculosis incidence: Experiences and perceptions from health care workers in Sub-Saharan Africa and South-East Asia.

Decentralizing childhood tuberculosis services, including diagnosis, is now recommended by the WHO and could contribute to increasing tuberculosis detection in high burden countries. However, implementing microbiological tests and clinical evaluation could be challenging for health care workers (HCWs) in Primary Health Centers (PHCs) and even District Hospitals (DHs). We sought to assess the acceptability of decentralizing a comprehensive childhood tuberculosis diagnosis package from HCWs' perspective. We conducted implementation research nested within the TB-Speed Decentralization study. HCWs from two health districts of Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Sierra Leone, and Uganda implemented systematic screening, nasopharyngeal aspirates (NPA) and stool sample collection with molecular testing, clinical evaluation and chest X-ray (CXR) interpretation. We investigated their experiences and perceptions in delivering the diagnostic package components in 2020-21 using individual semi-structured interviews. We conducted thematic analysis, supported by the Theoretical Framework of Acceptability. HCWs (n = 130, 55% female, median age 36 years, 53% nurses, 72% PHC-based) perceived that systematic screening, although increasing workload, was beneficial as it improved childhood tuberculosis awareness. Most HCWs shared satisfaction and confidence in performing NPA, despite procedure duration, need to involve parents/colleagues and discomfort for children. HCWs shared positive attitudes towards stool sample-collection but were frustrated by delayed stool collection associated with cultural practices, transport and distance challenges. Molecular testing, conducted by nurses or laboratory technicians, was perceived as providing quality results, contributing to diagnosis. Clinical evaluation and diagnosis raised self-efficacy issues and need for continuous training and clinical mentoring. HCWs valued CXR, however complained that technical and logistical problems limited access to digital reports. Referral from PHC to DH was experienced as burdensome. HCWs at DH and PHC-levels perceived and experienced decentralized childhood tuberculosis diagnosis as acceptable. Implementation however could be hampered by feasibility issues, and calls for innovative referral mechanisms for patients, samples and CXR.

Losing vigilance in diagnosing pulmonary alveolar microlithiasis: A report on four cases

Pulmonary alveolar microlithiasis (PAM) is a rare chronic lung disease characterized by calcium and phosphate deposition in the alveolar lumen throughout the parenchyma of both lungs, with predominance in the middle and lower lung fields. It is caused by mutations in the recessive gene, SLC34A2, on the autosomal chromosome. In this article, we characterize four cases of PAM and analyze the loss of diagnostic vigilance in two of them. Patients came to medical facilities with clinical manifestations such as cough, shortness of breath, chest pain, and fatigue. The initial diagnosis was unclear in two cases because the X-ray film’s quality was not good enough and the medical staff had little experience in clinical and chest X-ray interpretations for PAM. The definitive diagnosis was based on a combination of high-resolution computed tomography (CT) and bronchoalveolar lavage fluid testing. In addition, chest X-ray and high-resolution CT enable the assessment of the stage, progression, and severity of the disease. There is currently no specific treatment for PAM other than lung transplantation.

Automated AI Detection Tool for Ankle Fractures Using X-Rays and Smart Devices

Category: Ankle; Trauma Introduction/Purpose: The use of artificial intelligence (AI) is particularly salient to visually oriented medical professions, especially orthopedics. The most prominent use of AI in orthopedics comes in the form of medical imaging examinations. AI has a huge potential to help doctors make diagnoses by acting as a second pair of eyes. Some results suggested a very high level of agreement between AI models’ and the clinician’s assessments of radiographs. Furthermore, it was also suggested that the sensitivity and specificity of emergency medicine physicians for detecting some pathologies are significantly improved when aided by an AI tool. In view of these observations, our study aimed to create an AI-based ankle fracture detection tool that can be used on smart devices for X-ray interpretation. Methods: We examined 2,193 patients’ charts from 2 academic and 1 community hospital in Boston. We retrieved the anteroposterior (AP), oblique, and lateral ankle X-rays of each patient. Patients with ankle fractures and adults older than 18 years met our inclusion criteria. We excluded patients younger than 18 years old and those with any artifact, such as a cast, screws, or other artifacts in their X-rays. The study comprised 352 healthy controls and a total of 579 ankle fracture patients. Other than the digital images obtained from Electronic patient records (EPR), we used two different smart devices, a cellphone, and a tablet, to capture images from the monitor screen. Using Machine Learning models, we developed a fracture detection model using all three types of imaging and named it the “combination model”. We subsequently tested the combination model on digital X-rays, smart devices, and on both datasets together (Table 1). Results: We extracted the X-rays of a total of 931 patients in this study. Following the development and testing of our AI models, we noticed that all performed well with AUCs and accuracies above 0.85 and 0.86, respectively (Table1). The best performance was found when the combination model was tested on images taken from the camera of our smart devices, with an AUC of 0.88, a sensitivity of 0.86, and an accuracy of 0.89. This performance was closely followed by that of our model tested on a mix of both smart devices and original digital images with an AUC of 0.88, a sensitivity of 0.86, and an accuracy of 0.88. Conclusion: Our AI-based tool showed promising performance in the detection of ankle fractures using smart devices and images obtained from the monitor screen. We were able to reach an accuracy of diagnosis on smart device-captured images that were comparable with the original digital X-rays. The outcome of this study can be used to help providers who lack sufficient experience in detecting fractures. It can also be used for educational purposes for trainees in this field.

