Errores en la interpretación de la radiografía de tórax en el lactante: ¿qué tenemos que saber de la técnica? / Errors in the Interpretation of chest X-rays in Infants: what do we need to know about the technique?

La radiografía de tórax sigue siendo fundamental para la evaluación de patologías torácicas en lactantes. Antes de iniciar su interpretación, muchos autores sugieren revisar la técnica radiológica, ya que a esta edad se presentan varias particularidades técnicas que deben tenerse en cuenta para evitar errores interpretativos y no confundir hallazgos técnicos con patologías. Entre estas particularidades técnicas se deben evaluar: el centraje transversal o rotación, el centraje longitudinal o posición lordótica, el grado de inspiración, la posición de la vía aérea superior, la penetración o exposición de la radiografía, tipos de proyecciones y el movimiento. El objetivo de esta revisión es comentar y ejemplificar las peculiaridades técnicas que presenta la radiografía de tórax en lactantes y que pueden llevar a interpretaciones erróneas.

The chest X-ray remains essential for evaluating thoracic pathology in infants. Before beginning its interpretation, many authors recommend assessing the radiographic technique, as several technical peculiarities must be considered at this age to avoid interpretive errors and prevent mistaking technical artifacts for pathology. The technical aspects to be evaluated include transverse centering or rotation, longitudinal centering or lordotic position, degree of inspiration, upper airway positioning, radiograph penetration or exposure, projection types, and movement. The objective of this review is to discuss and illustrate the technical peculiarities of infant chest X-rays that can lead to erroneous interpretations.

A modified Tseng algorithm approach to restoring thoracic diseases' computerized tomography images.

It is well-known that the Tseng algorithm and its modifications have been successfully employed in approximating zeros of the sum of monotone operators. In this study, we restored various thoracic diseases' computerized tomography (CT) images, which were degraded with a known blur function and additive noise, using a modified Tseng algorithm. The test images used in the study depict calcification of the Aorta, Subcutaneous Emphysema, Tortuous Aorta, Pneumomediastinum, and Pneumoperitoneum. Additionally, we employed well-known image restoration tools to enhance image quality and compared the quality of restored images with the originals. Finally, the study demonstrates the potential to advance monotone inclusion problem-solving, particularly in the field of medical image recovery.

Radiation-free and injection-free imaging of the paediatric chest using a magnetic resonance imaging protocol including zero time echo sequence (3D-ZTE).

Computed tomography (CT) is commonly used for paediatric thoracic diseases but involves radiation exposure and often requires intravenous contrast. We evaluated the performance of a magnetic resonance imaging (MRI) protocol including a 3D zero echo time (3D-ZTE) sequence for radiation-free and contrast-free imaging of the paediatric chest. In this prospective, single-centre study, children aged 6-16 years underwent chest CT and MRI within 48 h. CT and MRI exams were independently assessed by two paediatric radiologists. The primary outcome was the image quality of the 3D-ZTE sequence using a scoring system based on the acceptability of the images obtained and visibility of bronchial structures, vessels and fissures. Secondary outcomes included radiologists' ability to detect lung lesions on 3D-ZTE MRI images compared with CT images. Seventy-two children were included. Overall, the image quality achieved with the 3D-ZTE MRI sequence was inferior to that of CT for visualising pulmonary structures, with satisfactory lung image quality observed for 81.9% (59/72) and 100% (72/72) of patients, respectively. However, MRI sensitivity was excellent (above 90%) for the detection of certain lesions such as lung consolidation, proximal mucoid impactions, pulmonary cysts, ground glass opacities and honeycombing. Intermodality agreement (MRI versus CT) was consistently higher for the senior reader compared to the junior reader. CONCLUSION: Despite its overall lower image quality compared to CT, and the additional years of experience required for accurate interpretation, the 3D-ZTE MRI sequence demonstrated excellent sensitivity for several lesions, making it an appropriate imaging method in certain indications. WHAT IS KNOWN: • Chest radiography and CT are the main imaging modalities for paediatric thoracic diseases but involve radiation exposure and CT often requires IV contrast. • MRI is promising for radiation-free lung imaging in children but faces challenges of low signal-to-noise ratio and motion artefacts. WHAT IS NEW: • An MRI protocol including a 3D zero echo time (ZTE) sequence allows satisfactory visualisation of lung parenchyma in 82% of children. • Despite overall inferior image quality compared to CT, MRI demonstrated excellent sensitivity for several lesions, making it an appropriate imaging method in certain indications.

