Diagnosis Of Pneumothorax Research Articles

Efficacy of portable ultrasonography for early detection of pneumothorax following lung biopsy.

Pneumothorax is a notable complication of lung biopsy, and its early detection is crucial. This study aimed to compare the sensitivities of handheld portable lung ultrasonography and chest radiography for identifying early pneumothorax after lung biopsy. Both upright chest radiography and lung ultrasonography were performed 3 and 24 hours after lung biopsy. The disappearance of lung sliding and the appearance of lung points were considered evidence of pneumothorax on lung ultrasonography. Of the 86 patients in this study, 23 were diagnosed with pneumothorax within 24 hours of biopsy. No significant differences were observed in sex, age, or baseline lung function between the pneumothorax and non-pneumothorax groups. The sensitivities of lung ultrasonography and chest radiography for pneumothorax detection were 73.9% and 47.8%, respectively, at 3 hours and 91.3% and 78.3%, respectively, at 24 hours. Furthermore, at 3 hours, the area under the curve for lung ultrasonography for diagnosing pneumothorax was significantly higher than that for chest radiography (0.870 vs. 0.739, p = 0.043); however, the difference was not significant at 24 hours (p = 0.254). These preliminary results show lung ultrasonography is more sensitive than chest radiography to detect early pneumothorax after lung biopsy and may be advantageous for the rapid diagnosis of pneumothorax.

Pneumothorax in patients with COVID-19 pneumonia in the intensive care unit: an indicator of poor prognosis

Aims: The objective of this study was to investigate whether the incidence and development of pneumothorax in patients hospitalized in the intensive care unit (ICU) with coronavirus disease 2019 (COVID-19) pneumonia is associated with patient prognosis. Methods: This retrospective, cohort, descriptive study was initiated following approval from the ethics committee. The study was conducted on patients with confirmed COVID-19 pneumonia admitted to the tertiary ICU between March 2020 and March 2022. Data were collected from the patient registry system and ICU files. The patients were divided into two groups: those who developed pneumothorax and those who did not. The factors associated with mortality in the ICU were evaluated by univariate analysis and multiple logistic regression analysis. Results: The study included a total of 397 patients with confirmed cases of COVID-19 infection and pneumonia who were admitted to the ICU between March 2020 and March 2022. The mean age of the patients was 62±15 years. Of the patients, 56.1% were male. Pneumothorax was identified in 6.8% of patients. In addition to pneumothorax, six patients (1.5%) exhibited pneumomediastinum. The mortality rate was observed to be 40.5% among the total patient population. The mortality rate was 81.5% in the group with pneumothorax and 37.6% in the group without pneumothorax. The median time to mortality was 6 days (range 1–29 days) following the diagnosis of pneumothorax. Pneumothorax alone increased the likelihood of mortality in the ICU sevenfold (OR 7.3, 95% CI=2.70–19.75) and twofold when other variables were taken into account (OR 2, 95% CI=0.57-6.99). Conclusion: Pneumothorax is a common and fatal complication affecting mortality in patients with COVID-19 pneumonia in the ICU, despite the use of protective ventilation strategies. Particular caution should be exercised in patients receiving respiratory support in the ICU and in patients with a severe inflammatory response.

DICOM LUT is a Key Step in Medical Image Preprocessing Towards AI Generalizability.

Image pre-processing has significant impact on performance of deep learning models in medicine; yet, there is no standardized method for DICOM pre-processing. In this study, we investigate the impact of two commonly used image preprocessing techniques, histogram equalization (HE) and values-of-interest look-up-table (VOI-LUT) transformations on the performance deep learning classifiers for chest X-rays (CXR). We generated two baseline datasets (raw pixel and standard DICOM processed) from our internal CXR dataset and then enhanced both with HE to create four distinct datasets. Four independent deep learning models for diagnosis of pneumothorax were trained and evaluated on two external datasets. Results reveal that HE enhancement significantly affects model performance, particularly in terms of generalizability. Models trained solely on HE-enhanced datasets exhibit poorer performance on external validation sets, suggesting potential overfitting and information loss. These models also exhibit shortcut learning, relying on spurious correlations in the training data for their prediction. This study highlights the importance of machine learning practitioners being aware of preprocessing techniques applied to datasets and their potential impacts on model performance, as well as need for including preprocessing information when sharing datasets. Additionally, this research underscores the necessity of using pixel values closer to clinical standards during dataset curation to improve model robustness and mitigate the risk of information loss.