AI-clinician collaboration via disagreement prediction: A decision pipeline and retrospective analysis of real-world radiologist-AI interactions

Clinical decision support tools can improve diagnostic performance or reduce variability, but they are also subject to post-deployment underperformance. Although using AI in an assistive setting offsets many concerns with autonomous AI in medicine, systems that present all predictions equivalently fail to protect against key AI safety concerns. We design a decision pipeline that supports the diagnostic model with an ecosystem of models, integrating disagreement prediction, clinical significance categorization, and prediction quality modeling to guide prediction presentation. We characterize disagreement using data from a deployed chest X-ray interpretation aid and compare clinician burden in this proposed pipeline to the diagnostic model in isolation. The average disagreement rate is 6.5%, and the expected burden reduction is 4.8%, even if 5% of disagreements on urgent findings receive a second read. We conclude that, in our production setting, we can adequately balance risk mitigation with clinician burden if disagreement false positives are reduced.

Testing the performance, adequacy, and applicability of an artificial intelligence model for pediatric pneumonia diagnosis

BackgroundCommunity-acquired Pneumonia (CAP) is a common childhood infectious disease. Deep learning models show promise in X-ray interpretation and diagnosis, but their validation should be extended due to limitations in the current validation workflow. To extend the standard validation workflow we propose doing a pilot test with the next characteristics. First, the assumption of perfect ground truth (100% sensitive and specific) is unrealistic, as high intra and inter-observer variability have been reported. To address this, we propose using Bayesian latent class models (BLCA) to estimate accuracy during the pilot. Additionally, assessing only the performance of a model without considering its applicability and acceptance by physicians is insufficient if we hope to integrate AI systems into day-to-day clinical practice. Therefore, we propose employing explainable artificial intelligence (XAI) methods during the pilot test to involve physicians and evaluate how well a Deep Learning model is accepted and how helpful it is for routine decisions as well as analyze its limitations by assessing the etiology. This study aims to apply the proposed pilot to test a deep Convolutional Neural Network (CNN)-based model for identifying consolidation in pediatric chest-X-ray (CXR) images already validated using the standard workflow. MethodsFor the standard validation workflow, a total of 5856 public CXRs and 950 private CXRs were used to train and validate the performance of the CNN model. The performance of the model was estimated assuming a perfect ground truth. For the pilot test proposed in this article, a total of 190 pediatric chest-X-ray (CXRs) images were used to test the CNN model support decision tool (SDT). The performance of the model on the pilot test was estimated using extensions of the two-test Bayesian Latent-Class model (BLCA). The sensitivity, specificity, and accuracy of the model were also assessed. The clinical characteristics of the patients were compared according to the model performance. The adequacy and applicability of the SDT was tested using XAI techniques. The adequacy of the SDT was assessed by asking two senior physicians the agreement rate with the SDT. The applicability was tested by asking three medical residents before and after using the SDT and the agreement between experts was calculated using the kappa index. ResultsThe CRXs of the pilot test were labeled by the panel of experts into consolidation (124/176, 70.4%) and no-consolidation/other infiltrates (52/176, 29.5%). A total of 31/176 (17.6%) discrepancies were found between the model and the panel of experts with a kappa index of 0.6. The sensitivity and specificity reached a median of 90.9 (95% Credible Interval (CrI), 81.2–99.9) and 77.7 (95% CrI, 63.3–98.1), respectively. The senior physicians reported a high agreement rate (70%) with the system in identifying logical consolidation patterns. The three medical residents reached a higher agreement using SDT than alone with experts (0.66±0.1 vs. 0.75±0.2). ConclusionsThrough the pilot test, we have successfully verified that the deep learning model was underestimated when a perfect ground truth was considered. Furthermore, by conducting adequacy and applicability tests, we can ensure that the model is able to identify logical patterns within the CXRs and that augmenting clinicians with automated preliminary read assistants could accelerate their workflows and enhance accuracy in identifying consolidation in pediatric CXR images.

The diagnostic value of chest X-ray scanning by the help of Artificial Intelligence in Heart Failure (ART-IN-HF).