Concordancia diagnóstica de las imágenes torácicas para enfermedad por Covid-19 / Diagnostic concordance of thoracic images for Covid-19 disease

La reacción en cadena de la polimerasa de transcripción inversa (RT-PCR) es el estándard de oro para el diagnóstico de enfermedad por SARS-CoV-2. En el contexto de la pandemia con accesibilidad limitada a esta prueba, las imágenes diagnósticas aportaron hallazgos que sustentan la sospecha diagnóstica, evitando retrasos en atención médica. Objetivo. Determinar la sensibilidad, especificidad, valor predictivo positivo y negativo de las imágenes diagnósticas y su concordancia respecto al resultado de RT-PCR. Metodología. Estudio transversal analítico. Se comparó el resultado del reporte por imágenes con los resultados de RT-PCR en 138 pacientes. Se calculó la sensibilidad, especificidad, valor predictivo positivo y valor predictivo negativo para los rayos X de tórax y tomografía computarizada para el diagnóstico de infección por SARS-CoV-2. Se utilizó el índice Kappa de Cohen y el factor de Bayes para medir la concordancia y fuerza de asociación entre las variables. Resultados. La tomografía computarizada presentó una sensibilidad de 92,9 %, una especificidad del 64 %, un valor predictivo positivo de 92,1 % y un valor predictivo negativo de 66,7 %; mientras que, los rayos X presentaron una sensibilidad del 86 %, una especificidad del 52,9 %, un valor predictivo positivo de 92,9 % y un valor predictivo negativo del 34,6 %. Conclusión. La tomografía mostró concordancia diagnóstica moderada; su utilidad es mayor en casos de sospecha clínica moderada-alta, discrepancia diagnóstica o confirmación de complicaciones. Los rayos X mostraron concordancia diagnóstica baja; este método es de utilidad en casos de alta sospecha clínica, pero necesita comprobación con RT-PCR para un diagnóstico definitivo.

Reverse transcription polymerase chain reaction (RT-PCR) is the gold standard method for diagnosing SARS-CoV-2 disease. However, due to limited accessibility to this test during the pandemic, diagnostic imaging was used to support diagnostic suspicion and avoid delays in medical care. Objective. Determine the accuracy of diagnostic imaging (chest X-ray and computed tomography) in diagnosing SARS-CoV-2 infection, compared to RT-PCR result. Methodology.An analytical cross-sectional study was conducted. The imaging reports of 138 patients were compared with their RT-PCR results to calculate sensitivity, specificity, positive predictive value, and negative predictive value for both chest X-ray and computed tomography. Concordance between the imaging results and RT-PCR was measured using Cohen's Kappa index and Bayes factor. Results. Computed tomography showed a sensitivity of 92.9 %, a specificity of 64 %, a positive predictive value of 92.1 %, and a negative predictive value of 66.7 %. On the other hand, X-rays showed a sensitivity of 86 %, a specificity of 52.9 %, a positive predictive value of 92.9 %, and a negative predictive value of 34.6 %. Conclusion. Computed tomography showed moderate diagnostic concordance and is particularly useful in cases of moderate to high clinical suspicion, diagnostic discrepancy, or the need to confirm complications. On the other hand, X-rays showed low diagnostic concordance and should be used in combination with RT-PCR for a definitive diagnosis, especially in cases of high clinical suspicion

Towards long-tailed, multi-label disease classification from chest X-ray: Overview of the CXR-LT challenge.

