Detection Of Pneumothorax Research Articles

DIAGNOSTIC ACCURACY OF CHEST X-RAY IN THE DETECTION OF PNEUMOTHORAX IN BLUNT CHEST TRAUMA PATIENTS WITH CT AS THE GOLD STANDARD

Background: To study the diagnostic accuracy of chest X-ray in the detection of pneumothorax in blunt chest trauma patients with CT as the Gold Standard
 Methods: The present study was conducted from 31th July 2018 to 30th July 2019. A total of 36 patients were enrolled in the study.
 Results: On Chest X-Ray Supine AP view, pneumothorax was detected in 11 of 24 patients. The sensitivity of Chest X-Ray Supine AP view was 45.83%, specificity was 100%, positive predictive value (PPV) was 100%, negative predictive value (NPV) was 48% and accuracy was 63.89% for the diagnosis of pneumothorax.
 Conclusion: A Chest radiograph is the most preferred and relevant primary investigation in the diagnosis of pneumothorax even with the various advanced techniques that are available. X-ray being relatively cheaper and is easily available even at the peripheral centers at the primary health care level.
 Keywords: X-ray, CT, Pneumothorax

DIAGNOSTIC ACCURACY OF ULTRASOUND IN THE DETECTION OF PNEUMOTHORAX IN CHEST TRAUMA PATIENTS WITH CT AS THE GOLD STANDARD

Background: To study the diagnostic accuracy of ultrasound in the detection of pneumothorax in chest trauma patients with CT as the Gold Standard
 Methods: The present study was conducted from 31th July 2018 to 30th July 2019. A total of 36 patients were enrolled in the study.
 Results: By chest ultrasound, pneumothorax was detected in 15 of 24 patients. The sensitivity of chest ultrasound for the diagnosis of pneumothorax was 62.5%, specificity was 100%, positive predictive value (PPV) was 100%, negative predictive value (NPV) was 54.14% and accuracy was 75%.
 Conclusion: Chest ultrasound can play an important role in the emergency department aiding a physician for bedside rapid and accurate diagnosis of pneumothorax without interruption in the resuscitation process and without transferring the patient to the radiology section.
 Keywords: Ultrasound, CT, Pneumothorax

Deep Learning Systems for Pneumothorax Detection on Chest Radiographs: A Multicenter External Validation Study.

To assess the generalizability of a deep learning pneumothorax detection model on datasets from multiple external institutions and examine patient and acquisition factors that might influence performance. In this retrospective study, a deep learning model was trained for pneumothorax detection by merging two large open-source chest radiograph datasets: ChestX-ray14 and CheXpert. It was then tested on six external datasets from multiple independent institutions (labeled A-F) in a retrospective case-control design (data acquired between 2016 and 2019 from institutions A-E; institution F consisted of data from the MIMIC-CXR dataset). Performance on each dataset was evaluated by using area under the receiver operating characteristic curve (AUC) analysis, sensitivity, specificity, and positive and negative predictive values, with two radiologists in consensus being used as the reference standard. Patient and acquisition factors that influenced performance were analyzed. The AUCs for pneumothorax detection for external institutions A-F were 0.91 (95% CI: 0.88, 0.94), 0.97 (95% CI: 0.94, 0.99), 0.91 (95% CI: 0.85, 0.97), 0.98 (95% CI: 0.96, 1.0), 0.97 (95% CI: 0.95, 0.99), and 0.92 (95% CI: 0.90, 0.95), respectively, compared with the internal test AUC of 0.93 (95% CI: 0.92, 0.93). The model had lower performance for small compared with large pneumothoraces (AUC, 0.88 [95% CI: 0.85, 0.91] vs AUC, 0.96 [95% CI: 0.95, 0.97]; P = .005). Model performance was not different when a chest tube was present or absent on the radiographs (AUC, 0.95 [95% CI: 0.92, 0.97] vs AUC, 0.94 [95% CI: 0.92, 0.05]; P > .99). A deep learning model trained with a large volume of data on the task of pneumothorax detection was able to generalize well to multiple external datasets with patient demographics and technical parameters independent of the training data.Keywords: Thorax, Computer Applications-Detection/DiagnosisSee also commentary by Jacobson and Krupinski in this issue.Supplemental material is available for this article.©RSNA, 2021.