Typical signs of heart failure (HF), like increased cardiothoracic ratio (CTR) and pleural effusion, can be seen on X-ray. Artificial Intelligence (AI) can help in the early and quicker diagnosis of HF. The study's goal was to demonstrate that the AI interpretation of chest X-rays can assist the clinician in diagnosing HF. Patients older than 45 years were included in the study. The study analyzed 10 100 deidentified outpatient chest X-rays by AI algorithm. The AI-generated report was later verified by an independent radiologist. Patients with CTR > 0.5 and pleural effusion were marked as potential HF. Flagged patients underwent confirmatory tests, and those labeled as negative also underwent further investigations to rule out HF. Out of 10 100, the AI algorithm detected 183 (1.8%) patients with increased CTR and pleural effusion on chest X-rays. One hundred and six out of 183 underwent diagnostic tests. Eighty-two (77%) out of 106 were diagnosed with HF according to current guidelines. From the remaining 9917 patients, 106 patients were randomly selected. Nine (8%) out of them were diagnosed with HF. The positive predictive value of AI for diagnosing HF is 77%, and the negative predictive value is 91%. More than half (54.9%) of newly diagnosed patients had HF with preserved ejection fraction. HF is a risky condition with nonspecific symptoms that are difficult to diagnose, especially in the early stages. Using AI assistance for X-ray interpretation can be helpful for early diagnosis of HF especially HF with preserved ejection fraction.

The Effects of Various Teaching Methods on Chest X-ray Interpretation Skills Among Medical Students and Interns: A Systematic Review.

Chest X-ray (CXR) is a common tool used in medical practice. Medical students and interns should acquire knowledge of CXR interpretation, as it is an essential diagnostic tool for a large spectrum of diseases. This systematic review aimed to compare the effect of different intervention techniques on the competency of medical students and interns to demonstrate the level of confidence and competence in interpreting common presentations of CXRs. The population, intervention, comparison, and outcomes (PICO) framework was used to formulate the review question. All related articles in five databases (PubMed, Web of Science, Scopus, Medline, and Embase) were retrieved and the search was completed in March 2023with no limiters on date and time. The number of relevant studies was 469. A multi-level approach through the Rayyan platform was used for the screening and exclusion processes. Eleven articles were included in the systematic review consisting of eight randomized controlled trials, one quasi-experimental study, one cross-sectional study, and one interventional cohort. Results showed significant effects of teaching methods utilizing deductive or inductive approach, clinical history, patient care comfort survey, and SAFMEDS (Say-All-Fast-Minute-Every-Day-Shuffled). Contrarily, no significant effect was shown by flipped classroom models and mixed and blocked practice, peer-assisted learning vs. expert-assisted learning, and Chester, an artificial intelligence tool. This review identified beneficial approaches that may enhance the learning outcomes of interpreting CXRs for medical students and interns, highlighting the remarkable impact of SAFMEDSon medical students' ability to identify CXR findings as well as the availability and practicality of online and e-learning resources for students.

Learning from the machine: AI assistance is not an effective learning tool for resident education in chest x-ray interpretation.

To assess whether a computer-aided detection (CADe) system could serve as a learning tool for radiology residents in chest X-ray (CXR) interpretation. Eight radiology residents were asked to interpret 500 CXRs for the detection of five abnormalities, namely pneumothorax, pleural effusion, alveolar syndrome, lung nodule, and mediastinal mass. After interpreting 150 CXRs, the residents were divided into 2 groups of equivalent performance and experience. Subsequently, group 1 interpreted 200 CXRs from the "intervention dataset" using a CADe as a second reader, while group 2 served as a control by interpreting the same CXRs without the use of CADe. Finally, the 2 groups interpreted another 150 CXRs without the use of CADe. The sensitivity, specificity, and accuracy before, during, and after the intervention were compared. Before the intervention, the median individual sensitivity, specificity, and accuracy of the eight radiology residents were 43% (range: 35-57%), 90% (range: 82-96%), and 81% (range: 76-84%), respectively. With the use of CADe, residents from group 1 had a significantly higher overall sensitivity (53% [n = 431/816] vs 43% [n = 349/816], p < 0.001), specificity (94% [i = 3206/3428] vs 90% [n = 3127/3477], p < 0.001), and accuracy (86% [n = 3637/4244] vs 81% [n = 3476/4293], p < 0.001), compared to the control group. After the intervention, there were no significant differences between group 1 and group 2 regarding the overall sensitivity (44% [n = 309/696] vs 46% [n = 317/696], p = 0.666), specificity (90% [n = 2294/2541] vs 90% [n = 2285/2542], p = 0.642), or accuracy (80% [n = 2603/3237] vs 80% [n = 2602/3238], p = 0.955). Although it improves radiology residents' performances for interpreting CXRs, a CADe system alone did not appear to be an effective learning tool and should not replace teaching. Although the use of artificial intelligence improves radiology residents' performance in chest X-rays interpretation, artificial intelligence cannot be used alone as a learning tool and should not replace dedicated teaching. • With CADe as a second reader, residents had a significantly higher sensitivity (53% vs 43%, p < 0.001), specificity (94% vs 90%, p < 0.001), and accuracy (86% vs 81%, p < 0.001), compared to residents without CADe. • After removing access to the CADe system, residents' sensitivity (44% vs 46%, p = 0.666), specificity (90% vs 90%, p = 0.642), and accuracy (80% vs 80%, p = 0.955) returned to that of the level for the group without CADe.