Many real-world image recognition problems, such as diagnostic medical imaging exams, are "long-tailed" - there are a few common findings followed by many more relatively rare conditions. In chest radiography, diagnosis is both a long-tailed and multi-label problem, as patients often present with multiple findings simultaneously. While researchers have begun to study the problem of long-tailed learning in medical image recognition, few have studied the interaction of label imbalance and label co-occurrence posed by long-tailed, multi-label disease classification. To engage with the research community on this emerging topic, we conducted an open challenge, CXR-LT, on long-tailed, multi-label thorax disease classification from chest X-rays (CXRs). We publicly release a large-scale benchmark dataset of over 350,000 CXRs, each labeled with at least one of 26 clinical findings following a long-tailed distribution. We synthesize common themes of top-performing solutions, providing practical recommendations for long-tailed, multi-label medical image classification. Finally, we use these insights to propose a path forward involving vision-language foundation models for few- and zero-shot disease classification.

Congenital Chest Lesions and Interventions.

Anomalies of the fetal chest require advanced imaging with ultrasound and MR imaging as well as expertise on the part of the interpreting pediatric radiologist. Congenital diaphragmatic hernia and congenital lung malformation are the most frequently seen, and in both conditions, the radiologist should provide both detailed anatomic description and measurement data for prognostication. This article provides a detailed approach to imaging the anatomy, in-depth explanation of available measurements and prognostic value, and keys to identifying candidates for fetal intervention. Less common congenital lung tumors and mediastinal and chest wall masses are also reviewed.

Disorders with Ophthalmic and Thoracic Involvement.

A variety of systemic conditions involve the thorax and the eyes. While subtle or nonspecific eye symptoms can be the initial clinical manifestation of some disorders, there can be additional manifestations in the thorax that lead to a specific diagnosis and affect patient outcomes. For instance, the initial clinical manifestation of Sjögren syndrome is dry eye or xerophthalmia; however, the presence of Sjögren lung disease represents a fourfold increase in mortality. Likewise, patients with acute sarcoidosis can initially present with pain and redness of the eye from uveitis in addition to fever and parotitis. Nearly 90% of patients with sarcoidosis have thoracic involvement, and the ophthalmologic symptoms can precede the thoracic symptoms by several years in some cases. Furthermore, a diagnosis made in one system can result in the screening of other organs as well as prompt genetic evaluation and examination of family members, such as in the setting of Marfan syndrome or Ehlers-Danlos syndrome. Multimodality imaging, particularly CT and MRI, plays a vital role in identification and characterization of these conditions. While it is helpful for ophthalmologists to be knowledgeable about these conditions and their associations so that they can order the pertinent radiologic studies, it is also important for radiologists to use the clues from ophthalmologic examination in addition to imaging findings to suggest a specific diagnosis. Systemic conditions with thoracic and ophthalmologic manifestations can be categorized as infectious, inflammatory, autoimmune, neoplastic, or hereditary in origin. The authors describe a spectrum of these conditions based on their underlying cause. ©RSNA, 2024.

Automated thorax disease diagnosis using multi-branch residual attention network.

Chest X-ray (CXR) is an extensively utilized radiological modality for supporting the diagnosis of chest diseases. However, existing research approaches suffer from limitations in effectively integrating multi-scale CXR image features and are also hindered by imbalanced datasets. Therefore, there is a pressing need for further advancement in computer-aided diagnosis (CAD) of thoracic diseases. To tackle these challenges, we propose a multi-branch residual attention network (MBRANet) for thoracic disease diagnosis. MBRANet comprises three components. Firstly, to address the issue of inadequate extraction of spatial and positional information by the convolutional layer, a novel residual structure incorporating a coordinate attention (CA) module is proposed to extract features at multiple scales. Next, based on the concept of a Feature Pyramid Network (FPN), we perform multi-scale feature fusion in the following manner. Thirdly, we propose a novel Multi-Branch Feature Classifier (MFC) approach, which leverages the class-specific residual attention (CSRA) module for classification instead of relying solely on the fully connected layer. In addition, the designed BCEWithLabelSmoothing loss function improves the generalization ability and mitigates the problem of class imbalance by introducing a smoothing factor. We evaluated MBRANet on the ChestX-Ray14, CheXpert, MIMIC-CXR, and IU X-Ray datasets and achieved average AUCs of 0.841, 0.895, 0.805, and 0.745, respectively. Our method outperformed state-of-the-art baselines on these benchmark datasets.