Clinical Validation Is the Key to Adopting AI in Clinical Practice.

Clinical Validation Is the Key to Adopting AI in Clinical Practice.

Automated Radiology Alert System for Pneumothorax Detection on Chest Radiographs Improves Efficiency and Diagnostic Performance.

We aimed to set up an Automated Radiology Alert System (ARAS) for the detection of pneumothorax in chest radiographs by a deep learning model, and to compare its efficiency and diagnostic performance with the existing Manual Radiology Alert System (MRAS) at the tertiary medical center. This study retrospectively collected 1235 chest radiographs with pneumothorax labeling from 2013 to 2019, and 337 chest radiographs with negative findings in 2019 were separated into training and validation datasets for the deep learning model of ARAS. The efficiency before and after using the model was compared in terms of alert time and report time. During parallel running of the two systems from September to October 2020, chest radiographs prospectively acquired in the emergency department with age more than 6 years served as the testing dataset for comparison of diagnostic performance. The efficiency was improved after using the model, with mean alert time improving from 8.45 min to 0.69 min and the mean report time from 2.81 days to 1.59 days. The comparison of the diagnostic performance of both systems using 3739 chest radiographs acquired during parallel running showed that the ARAS was better than the MRAS as assessed in terms of sensitivity (recall), area under receiver operating characteristic curve, and F1 score (0.837 vs. 0.256, 0.914 vs. 0.628, and 0.754 vs. 0.407, respectively), but worse in terms of positive predictive value (PPV) (precision) (0.686 vs. 1.000). This study had successfully designed a deep learning model for pneumothorax detection on chest radiographs and set up an ARAS with improved efficiency and overall diagnostic performance.

Pulmonary tuberculosis presenting as tension pneumothorax

Although detection of spontaneous pneumothorax during evolvement active pulmonary tuberculosis is not uncommon, presentation of pulmonary tuberculosis as tension pneumothorax is rare. There is limited literature regarding pneumothorax complicating pulmonary tuberculosis. Pneumothorax due to tuberculosis can be managed by intercostal drainage in the majority; only a minority needs surgical intervention. Many patients need intercostal drainage for a prolonged period. We report a patient presenting with a tension pneumothorax who was found to have active pulmonary tuberculosis. His pneumothorax was managed successfully with prolonged intercostal drainage. Lessons we learnt during the management of this patient is discussed.

Chest sonography versus chest radiograph in children admitted to paediatric intensive care - A prospective study.

There is a paucity of studies on the correlation between chest radiograph and ultrasound (US) in children. Our objective was to study the correlation between bedside chest radiograph and ultrasound findings in 413 children with 1002 episodes of chest radiograph and US enrolled for a prospective, double-blinded observational study in a multidisciplinary paediatric intensive care unit. Weighted κ statistic for agreement was different for right and left lungs and varied from 50% for left pleural effusion to 98% for right pneumothorax. Pulmonary oedema, pneumothorax and pleural effusion were diagnosed by ultrasound alone in a significantly higher number of patients as compared to chest radiograph (P = 0.001). Chest ultrasound is therefore deemed more sensitive than chest radiograph in detection of pleural effusion, pulmonary oedema and pneumothorax.

Detection and Semiquantitative Analysis of Cardiomegaly, Pneumothorax, and Pleural Effusion on Chest Radiographs.