Lung ultrasound in neonates and children with cardiac diseases with focus on post cardiac surgical period: time for systematic use-an expert opinion report by the Association for European Paediatric and Congenital Cardiology Imaging Working Group.

Despite lung ultrasound (LUS) gaining consensus for the diagnosis of pulmonary complication in paediatric acute care setting and in adult cardiology, its use in paediatric cardiology remains limited. The aim of the present investigation is to provide an expert opinion on the applications of LUS in neonates and children with congenital heart disease, with a special focus on the post-surgical period. A complete guide for identification of landmarks and major signs (A and B lines) and their characteristics is provided. Diagnostic criteria, tips, and tricks for the diagnosis, and differential diagnosis of common pulmonary diseases such as pleural effusion, pneumonia, and consolidation are provided. To perform diagnosis of pneumothorax is illustrated. Applications of LUS for evaluation of hemidiaphragm motility and for a comprehensive assessment of retrosternal area are also discussed. The use of LUS for guidance of minor, common interventional procedures such as lung recruitment and drainage insertion is also described. The report also highlights current gaps of knowledge, including the difficulty for quantitative estimation of pleural effusion and atelectasis, and future prospective. There is sufficient evidence to support a systematic use of LUS for the diagnosis and follow-up of neonates and children with cardiac disease especially those undergoing paediatric cardiac surgery. LUS is an easy, accurate, fast, cheap, and radiation-free tool that should become a routine in daily practice.

A New Evaluation Model for Traumatic Severe Pneumothorax Based on Interpretable Machine Learning

Traumatic pneumothorax is a complex condition that is challenging to diagnose, particularly in hospitals, underdeveloped areas, and during mass casualty events. This study aimed to evaluate the potential of machine learning (ML) for diagnosing and assessing traumatic pneumothorax. We extracted 33 vital signs and blood gas parameters from the MIMIC-IV database, selecting 12 clinically significant features as inputs to four ML algorithms: extreme gradient boosting (XGBoost), artificial neural network (ANN), support vector machine (SVM), and k-nearest neighbors (KNN). Five-fold cross-validation was used to train and test the models, with external validation performed on the EICU database. Model performance was evaluated using AUROC, recall, and accuracy, with SHAP interpretability employed to understand feature importance. In total, 3871 participants from the MIMIC-IV database and 22,022 participants from the EICU database were analyzed. Hemoglobin, Oxygenation Index, and pH were found to be key indicators of severe traumatic pneumothorax. XGBoost exhibited the best performance, achieving an AUROC of 0.979 (95% CI: [0.966, 0.989]) on the MIMIC-IV dataset and 0.806 (95% CI: [0.740, 0.864]) on the EICU dataset. The results suggest that ML, particularly XGBoost, is faster and more convenient than traditional imaging methods, making it well-suited for emergency or mass casualty situations. ML algorithms show promise for initial diagnosis of traumatic pneumothorax, with XGBoost demonstrating strong interpretability and robust external validation.

Diagnostic accuracy of prehospital ultrasound in detecting lung injury in patients with trauma: a systematic review and meta-analysis

BackgroundUltrasound is now readily available in the prehospital setting and its use has been highlighted as one of the top research priorities in prehospital care. Clinical examination remains the standard...

Radiomics-based machine learning for automated detection of Pneumothorax in CT scans.

