Enhancing thoracic disease detection using chest X-rays from PubMed Central Open Access.

Large chest X-rays (CXR) datasets have been collected to train deep learning models to detect thorax pathology on CXR. However, most CXR datasets are from single-center studies and the collected pathologies are often imbalanced. The aim of this study was to automatically construct a public, weakly-labeled CXR database from articles in PubMed Central Open Access (PMC-OA) and to assess model performance on CXR pathology classification by using this database as additional training data. Our framework includes text extraction, CXR pathology verification, subfigure separation, and image modality classification. We have extensively validated the utility of the automatically generated image database on thoracic disease detection tasks, including Hernia, Lung Lesion, Pneumonia, and pneumothorax. We pick these diseases due to their historically poor performance in existing datasets: the NIH-CXR dataset (112,120 CXR) and the MIMIC-CXR dataset (243,324 CXR). We find that classifiers fine-tuned with additional PMC-CXR extracted by the proposed framework consistently and significantly achieved better performance than those without (e.g., Hernia: 0.9335 vs 0.9154; Lung Lesion: 0.7394 vs. 0.7207; Pneumonia: 0.7074 vs. 0.6709; Pneumothorax 0.8185 vs. 0.7517, all in AUC with p< 0.0001) for CXR pathology detection. In contrast to previous approaches that manually submit the medical images to the repository, our framework can automatically collect figures and their accompanied figure legends. Compared to previous studies, the proposed framework improved subfigure segmentation and incorporates our advanced self-developed NLP technique for CXR pathology verification. We hope it complements existing resources and improves our ability to make biomedical image data findable, accessible, interoperable, and reusable.

Tension pneumothorax in a patient with Covid-19 pneumonia initially misdiagnosed as volvulus.

Tension pneumothorax is a condition that can present with a myriad of symptoms, including chest pain, shortness of breath, rapid breathing, and tachycardia. If left untreated, these signs and symptoms can progress to shock causing haemodynamic collapse and even death. At times, it may be difficult to identify tension pneumothorax. We present the case of a 59-year-old male with a prolonged initial hospital course that eventually led to a diagnosis of tension pneumothorax with the use of CT scans rather than traditional X-rays. This case reinforces the idea that clinicians should have a wide differential diagnosis in mind when dealing with vague symptoms and should not hesitate to use different diagnostic modalities to help confirm a diagnosis.

Spontaneous pneumothorax in COVID-19: a single center experience.

OBJECTIVE: The effect of pulmonary complications of COVID-19, such as pneumothorax, pneumomediastinum, and subcutaneous emphysema, is still unclear. This study aimed at investigating the relationship between COVID-19 and spontaneous pneumothorax. PATIENTS AND METHODS: This study was conducted as a single-center retrospective study. Groups were assigned as study and control groups. The study group (n=120) included patients who were followed up in ICU and developed pneumothorax during their follow-up. The control group (n=120) included patients who did not develop a pneumothorax in ICU and who had been randomly selected using hospital records. Demographic findings, laboratory parameters, radiological findings, clinical management, patients' follow-up patterns, and survival status of the patients were recorded. RESULTS: There was a significant relationship between gender, outcome, last hospitalization, general condition, first follow-up, intubation, uptake tomography, uptake rate, CO-RADS, and involvement variables between groups (p<0.05). In the survival analysis performed in the control and study groups, a significant difference was obtained between the averages of the two groups (LogRank=3.944, p<0.05). Intubation and mortality rates of the patients who developed pneumothorax during the patient follow-ups were significantly higher than the control group. CONCLUSIONS: We found that patients diagnosed with COVID-19 who developed pneumothorax during intensive care follow-up had a higher hospital stay and intubation rate. The pneumothorax rate was also higher in follow-up methods such as noninvasive/HFO providing PEEP to the patients. The data in our study may help reducing mortality by shedding light on the early prevention and recognition of pneumothorax in critically ill patients diagnosed with COVID-19.