To develop and evaluate deep learning models for the detection and semiquantitative analysis of cardiomegaly, pneumothorax, and pleural effusion on chest radiographs. In this retrospective study, models were trained for lesion detection or for lung segmentation. The first dataset for lesion detection consisted of 2838 chest radiographs from 2638 patients (obtained between November 2018 and January 2020) containing findings positive for cardiomegaly, pneumothorax, and pleural effusion that were used in developing Mask region-based convolutional neural networks plus Point-based Rendering models. Separate detection models were trained for each disease. The second dataset was from two public datasets, which included 704 chest radiographs for training and testing a U-Net for lung segmentation. Based on accurate detection and segmentation, semiquantitative indexes were calculated for cardiomegaly (cardiothoracic ratio), pneumothorax (lung compression degree), and pleural effusion (grade of pleural effusion). Detection performance was evaluated by average precision (AP) and free-response receiver operating characteristic (FROC) curve score with the intersection over union greater than 75% (AP75; FROC score75). Segmentation performance was evaluated by Dice similarity coefficient. The detection models achieved high accuracy for detecting cardiomegaly (AP75, 98.0%; FROC score75, 0.985), pneumothorax (AP75, 71.2%; FROC score75, 0.728), and pleural effusion (AP75, 78.2%; FROC score75, 0.802), and they also weakened boundary aliasing. The segmentation effect of the lung field (Dice, 0.960), cardiomegaly (Dice, 0.935), pneumothorax (Dice, 0.827), and pleural effusion (Dice, 0.826) was good, which provided important support for semiquantitative analysis. The developed models could detect cardiomegaly, pneumothorax, and pleural effusion, and semiquantitative indexes could be calculated from segmentations.Keywords: Computer-Aided Diagnosis (CAD), Thorax, CardiacSupplemental material is available for this article.© RSNA, 2021.

Deep Learning for Diagnosis and Segmentation of Pneumothorax: The Results on the Kaggle Competition and Validation Against Radiologists

Pneumothorax is potentially a life-threatening disease that requires urgent diagnosis and treatment. The chest X-ray is the diagnostic modality of choice when pneumothorax is suspected. The computer-aided diagnosis of pneumothorax has received a dramatic boost in the last few years due to deep learning advances and the first public pneumothorax diagnosis competition with 15257 chest X-rays manually annotated by a team of 19 radiologists. This paper describes one of the top frameworks that participated in the competition. The framework investigates the benefits of combining the Unet convolutional neural network with various backbones, namely ResNet34, SE-ResNext50, SE-ResNext101, and DenseNet121. The paper presents a step-by-step instruction for the framework application, including data augmentation, and different pre- and post-processing steps. The performance of the framework was of 0.8574 measured in terms of the Dice coefficient. The second contribution of the paper is the comparison of the deep learning framework against three experienced radiologists on the pneumothorax detection and segmentation on challenging X-rays. We also evaluated how diagnostic confidence of radiologists affects the accuracy of the diagnosis and observed that the deep learning framework and radiologists find the same X-rays to be easy/difficult to analyze (p-value <1e4). Finally, the methodology of all top-performing teams from the competition leaderboard was analyzed to find the consistent methodological patterns of accurate pneumothorax detection and segmentation.

Hemopneumothorax detection through the process of artificial evolution - a feasibility study

BackgroundTension pneumothorax is one of the leading causes of preventable death on the battlefield. Current prehospital diagnosis relies on a subjective clinical impression complemented by a manual thoracic and respiratory examination. These techniques are not fully applicable in field conditions and on the battlefield, where situational and environmental factors may impair clinical capabilities. We aimed to assemble a device able to sample, analyze, and classify the unique acoustic signatures of pneumothorax and hemothorax.MethodsAcoustic data was obtained with simultaneous use of two sensitive digital stethoscopes from the chest wall of an ex-vivo porcine model. Twelve second samples of acoustic data were obtained from the in-house assembled digital stethoscope system during mechanical ventilation. The thoracic cavity was injected with increasing volumes of 200, 400, 600, 800, and 1000 ml of air or saline to simulate pneumothorax and hemothorax, respectively. The data was analyzed using a multi-objective genetic algorithm that was used to develop an optimal mathematical detector through the process of artificial evolution, a cutting-edge approach in the artificial intelligence discipline.ResultsThe in-house assembled dual digital stethoscope system and developed genetic algorithm achieved an accuracy, sensitivity and specificity ranging from 64 to 100%, 63 to 100%, and 63 to 100%, respectively, in classifying acoustic signal as associated with pneumothorax or hemothorax at fluid injection levels of 400 ml or more, and regardless of background noise.ConclusionsWe present a novel, objective device for rapid diagnosis of potentially lethal thoracic injuries. With further optimization, such a device could provide real-time detection and monitoring of pneumothorax and hemothorax in battlefield conditions.