The increasing complexity of diagnostic imaging often leads to misinterpretations and diagnostic errors, particularly in critical conditions such as pneumothorax. This study addresses the pressing need for improved diagnostic accuracy in CT scans by developing an intelligent model that leverages radiomics features and machine learning techniques. By enhancing the detection of pneumothorax, this research aims to mitigate diagnostic errors and accelerate the process of image interpretation, ultimately improving patient outcomes. Data used in this study was extracted from the medical records of 175 patients with suspected pneumothorax. The collected images were preprocessed in Matlab software. Radiomics features were extracted from each image and finally, the machine learning models were implemented on these features. The used machine learning algorithms are Gradient Tree Boosting (GBM), eXtreme Gradient Boosting (XGBoost), and Light GBM. To evaluate the performance of models, various evaluation criteria such as precision, accuracy, specificity, sensitivity, F1 score, Area Under the Receiver Operating Characteristic (ROC) Curve (AUC), and misclassification were calculated. According to the calculated evaluation criteria, in terms of accuracy, the Gradient Boosting Machine (GBM) model achieved the highest performance with an accuracy of 98.97%, followed closely by the XGBoost model at 98.29%. For precision, the GBM model outperformed the other models, recording a precision value of 99.55%. Regarding sensitivity, all three models-GBM, XGBoost, and LightGBM (LGBM)-demonstrated strong performance, with sensitivity values of 99%, 99%, and 100%, respectively, indicating minimal variation among them. The artificial intelligence models used in this study have significant potential to enhance patient care by supporting radiologists and other clinicians in the diagnosis of pneumothorax. These models can facilitate the prioritization of positive cases, expedite evaluations, and ultimately improve patient outcomes.

Features of Diagnosis and Treatment of Endometriosis-Associated Pneumothorax (Clinical Case)

Features of Diagnosis and Treatment of Endometriosis-Associated Pneumothorax (Clinical Case)

Impact of Hyaneb in the management of severe acute bronchiolitis in pediatric patients: a single-center retrospective study

BackgroundTo determine the impact of Hyaneb (7% hypertonic saline and 0.1% hyaluronic acid) on severe acute bronchiolitis management in pediatric patients. The aim of this study is to evaluate the possible benefits and safety of Hyaneb in managing severe acute bronchiolitis.MethodsThis retrospective observational study was conducted in a tertiary healthcare center. The investigation included patients aged < 2 years who were admitted to the pediatric intensive care unit (PICU) of the Clínico San Carlos Hospital. Patients were excluded if they had a radiological diagnosis of pneumothorax on admission, infection by other microorganisms, bronchiolitis for > 5 days, need for respiratory or hemodynamic support with extracorporeal techniques, or known allergies or intolerance to any Hyaneb compounds. We evaluated the potential benefit and safety of using Hyaneb in severe acute bronchiolitis between 2019 and 2020 by comparing two groups of patients: one that received Hyaneb and the other that did not.ResultsA total of 49 children were included in the study, and 51.7% received Hyaneb. The median age was 1 month (interquartile range [IQR] 1.5–2.0), and the median weight was 4.7 ± 1.3 kg. Patient history included premature birth in 10% of the sample population. The results were not significant for the length of stay in the PICU and the respiratory support score (p = 0.114 and p = 0.06, respectively). However, there was a significant difference (p = 0.047) in the use of adrenaline rescue between the groups. These data were no longer significant when adjusted for age, Wood Downes scale, and weight (p = 0.056).ConclusionsUsing nebulized Hyaneb seems safe in patients less than a month old and weighing less than 5 kg. A reduction in PICU stay or respiratory support length could not be confirmed.

Automated Analysis of Ultrasound for the Diagnosis of Pneumothorax: A Systematic Review.

Automated analysis of lung ultrasound for pneumothorax detection is an emerging technology with potential applications across various clinical settings. This systematic review aims to evaluate the current evidence for the efficacy of automated analysis techniques in diagnosing pneumothorax using lung ultrasound images and videos. A literature search was conducted in the MEDLINE (Medical Literature Analysis and Retrieval System Online), Embase (Excerpta Medica Database), Web of Science, and Scopus databases up to July 5, 2024. Two reviewers screened articles, extracted data, and assessed the risk of bias using a modified QUADAS-2 (the revised version of the Quality Assessment of Diagnostic Accuracy Studies) tool. Ten studies met the inclusion criteria, with nine published after 2020. Eight studies were retrospective, and nine used physician analysis of ultrasound as the reference standard. Convolutional neural networks were the predominant classification method. The majority of studies analysed the presence or absence of lung sliding. Significant heterogeneity in study designs, outcome metrics, and methodology limited direct comparisons of efficacy. Automated analysis of lung ultrasound for pneumothorax detection shows promise; however, the field is still in its early stages. Challenges identified include the limitations of lung sliding as a diagnostic feature, variability in ultrasound hardware and protocols, and accounting for low-quality scans. Future research should focus on larger, more diverse datasets, standardized evaluation metrics, and robust external validation.