Which is a real valuable screening tool for lung cancer and measure thoracic diseases, chest radiography or low-dose computed tomography?: A review on the current status of Japan and other countries.

Chest radiography (CR) has been used as a screening tool for lung cancer and the use of low-dose computed tomography (LDCT) is not recommended in Japan. We need to reconsider whether CR really contributes to the early detection of lung cancer. In addition, we have not well discussed about other major thoracic disease detection by CR and LDCT compared with lung cancer despite of its high frequency. We review the usefulness of CR and LDCT as veridical screening tools for lung cancer and other thoracic diseases. In the case of lung cancer, many studies showed that LDCT has capability of early detection and improving outcomes compared with CR. Recent large randomized trial also supports former results. In the case of chronic obstructive pulmonary disease (COPD), LDCT contributes to early detection and leads to the implementation of smoking cessation treatments. In the case of pulmonary infections, LDCT can reveal tiny inflammatory changes that are not observed on CR, though many of these cases improve spontaneously. Therefore, LDCT screening for pulmonary infections may be less useful. CR screening is more suitable for the detection of pulmonary infections. In the case of cardiovascular disease (CVD), CR may be a better screening tool for detecting cardiomegaly, whereas LDCT may be a more useful tool for detecting vascular changes. Therefore, the current status of thoracic disease screening is that LDCT may be a better screening tool for detecting lung cancer, COPD, and vascular changes. CR may be a suitable screening tool for pulmonary infections and cardiomegaly.

Improving Computer-Aided Thoracic Disease Diagnosis through Comparative Analysis Using Chest X-ray Images Taken at Different Times.

Medical professionals in thoracic medicine routinely analyze chest X-ray images, often comparing pairs of images taken at different times to detect lesions or anomalies in patients. This research aims to design a computer-aided diagnosis system that enhances the efficiency of thoracic physicians in comparing and diagnosing X-ray images, ultimately reducing misjudgments. The proposed system encompasses four key components: segmentation, alignment, comparison, and classification of lung X-ray images. Utilizing a public NIH Chest X-ray14 dataset and a local dataset gathered by the Chiayi Christian Hospital in Taiwan, the efficacy of both the traditional methods and deep-learning methods were compared. Experimental results indicate that, in both the segmentation and alignment stages, the deep-learning method outperforms the traditional method, achieving higher average IoU, detection rates, and significantly reduced processing time. In the comparison stage, we designed nonlinear transfer functions to highlight the differences between pre- and post-images through heat maps. In the classification stage, single-input and dual-input network architectures were proposed. The inclusion of difference information in single-input networks enhances AUC by approximately 1%, and dual-input networks achieve a 1.2-1.4% AUC increase, underscoring the importance of difference images in lung disease identification and classification based on chest X-ray images. While the proposed system is still in its early stages and far from clinical application, the results demonstrate potential steps forward in the development of a comprehensive computer-aided diagnostic system for comparative analysis of chest X-ray images.

Spectrum of Thoracic Imaging Findings in the Setting of Substance Abuse.

ABSTRACT: Substance abuse continues to be prevalent nationwide and can lead to a myriad of chest pathologies. Imaging findings are vast and can include nodules, masses, ground-glass opacities, airspace disease, and cysts. Radiologists with awareness of these manifestations can assist in early identification of disease in situations where information is unable to be obtained from the patient. This review focuses on thoracic imaging findings associated with various forms of substance abuse, which are organized by portal of entry into the thorax: inhalation, ingestion, and injection.