Barotrauma in COVID 19: Incidence, pathophysiology, and effect on prognosis.

OBJECTIVES: To investigate the incidence, risk factors, and outcomes of barotrauma (pneumomediastinum and subcutaneous emphysema) in mechanically ventilated COVID-19 patients. To describe the chest radiography patterns of barotrauma and understand the development in relation to mechanical ventilation and patient mortality. METHODS: We performed a retrospective study of 363 patients with COVID-19 from March 1 to April 8, 2020. Primary outcomes were pneumomediastinum or subcutaneous emphysema with or without pneumothorax, pneumoperitoneum, or pneumoretroperitoneum. The secondary outcomes were length of intubation and death. In patients with pneumomediastinum and/or subcutaneous emphysema, we conducted an imaging review to determine the timeline of barotrauma development. RESULTS: Forty three out of 363 (12%) patients developed barotrauma radiographically. The median time to development of either pneumomediastinum or subcutaneous emphysema was 2 days (IQR 1.0-4.5) after intubation and the median time to pneumothorax was 7 days (IQR 2.0-10.0). The overall incidence of pneumothorax was 28/363 (8%) with an incidence of 17/43 (40%) in the barotrauma cohort and 11/320 (3%) in those without barotrauma (p ≤ 0.001). In total, 257/363 (71%) patients died with an increase in mortality in those with barotrauma 33/43 (77%) vs. 224/320 (70%). When adjusting for covariates, barotrauma was associated with increased odds of death (OR 2.99, 95% CI 1.25-7.17). CONCLUSION: Barotrauma is a frequent complication of mechanically ventilated COVID-19 patients. In comparison to intubated COVID-19 patients without barotrauma, there is a higher rate of pneumothorax and an increased risk of death.

An extremely rare case of COVID-19: bilateral spontaneous pneumothorax.

COVID-19 is a disease caused by the SARS-CoV-2 infection which causes pneumonic infiltration in both lungs. This research paper is meant to present our case of a 31-year-old male who experienced bilateral spontaneous pneumothorax. The rarity of this condition cannot be underestimated considering pneumothorax from COVID-19 is quite uncommon.

Incidence and Clinical Features of Pneumomediastinum and Pneumothorax in COVID-19 Pneumonia.

BACKGROUND: Pneumothorax (PTX) and pneumomediastinum (PM), collectively termed here "air leak", are now well described complications of severe COVID-19 pneumonia across several case series. The incidence is thought to be approximately 1% but is not definitively known. OBJECTIVES: To report the incidence and describe the demographic features, risk factors and outcomes of patients with air leak as a complication of COVID-19. METHODS: A retrospective observational study on all adult patients with COVID-19 admitted to Watford General Hospital, West Hertfordshire NHS Trust between March 1st 2020 and Feb 28th 2021. Patients with air leak were identified after reviewing both chest radiographs (CXRs) and axial imaging (CT Thorax) with confirmatory radiology reports inclusive of the terms PTX and/or PM. RESULTS: Air leak occurred with an incidence of 0.56%. Patients with air leak were younger and had evidence of more severe disease at presentation, including a higher median CRP and number of abnormal zones affected on chest radiograph. Asthma was a significant risk factor in the development of air leak (OR 13.4 [4.7-36.4]), both spontaneously and following positive pressure ventilation. CPAP and IMV were also associated with a greater than six fold increase in the risk of air leak (OR 6.4 [2.5-16.6] and 9.8 [3.7-27.8] respectively). PTX, with or without PM, in the context of COVID-19 pneumonia was almost universally fatal whereas those with alone PM had a lower risk of death. CONCLUSION: Despite the global vaccination programme, patients continue to develop severe COVID-19 disease and may require respiratory support. This study demonstrates the importance of identifying that deterioration in such patients may be resultant from PTX or PM, particularly in asthmatics and those managed with positive pressure ventilation.