Pleural ultrasound for pneumothorax diagnosis after computerised tomography-guided biopsy

Pleural ultrasound for pneumothorax diagnosis after computerised tomography-guided biopsy

Treatment Strategy for Catamenial and Endometriosis-Related Recurrent Spontaneous Pneumothorax

RELEVANCE. The problem of treating thoracic endometriosis complicated by recurrent spontaneous catamenial pneumothorax is associated with the widespread prevalence of endometriosis, the complexity of its treatment and unsatisfactory long-term outcomes. The diagnosis of catamenial pneumothorax can only be established with a carefully collected history and confirmation of its recurrent nature. Favorable results are achieved with complex therapy carried out by the thoracic surgeon and gynecologist, consisting of an adequate volume of surgical intervention and hormonal therapy with gonadotropin releasing hormone agonists.AIM OF STUDY. To improve the diagnosis of thoracic endometriosis and recurrent catamenial and endometriosis-related spontaneous pneumothorax. To optimize the currently available treatment tactics by reducing the incidence of early relapses of catamenial pneumothorax and creating favorable conditions for the formation of reliable pleurodesis.MATERIAL AND METHODS. We studied the immediate outcomes of complex treatment of 30 patients (14.7%) with right-sided recurrent catamenial (28) and endometriosis-related spontaneous pneumothorax (2), with a median age of 41 (37;44) years, for the period from 2011 to the first half of 2023. Surgical treatment via video-assisted thoracoscopic access was performed in 26 patients (86.7%). After surgery, all the patients were recommended a six-month course of hormonal therapy with gonadotropin releasing hormone agonists.RESULTS. Various types of mechanical pleurodesis were performed in 25 patients (96.1%), resection of the diaphragm — in 17 (65.4%), alloplastic reconstruction with a synthetic mesh implant — in 12 patients (46.1%), pulmonary resection — in 9 (34.6%). In 12 of 26 patients (46.1%) in the early postoperative period, 6 (3; 6.75) days after surgery, a recurrence of catamenial pneumothorax developed, which subsequently required repeated interventions. In 5 patients, the relapse clearly coincided with the menstrual cycle. Early recurrence of pneumothorax required repeated drainage of the pleural cavity in all the patients, chemical pleurodesis in 2, and prolonged drainage of the pleural cavity in 1.CONCLUSION. Surgeries involving resection and reconstruction of the diaphragm with a mesh implant, pleurectomy, lung resection do not ensure the removal of all endometrioid heterotopias of the diaphragm and lung, therefore, during the first mensis after surgery, in the early postoperative period, 46.1% of the patients had a recurrence of catamenial pneumothorax, with no formed pleurodesis, which contributes to the possibility of further recurrences of pneumothorax. Improving the immediate outcomes of complex treatment of catamenial pneumothorax is possible with the preventive administration of hormonal therapy 1.5–2 months before the planned surgical intervention, which requires a multidisciplinary approach of the thoracic surgeon and gynecologist.

Clinical challenges and management of late presenting congenital diaphragmatic hernia mimicking tension pneumothorax in a child: a case report and review of literatures