EBUS-guided cryobiopsy in the diagnosis of thoracic disorders.

BACKGROUND: Endobronchial Ultrasound (EBUS) has emerged as a crucial tool for diagnosing intrathoracic disorders, particularly in the staging of lung cancer. However, its diagnostic capabilities in the context of benign and rare diseases remain a subject of debate. AIM: to investigate the diagnostic yield and safety of EBUS-transbronchial mediastinal cryobiopsy (EBUS-TMC) in comparison to EBUS-transbronchial needle aspiration (TBNA) for a broad spectrum of intrathoracic diseases. METHODS: a single-centre retrospective observational study conducted on 48 patients who underwent both EBUS-TBNA and endobronchial ultrasound-transbronchial mediastinal cryobiopsy (EBUS-TMC) in the same procedure between August 2021 and October 2023. RESULTS: The overall diagnostic yield of EBUS-TMC surpassed that of EBUS-TBNA (95.8% vs 54.1 %), notably excelling in the diagnosis of sarcoidosis (92.8% vs 78.5 %), rare mediastinal disorders (100% vs 0 %), hyperplastic lymphadenopathy (100% vs 0 %), and lymphoproliferative disease (100% vs 0 %). No significant differences were observed in the diagnosis of NSCLC and SCLC. Samples obtained through EBUS-TMC facilitated the acquisition of NGS and immunohistochemical analyses more readily. CONCLUSION: EBUS-TMC may contribute to the precise diagnosis and subtyping of mediastinal diseases, especially lymphomas and rare mediastinal tumors, thereby reducing the number of non-diagnostic procedures.

Robust Stochastic Neural Ensemble Learning With Noisy Labels for Thoracic Disease Classification.

Chest radiography is the most common radiology examination for thoracic disease diagnosis, such as pneumonia. A tremendous number of chest X-rays prompt data-driven deep learning models in constructing computer-aided diagnosis systems for thoracic diseases. However, in realistic radiology practice, a deep learning-based model often suffers from performance degradation when trained on data with noisy labels possibly caused by different types of annotation biases. To this end, we present a novel stochastic neural ensemble learning (SNEL) framework for robust thoracic disease diagnosis using chest X-rays. The core idea of our method is to learn from noisy labels by constructing model ensembles and designing noise-robust loss functions. Specifically, we propose a fast neural ensemble method that collects parameters simultaneously across model instances and along optimization trajectories. Moreover, we propose a loss function that both optimizes a robust measure and characterizes a diversity measure of ensembles. We evaluated our proposed SNEL method on three publicly available hospital-scale chest X-ray datasets. The experimental results indicate that our method outperforms competing methods and demonstrate the effectiveness and robustness of our method in learning from noisy labels. Our code is available at https://github.com/hywang01/SNEL.

Interventional Pulmonology: Extending the Breadth of Thoracic Care.

Interventional pulmonary medicine has developed as a subspecialty focused on the management of patients with complex thoracic disease. Leveraging minimally invasive techniques, interventional pulmonologists diagnose and treat pathologies that previously required more invasive options such as surgery. By mitigating procedural risk, interventional pulmonologists have extended the reach of care to a wider pool of vulnerable patients who require therapy. Endoscopic innovations, including endobronchial ultrasound and robotic and electromagnetic bronchoscopy, have enhanced the ability to perform diagnostic procedures on an ambulatory basis. Therapeutic procedures for patients with symptomatic airway disease, pleural disease, and severe emphysema have provided the ability to palliate symptoms. The combination of medical and procedural expertise has made interventional pulmonologists an integral part of comprehensive care teams for patients with oncologic, airway, and pleural needs. This review surveys key areas in which interventional pulmonologists have impacted the care of thoracic disease through bronchoscopic intervention.