Deep Learning-Based Computer-Aided Pneumothorax Detection Using Chest X-ray Images.

Pneumothorax is a thoracic disease leading to failure of the respiratory system, cardiac arrest, or in extreme cases, death. Chest X-ray (CXR) imaging is the primary diagnostic imaging technique for the diagnosis of pneumothorax. A computerized diagnosis system can detect pneumothorax in chest radiographic images, which provide substantial benefits in disease diagnosis. In the present work, a deep learning neural network model is proposed to detect the regions of pneumothoraces in the chest X-ray images. The model incorporates a Mask Regional Convolutional Neural Network (Mask RCNN) framework and transfer learning with ResNet101 as a backbone feature pyramid network (FPN). The proposed model was trained on a pneumothorax dataset prepared by the Society for Imaging Informatics in Medicine in association with American college of Radiology (SIIM-ACR). The present work compares the operation of the proposed MRCNN model based on ResNet101 as an FPN with the conventional model based on ResNet50 as an FPN. The proposed model had lower class loss, bounding box loss, and mask loss as compared to the conventional model based on ResNet50 as an FPN. Both models were simulated with a learning rate of 0.0004 and 0.0006 with 10 and 12 epochs, respectively.

Spontaneous Pneumomediastinum in COVID-19 Pneumonia: A Rare Occurence.

Spontaneous pneumomediastinum with COVID pneumonia is a rare occurrence with or without accompanying subcutaneous emphysema or pneumothorax. The aim of this study was to establish relation of this complication to severity of lung disease and its clinical outcome. The study was conducted for a period of seven months from April to October 2020 in the CT Department of Armed Forces Institute of Radiology and Imaging (AFIRI), Rawalpindi, Pakistan. All COVID positive patients having spontaneous pneumomediastinum on high resolution CT (HRCT) chest were included (n=14). These patients were assessed for severity of lung disease as per CT severity score (CTSS), and were followed up for their clinical outcome. All patients with spontaneous pneumomediastinum had moderate to severe degree of COVID pneumonia; mortality in patients with pneumomediastinum was 50%; and was seen in those patients who had greater severity of lung disease as per the CTSS. Key Words: Spontaneous, Pneumomediastinum, COVID, Pneumothorax, Subcutaneous, Emphysema.

Post COVID-19 pneumo-hematocele: clinical presentation and treatment pathway.

OBJECTIVES: The appearance of characteristic pulmonary lesions has been noted after COVID-19, being described as post-COVID-19 pneumo-hematocele. The aim of this study is to describe the clinical, histopathologic, and imaging features of pneumo-hematocele and to suggest a treatment algorithm for these patients. METHODS: A retrospective study was performed in patients admitted with a diagnosis of SARS-CoV2 infection from March 2020 to September 2021 who presented a pneumo-hematocele on imaging studies. Clinical and demographic variables were recorded, and CT scans were analyzed. A secondary analysis was performed to estimate the risk provided by the pneumo-hematocele diameter of developing pneumothorax. RESULTS: 37 patients were diagnosed with pneumo-hematoceles, 97.3% were males with a median age of 41 ± 13 years and 51% were smokers. The mean diameter of the pneumatocele was 6.3 ± 2.8 cm; they were more common on the subpleural surface and in the inferior lobe. Thirty patients had ruptured pneumo-hematoceles and developed pneumothorax (81.1%); a total of 26 patients required surgery (70.3%). Lesions measuring 5 cm had a high risk of rupture (OR 6.8, CI 95% 1.1-42); those measuring 3 cm were prone to this complication. For each centimeter that the pneumo-hematocele diameter increases, the OR for rupture increases 1.5. CONCLUSIONS: It appears that post-COVID-19 pneumo-hematocele occurs secondary to encapsulation of blood accumulation inside the lung, as a result of micro-capillary bleeding, with partial reabsorption of blood and subsequent air filling. We recommend surgery for patients with pneumo-hematoceles of 5 cm and those with persistent lesions of 3 cm. TRIAL REGISTRATION: Clinical Trial Registration: NCT05067881.