BackgroundCongenital diaphragmatic hernia(CDH) is a rare congenital anomaly characterized by herniation of abdominal contents into thoracic cavity through a defect in diaphragm. While commonly diagnosed prenatally or in neonatal period, late-presenting CDH can occur and may mimic other thoracic emergencies such as tension pneumothorax, complicating diagnosis and management.Case presentationA two-year old male black child from Ethiopia presented to the emergency department with sudden onset of acute respiratory distress. Initial clinical assessment and chest radiography suggested a diagnosis of tension pneumothorax due to the presence of significant mediastinal shift and apparent pleural air. Despite insertion of chest tube, the child’s condition did not improve, raising suspicion of alternative diagnosis. Careful observation of initial chest x-ray and subsequent chest ultrasound revealed a left sided congenital diaphragmatic hernia with herniation of stomach and intestine into thoracic cavity compressing the left lung and causing mediastinal shift. After the diagnosis of CDH was confirmed, the child was stabilized and emergent surgical repair performed. Postoperative recovery was uneventful, and the child was discharged with no significant long-term complications.ConclusionThis case underscores the importance of considering CDH in the differential diagnosis of acute respiratory distress in a child. It highlights the diagnostic challenges and potential risks of emergency interventions based on initial misdiagnosis. Even if x -ray looks like typical of tension pneumothorax, it showed giant cystic air filled structure pushing the mediastinal structure to contralateral side with loss of left diaphragmatic outline which raised suspicion of congenital cystic lung mass or congenital diaphragmatic hernia. Advanced imaging and high index of suspicion are crucial for accurate diagnosis and timely management, ultimately improving patient outcomes. Consideration of alternative diagnosis when our initial intervention with insertion of chest tube fail to provide symptom improvement in suspected pneumothorax should raise suspicion of congenital diaphragmatic hernia like in our case.

Combat Medic-Performed Auscultation Versus Thoracic Ultrasound Image Interpretation for Pneumothorax Detection: Look or Listen?

Background Pneumothorax (PTX) is a potentially life-threatening condition encountered by U.S. Army combat medics on the battlefield. PTX, when left untreated, can progress to tension PTX, a leading cause of preventable death on the battlefield which can be difficult to diagnose based on physical exam alone due to variable physical exam findings. Prior literature shows medics can accurately use point-of-care ultrasound (POCUS) to diagnose PTX in cadaver and porcine models; however, no studies have directly compared the diagnostic accuracy of ultrasound (US) and physical exams performed by this population. We sought to compare medic diagnostic accuracy of a simulated PTX using these two diagnostic modalities. Methodology We conducted a prospective, observational study in which medics from the Flight Paramedic Program at Joint Base San Antonio-Sam Houston, TX received a standardized 30-minute training on physical exam and US diagnosis of PTX, followed by a 20-minute hands-on US familiarization session. Participants were then randomly selected into cohorts to evaluate 12 high-fidelity manikin lung fields and 12 thoracic US video clips for the presence of PTX in a simulated combat environment. Cohorts evaluated the same manikins and thoracic US video clips but in opposing sequences. Our primary outcome compared the sensitivity and specificity of PTX identification via thoracic US image interpretation and physical exam. Results In total, 21 medics evaluated 252 hemithoraces and interpreted 252 thoracic US images.We found a statistically significant difference favoring medics' sensitivity with US image interpretation over physical exam to detect PTX (85.7% vs. 72.4%, p = 0.004). There was no statistically significant difference in specificity between these modalities (81.9%vs. 69.4%, p = 0.139). Conclusions After brief training, medics demonstrated greater sensitivity to detect PTX using thoracic US image interpretation compared to physical exam in a simulated combat environment.Further studies involving medics' ultrasound image acquisition capability in human patients in austere combat environments are warranted.

ANÁLISE COMPARATIVA ENTRE A EFICÁCIA DIAGNÓSTICA DA RADIOGRAFIA E ULTRASSONOGRAFIA EM PACIENTES COM PNEUMOTÓRAX

INTRODUCTION: Pneumothorax is a serious medical condition, caused by the presence of air in the pleural cavity, which can lead to serious complications and even be fatal if not treated quickly. Chest radiography is the traditional diagnostic method, but it has limitations, and in this context, ultrasound has stood out as a promising alternative for the diagnosis of pneumothorax, having high sensitivity and specificity in some contexts. OBJECTIVES: This article aims to compare the diagnostic accuracy of ultrasound and chest radiography in the emergency department. METHODOLOGY: We carried out a broad search in several academic databases, using several Health Sciences Descriptors. The objective was to retrospectively compare the diagnostic accuracy of pneumothorax between radiography and ultrasound performed by non-radiologists in emergency patients. RESULTS: This study analyzed the effectiveness of chest radiography and ultrasound in diagnosing pneumothorax. 10 studies were included, 9 of which used the patient as the unit of analysis, with 1170 patients, and 460 with confirmed traumatic pneumothorax. 80% of studies presented a high risk of bias in patient selection. There was heterogeneity in the sensitivity of chest radiography, indicating the influence of the size of the pneumothorax, clinical conditions, and imaging techniques. Ultrasonography showed greater sensitivity and specificity than radiography; however, the results highlight the importance of considering different clinical contexts when choosing the best imaging modality to diagnose pneumothorax in emergencies. CONCLUSION: Thoracic ultrasound performed by non-radiologists is more accurate in diagnosing pneumothorax in emergency room trauma patients than chest radiography. The results highlight the potential of ultrasound as an effective diagnostic tool, easily integrating into trauma care protocols. Its use in medical training programs can improve the speed and accuracy of diagnosing critical conditions such as pneumothorax. The routine use of ultrasound in trauma management can bring positive changes to clinical practice, offering faster and safer interventions to patients.