Most patient conditions do not a priori debilitate the sensitivity of thoracic ultrasound in thoracic surgery-a prospective comparative study

BackgroundThe few existing studies on the accuracy of lung ultrasound in the detection of a postoperative pneumothorax after thoracic surgery differ in the sonographic technique and the inclusion criteria. Several conditions are considered unfavourable in the sonographic examination of the lung. We aim to test these conditions for their impact on the diagnostic accuracy of lung ultrasound.MethodsWe compared lung ultrasound and chest roentgenograms for the detection of a pneumothorax after lung-resecting surgery in two prospective trials (register ID DRKS00014557 and DRKS00020216). The ultrasound examiners and radiologists were blinded towards the corresponding findings. We performed posthoc subgroup analyses to determine the influence of various patient or surgery related conditions on the sensitivity and specificity of ultrasound in the detection of pneumothorax.ResultsWe performed 340 examinations in 208 patients. The covariates were age, gender, body mass index, smoking status, severity of chronic obstructive pulmonary disease, previous ipsilateral operation or irradiation, thoracotomy, postoperative skin emphysema, indwelling chest tube and X-ray in supine position. In univariate analysis, an indwelling chest-tube was associated with a higher sensitivity (58%, p = 0.04), and a postoperative subcutaneous emphysema with a lower specificity (73% vs. 88%, p = 0.02). None of the other subgroups differed in sensitivity or specificity from the total population .ConclusionsMost of the patient- or surgery related conditions usually considered unfavourable for lung ultrasound did not impair the sensitivity or specificity of lung ultrasound. Further studies should not excluce patients with these conditions, but test the accuracy under routine conditions.Trial registrationDRKS, DRKS00014557, registered 06/09/2018, https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00014557 and DRKS00020216, registered 03/12/2019, https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00020216

Pneumothorax detection in chest radiographs: optimizing artificial intelligence system for accuracy and confounding bias reduction using in-image annotations in algorithm training

ObjectivesDiagnostic accuracy of artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXR) is limited by the noisy annotation quality of public training data and confounding thoracic tubes (TT). We hypothesize that in-image annotations of the dehiscent visceral pleura for algorithm training boosts algorithm’s performance and suppresses confounders.MethodsOur single-center evaluation cohort of 3062 supine CXRs includes 760 PTX-positive cases with radiological annotations of PTX size and inserted TTs. Three step-by-step improved algorithms (differing in algorithm architecture, training data from public datasets/clinical sites, and in-image annotations included in algorithm training) were characterized by area under the receiver operating characteristics (AUROC) in detailed subgroup analyses and referenced to the well-established “CheXNet” algorithm.ResultsPerformances of established algorithms exclusively trained on publicly available data without in-image annotations are limited to AUROCs of 0.778 and strongly biased towards TTs that can completely eliminate algorithm’s discriminative power in individual subgroups. Contrarily, our final “algorithm 2” which was trained on a lower number of images but additionally with in-image annotations of the dehiscent pleura achieved an overall AUROC of 0.877 for unilateral PTX detection with a significantly reduced TT-related confounding bias.ConclusionsWe demonstrated strong limitations of an established PTX-detecting AI algorithm that can be significantly reduced by designing an AI system capable of learning to both classify and localize PTX. Our results are aimed at drawing attention to the necessity of high-quality in-image localization in training data to reduce the risks of unintentionally biasing the training process of pathology-detecting AI algorithms.Key Points• Established pneumothorax-detecting artificial intelligence algorithms trained on public training data are strongly limited and biased by confounding thoracic tubes.• We used high-quality in-image annotated training data to effectively boost algorithm performance and suppress the impact of confounding thoracic tubes.• Based on our results, we hypothesize that even hidden confounders might be effectively addressed by in-image annotations of pathology-related image features.