Pneumothorax and pneumomediastinum in patients with COVID-19: A retrospective study from tertiary care institute in India.

COVID-19 is associated with rarer extra-parenchymal manifestations, namely pneumothorax (PTX) and pneumomediastinum (PM) leading to complications and increased mortality. The study aims to describe the prevalence, risk factors for mortality, radiological characteristics and outcome of PTX/PM in patients admitted with COVID-19. This was a retrospective, single-centre, observational study in patients with confirmed COVID-19 presenting with non-iatrogenic PTX/PM from April 2020 to May 2021. Details pertaining to demographics, presentation, radiological characteristics, management and outcome were collected. Cases were classified into spontaneous and barotraumatic PTX/PM and a between-group comparison was performed using Chi-square and t-test. A total of 45 cases (mean age: 53.2 years, 82% males) out of 8,294 confirmed COVID-19 patients developed PTX/PM, the calculated incidence being 0.54%. 29 cases had spontaneous PTX/PM and the remaining 17 cases were attributed to barotrauma. The most common comorbidities were diabetes-mellitus (65.3%) and hypertension (42.3%). The majority of the cases had large PTX (62.1%) with tension in 8 cases (27.5%). There were predominant right-sided pneumothoraces and five were diagnosed with bronchopleural fistula. 37.7% of cases had associated subcutaneous emphysema. The median duration of PTX/PM from symptom onset was delayed at 22.5 and 17.6 days respectively. The mean CT severity score (CTSS) was 20.5 (± 4.9) with fibrosis (53.8%), bronchiectatic changes (50%) and cystic-cavitary changes (23%). There was no statistically significant difference between the spontaneous and barotrauma cohort. 71% of cases died and the majority belonged to the barotrauma cohort. It is imperative to consider the possibility of PTX/PM in patients having COVID-19, especially in those with deterioration in the disease course, both in spontaneously breathing and mechanically ventilated patients. These patients may also have a high incidence of death, reflecting the gravity of COVID-19.

COVID-19 presenting as pneumatoceles and spontaneous cavitary lesion as a late complication.

We report a case of an adult patient with COVID-19 pneumonia presenting as pneumatoceles as a late complication. These pneumatoceles are steroid-resistant and can predispose to cavitary lesions. These cystic lesions need close follow-up with repeat imaging as these can increase the risk of pneumothorax. It can take up to around 12 weeks for the spontaneous resolution of pneumatoceles.

Increased frequency of pneumothorax and pneumomediastinum in COVID-19 patients admitted in the ICU: A multicentre study from Mumbai, India.

BACKGROUND: There are limited data regarding the incidence of pneumothorax in COVID-19 patients as well as the impact of the same on patient outcomes. METHODS: A retrospective review of the medical records at three large tertiary care hospitals in Mumbai was performed to identify patients hospitalised with COVID-19 from March 2020 to October 2020. The presence of pneumothorax and/or pneumomediastinum was noted when chest radiographs or CT scans were performed. Demographic and clinical characteristics of patients who developed air leak were recorded. RESULTS: 4,906 patients with COVID-19 were admitted, with 1,324 (27%) having severe COVID-19 disease. The overall incidence of pneumothorax and/or pneumomediastinum in patients with severe disease was 3.2% (42/1,324). Eighteen patients had pneumothorax, 16 had pneumomediastinum and 8 patients had both. Fourteen patients (33.3%) developed this complication breathing spontaneously, 28 patients (66.6%) developed it during mechanical ventilation. Overall mortality in this cohort was 74%, compared with 17% in the COVID-19 patients without pneumothorax (p<0.001). CONCLUSIONS: Our study demonstrates that air leaks occur with a higher frequency in patients with COVID-19 than in other ICU patients. When present, such air leaks contributed to poor outcomes with almost 74% mortality rates in these patients.