Factors for increasing positive predictive value of pneumothorax detection on chest radiographs using artificial intelligence

This study evaluated the positive predictive value (PPV) of artificial intelligence (AI) in detecting pneumothorax on chest radiographs (CXRs) and its affecting factors. Patients determined to have pneumothorax on CXR by a commercial AI software from March to December 2021 were included retrospectively. The PPV was evaluated according to the true-positive (TP) and false-positive (FP) diagnosis determined by radiologists. To know the factors that might influence the results, logistic regression with generalized estimating equation was used. Among a total of 87,658 CXRs, 308 CXRs with 331 pneumothoraces from 283 patients were finally included. The overall PPV of AI about pneumothorax was 41.1% (TF:FP = 136:195). The PA view (odds ratio [OR], 29.837; 95% confidence interval [CI], 15.062–59.107), high abnormality score (OR, 1.081; 95% CI, 1.066–1.097), large amount of pneumothorax (OR, 1.005; 95% CI, 1.003–1.007), presence of ipsilateral atelectasis (OR, 3.508; 95% CI, 1.509–8.156) and a small amount of ipsilateral pleural effusion (OR, 5.277; 95% CI, 2.55–10.919) had significant effects on the increasing PPV. Therefore, PPV for pneumothorax diagnosis using AI can vary based on patients’ factors, image-acquisition protocols, and the presence of concurrent lesions on CXR.

Low Diagnostic Accuracy of Transthoracic Ultrasound for the Assessment of Spontaneous Pneumothorax in the Emergency Setting: A Multicentric Study.

Background: Pneumothorax (PNX) represents a common clinical condition in emergency departments (EDs), requiring prompt recognition and treatment. The role of transthoracic ultrasounds (TUSs) in the diagnosis of PNX is still debated. We aimed to prospectively evaluate the accuracy of TUSs in the detection of spontaneous PNX in EDs. Methods: A total of 637 consecutive adult patients who presented to the EDs of four Italian hospitals complaining of acutely onset chest pain and dyspnoea were included in the study. Exclusion criteria were previous traumatic events, cardiogenic causes of pain/dyspnoea and suspected tension PNX. The absence of "lung sliding" (B-mode) and the "bar-code" sign (M-mode) were considered indicative of PNX in a TUS. Accuracy, sensitivity, specificity, and positive and negative predictive values (PPVs, NPVs) were calculated using a chest CT scan as reference. Results: Spontaneous PNX occurred in 93 patients: of those, 83 (89.2%) were correctly identified by TUSs. However, 306 patients with suspected PNX at TUS were not confirmed by chest CTs. The diagnostic accuracy of both the absence of "lung sliding" and "bar-code" sign during TUS was 50.4% (95% CI: 46.4-54.3), sensitivity was 89.2% (95% CI: 81.1-94.7), specificity was 43.8% (95% CI: 39.5-48.0), the PPV was 21.3% (95% CI: 19.7-23.1) and the NPV was 96.0% (95% CI: 92.9-97.7). Conclusions: TUS showed high sensitivity but low specificity in the identification of PNX in EDs. Relying exclusively on TUSs results for patients' management in ED settings is neither suitable nor recommendable. TUS examination can be useful to strengthen the clinical suspicion of PNX, but its results should be confirmed by a chest X-ray or CT scan.