The effect of a short ultrasound training course on the performance of emergency medical technicians in terms of using extended-focused assessment with sonography in trauma (E-FAST) in the pre-hospital setting

Objective: The aim of this study was to determine the effect of an ultrasound training course on the performance of emergency medical technicians (EMTs) in terms of using extended-focused assessment with sonography in trauma (E-FAST) in the pre-hospital setting. Methods: In this quasi-experimental study, 107 EMTs affiliated to Shahid Beheshti and Kashan Universities of Medical Sciences, were randomly selected. At first, the EMTs were given a pre-test of five images concerning normal and abnormal ultrasound views of thoracic and abdominal cavities. Then, 6 hours of theoretical and practical sessions were done by emergency medicine specialists to teach the EMTs to use ultrasound for the detection of pneumothorax and free fluid. The EMTs were tested again in the form of 18 images post-test as well as a 13-item survey to evaluate their skills in taking correct 10 E-FAST. Results: A total of 107 EMTs with the mean age of 27±15.21 and average of 3.5 years work experience participated in this study. The mean of test score increased from 0.57± 0.19 to 0.81±0.08 (with P &lt; 0.001) out of the total score of 100. In the 4-graded survey, 98% of EMTs obtained good to excellent grades in working with ultrasound machine. Also, 50.5% of EMTs were good in taking Morison’s view and 46.6% and 45.6% had medium grades in pericardial and splenorenal views, respectively. In addition, 22.3% had weak grades in taking pericardial views. Conclusion: The results of the present study showed that the implementation of the training course significantly improved the accuracy of free fluid detection in thoracic and abdominal cavities by EMTs.

Application of deep learning-based chest CT auxiliary diagnosis system in emergency trauma patients

Objective: To investigate the diagnostic efficacy and potential application value of deep learning-based chest CT auxiliary diagnosis system in emergency trauma patients. Methods: A total of 403 patients, including 254 males and 149 females aged from 16 to 100 (50±19) years, who received emergency treatment for trauma and chest CT examination in the Eastern Theater General Hospital from September 2019 to November 2019 were retrospectively analyzed. Dr. Wise Lung Analyzer's chest CT auxiliary diagnosis system was applied to detect 5 types of injuries, including pneumothorax, pleural effusion/hemothorax, pulmonary contusion (shown as consolidation and ground glass opacity), rib fractures, and other fractures (including thoracic vertebrae, sternum, scapula and clavicle, etc.) and 6 other abnormalities (bullae, emphysema, pulmonary nodules, stripe, reticulation, pleural thickening). The diagnostic reference standards were labeled by two radiologists independently. The sensitivity and specificity of the auxiliary diagnosis system were evaluated. The imaging diagnostic reports were compared with the results of the auxiliary diagnosis system, and the diagnostic consistency between the two was calculated by using the Kappa test. Results: According to the reference standards, among the 403 patients, 29 were pneumothorax, 75 were pleural effusion/hemothorax, 131 were pulmonary contusion, 124 were rib fractures, and 63 were other fractures. The sensitivity and specificity of the auxiliary diagnosis system for detection of pneumothorax, pleural effusion/hemothorax, rib fractures, and other fractures were 96.6%, 97.6%, 80.0%, 99.7%, 99.2%, 83.9%, 84.1%, and 99.7%, respectively. The sensitivity of detecting lung contusion was 97.7%. There was a high consistency between the auxiliary diagnosis system and imaging diagnosis in the diagnosis of injuries, in which the kappa values of pneumothorax, pleural effusion, rib fracture and other fractures were 0.783, 0.821, 0.706 and 0.813, respectively (all P<0.001). Two cases of pneumothorax, three cases of pleural effusion/hemothorax, nine cases of rib fractures, and six cases of other fractures missed by imaging diagnosis were all detected by the auxiliary diagnosis system. The detection sensitivity of the auxiliary diagnosis system was higher for emphysema, pulmonary nodules and stripe (all>85%), but lower for bullae, reticulation and pleural thickening. Conclusions: The deep learning-based chest CT auxiliary diagnosis system could effectively assist chest CT to detect injuries in emergency trauma patients, which was expected to optimize the clinical workflow.