Recurrent spontaneous pneumothorax secondary to lung cystic lesions in a case of convalescent COVID-19: a case report and literature review

BackgroundWhile spontaneous pneumothorax has been documented in COVID-19 patients, reports on recurrent spontaneous pneumothorax due to cystic lesions in convalescent COVID-19 patients are scarce. The progression of these lung cystic lesions remains inadequately explored.Case presentation and literature reviewAn 81-year-old male, a non-smoker with a history of rheumatoid arthritis, presented with fever, cough, and expectoration for 14 days. Initially diagnosed with moderate COVID-19, he deteriorated to severe COVID-19 despite adherence to local treatment guidelines. Successive identification of three cystic lesions termed “bulla” or “pneumatocele”, and one cystic lesion with air-fluid level, referred to as “pneumo-hamatocele” (PHC), occurred in his lungs. Gradual improvement followed anti-inflammatory therapy and optimal supportive care. However, on day 42, sudden worsening dyspnea prompted a computed tomography (CT) scan, confirming a right spontaneous pneumothorax and subcutaneous emphysema, likely due to PHC rupture. Discharge followed chest tube implementation for pneumothorax resolution. On day 116, he returned to the hospital with mild exertional dyspnea. Chest CT revealed recurrent right pneumothorax from a remaining cyst in the right lung. Apart from our patient, literature retrieval identified 22 COVID-19 patients with spontaneous pneumothorax due to cystic lesions, with a male predominance (95.6%; 22/23). Diagnosis of pneumothorax and lung cystic lesions occurred around day 29.5 (range: 18–35) and day 26.4 (± 9.8) since symptom onset, respectively. Except for one patient whose pneumothorax occurred on day seven of illness, all patients eventually recovered.ConclusionsRecurrent spontaneous pneumothorax secondary to lung cystic lesions may manifest in convalescent COVID-19 patients, particularly males with COVID-19 pneumonia. Chest CT around 2 to 3 weeks post-symptom onset may be prudent to detect cystic lesion development and anticipate spontaneous pneumothorax.

Detecting Pneumothorax on Chest Radiograph Using Segmentation with Deep Learning

Introduction: Pneumothorax is a life-threatening condition that requires prompt recognition and therapy to prevent deterioration. Radiologist workload often precludes rapid assessment of the usual diagnostic modality, the chest radiograph, particularly after hours. The aim was to develop a deep learning model using a segmentation-based Deep Convolutional Neural Network (DCNN) to detect pneumothorax on chest radiographs to provide rapid and accurate pneumothorax diagnosis. Methods: This is a retrospective study of spontaneous pneumothorax at a single center, containing 130 positive and 70 negative radiographs. Subsequent manual contour mapping was performed to draw a mask of the pneumothorax. These image pairs were used to train a DCNN model (a modified AlexNet) after pretraining on the ImageNet dataset. Results: The DCNN achieved an accuracy of 0.83, with sensitivity of 98.1%, and specificity of 68.5%. Conclusion: This segmentation-based DCNN accuracy is comparable to previous categorization-based CDNN models, despite using a smaller sample size for training, while including the benefits of visual representation for clinician feedback. Segmentation-based DCNNs show promise in the development of accurate and clinically useful models for medical imaging.

The Value of BLUE Protocol in the Diagnosis of Acute Onset of Respiratory Distress in Critically Ill Patient

Abstract To detect the value of BLUE protocol in the diagnosis of acute onset of respiratory distress in critically ill patients This cross-sectional study conducted on 60 patients by performing ultrasonography on consecutive patients admitted to the ICU with acute onset of respiratory distress (dyspnea), comparing lung ultrasonography results on initial presentation with the final diagnosis with conventional method of diagnosis by the ICU team. The BLUE protocol proved its importance, availability, easiness of application, accuracy and its ability to reach a final correct diagnosis in 90%of the patients. The lung U/S showed much better results than the CXR, but still the CT chest was more superior to the lung U/S as the sensitivity of the BLUE protocol in the diagnosis of pneumonia, pulmonary edema, pneumothorax, pleural effusion and COPD was 91.4%, 92.8%, 90%, 88.8% and 78.6 respectively. The lung U/S (BLUE protocol) is a bedside, reliable, dynamic, rapid, and noninvasive technique, these results suggest it could have a significant role in the diagnostic workup of acute onset of respiratory distress in the ICU.