Current Trends in the Use of Ultrasound Over Chest X-Ray to Identify Pneumothoraces in ICU, Trauma, and ARDS Patients.

Pneumothoraces are a common and potentially fatal complication for critically ill patients in the trauma and intensive care units. Since its use for pneumothorax detection was first reported in 1987, ultrasound has been increasingly used for the detection of thoracic injuries. As ultrasound imaging has improved and operators have potentially become more proficient, it is important to analyze more recent trends in the sensitivities and specificities of ultrasound for the detection of pneumothorax. This literature review and meta-analysis identifies 17 studies that directly compare the sensitivity and specificity of ultrasound and anterior-posterior chest x-ray in the identification of pneumothorax among 2955 patients who developed 793 pneumothoraces as detected by gold standard CT scanning. For the 17 articles analyzed, the pooled sensitivity of trans-thoracic ultrasound was 75.07% (64.92%-85.22%), and the pooled specificity was 98.36% (97.45%-99.26%). The pooled sensitivity of CXR was 45.65% (36.04%-55.26%), and pooled specificity was 99.62% (99.00%-100%). While this review demonstrates an improved sensitivity in the detection of pneumothorax with ultrasound over AP chest x-rays, it did not find a significant trend or improvement in the sensitivity or specificity of ultrasound for detecting pneumothorax over time.

Diagnostic accuracy of a lung ultrasound protocol (Vet BLUE) for detection of pleural fluid, pneumothorax and lung pathology in dogs and cats.

To assess the accuracy of the lung ultrasound protocol Vet BLUE, using thoracic CT as the reference standard, for the detection of thoracic pathology in dogs and cats. Animals that had thoracic ultrasound and thoracic CT were prospectively recruited between May 2017 and September 2018. The Vet BLUE protocol was performed on animals at the time of admission by veterinarians with basic training in emergency ultrasound. A board-certified radiologist, blinded to the Vet BLUE findings, reviewed the CT images. CT was abnormal in 64.5% (20/31) animals. The number of CT sites positive for alveolar-interstitial syndrome was 24.2% (60/248). When using CT as the reference standard, detection of ≥3 B lines with thoracic ultrasound had a sensitivity of 18.33% and specificity of 98.4% for detection of site specific alveolar-interstitial syndrome. The sensitivity of Vet BLUE to detect alveolar-interstitial syndrome increased to 56.9% when including the presence of any B line as abnormal. Overall accuracy for detection of alveolar-interstitial syndrome based on these two criteria was 79% and 73%, respectively. Vet BLUE correctly identified consolidation in 58.3% (14/24) sites, pleural effusion in 66.6% (2/3) cases, pneumothorax in 33.3% (1/3) cases and intrathoracic mass in 25% (1/4) cases. The Vet BLUE protocol is a useful technique to detect alveolar-interstitial syndrome and other thoracic pathology but should not be used as a sole imaging method. Detection of ≥3 B lines is highly suggestive of alveolar-interstitial syndrome and warrants further diagnostics.

Immediate Pneumothorax Diagnosis by Surgical Residents Using Portable Ultrasound.

Feasibility of diagnosis of pneumothorax using handheld ultrasound by non-radiologists shows inconsistent results. The aim of this study is to evaluate the feasibility and accuracy of portable ultrasound for immediate diagnosis of pneumothorax by general surgery residents who underwent short training. Patients who presented to the emergency department of a university hospital with suspected pneumothorax between 10/2018 and 12/2019 were included in the study. Patients underwent ultrasound in 2 points of each hemithorax. Sensitivity and specificity for pneumothorax diagnosis by ultrasound and physical examination were calculated and compared with chest computed tomography (CT). Patients in whom a chest tube was placed prior to ultrasound examination and those who did not undergo a CT scan were excluded from the study. A total of 85 patients met the inclusion criteria. Mean age was 40.7 ± 20.2 years. Pneumothorax was found among 46 patients (54%) per chest CT, and of these, 21 (46%) underwent chest tube placement following imaging. Ultrasound showed the highest sensitivity and specificity (95.6% [95% confidence interval {CI} 85.16% to 99.47%] and 97.44% [95% CI 86.40% to 99.67%], respectively). Chest x-ray had the lowest sensitivity (47.8% [95% CI 32.89% to 63.05%]) for pneumothorax detection. Physical examination showed a moderate sensitivity and specificity (82.6% [95% CI 68.58% to 92.18%] and 77.89% [95% CI 60.67% to 88.87%], respectively) for the diagnosis of pneumothorax. We found high accuracy rates of 2-point ultrasound in immediate pneumothorax diagnosis when performed by surgical residents who underwent a short ultrasound training. This is a fast and repeatable test, and has the potential for successful implementation in prehospital and military scenarios as well, minimizing unnecessary chest tube placements.

Automatic Diagnosis of Pneumothorax From Chest Radiographs: A Systematic Literature Review

Among various medical imaging tools, chest radiographs are the most important and widely used diagnostic tool for detection of thoracic pathologies. Research is being carried out in order to propose robust automatic diagnostic tool for detection of pathologies from chest radiographs. Artificial Intelligence techniques especially deep learning methodologies have found to be giving promising results in automating the field of medicine. Lot of research has been done for automatic and fast detection of pneumothorax from chest radiographs while proposing several frameworks based on artificial intelligence and machine learning techniques. This study summarizes the existing literature for the automatic detection of pneumothorax from chest x-rays along with describing the available chest radiographs datasets. The comparative analysis of the literature is also provided in terms of goodness. Limitations of the existing literature along with the research gaps is also given for further investigation. The paper provides a brief overview of the present work for pneumothorax detection for helping the researchers in selection of optimal approach for future research.

Diagnosis of Traumatic Pneumothorax: A Comparison between Lung Ultrasound and Supine Chest Radiographs.

Background/ObjectiveTraumatic pneumothorax is an ominous condition necessitating urgent appropriate action. It is typically detected on chest X-rays; however, these may not be able to detect the presence of a subtle pneumothorax, especially in supine position. Lung ultrasound is emerging as a promising modality for detecting pneumothorax in trauma patients. The aim of our study was to compare ultrasound with supine chest radiography for the detection of pneumothorax in trauma patients.Materials and MethodsThis was a prospective, single-blinded study carried out on 212 adult thoracoabdominal trauma patients who underwent ultrasound FAST and supine (AP) chest radiography. During the FAST sonography, ultrasound thorax was done to rule out pneumothorax. Only those cases were considered (118) in which the presence or absence of pneumothorax could be confirmed on CT done subsequently or where pneumothorax was confirmed by air escape on chest tube placement, wherever indicated, and the results were compared with sonographic and chest X-ray findings.Observation/ResultsThere were 48 true positives on CT/chest tube insertion. Among these, ultrasound was able to correctly detect pneumothorax in 43 patients, while supine chest X-rays correctly identified 33 cases. Sensitivity of ultrasound was 89.6 vs. 68.8% of supine chest radiography. Lung ultrasound also had a higher negative predictive value as compared to supine chest X-rays.ConclusionsLung ultrasound is more sensitive in detecting traumatic pneumothorax than supine chest X-rays, in addition to having numerous other inherent advantages over chest radiography. It should be incorporated in the emergency assessment of thoracic trauma patients to rule out pneumothorax.Clinical significanceLung sonography is more sensitive in detecting traumatic pneumothorax than supine chest X-rays. No added equipment is required, and the procedure can be carried out at the time of doing ultrasound FAST, thus saving precious time in trauma patients.How to cite this articleBhoil R, Kumar R, Kaur J, Attri PK, Thakur R. Diagnosis of Traumatic Pneumothorax: A Comparison between Lung Ultrasound and Supine Chest Radiographs. Indian J Crit Care Med 2021;25(2):176–180.

Sensitivity of lung ultrasound for the detection of pneumothorax one day after pulmonary resection—a prospective observational study

Summary Background Chest X‑ray (CXR) after thoracic surgery contributes to patient discomfort and costs and is of limited therapeutic value. Lung ultrasound (LU) for pneumothorax may be an alternative to CXR, but diagnostic accuracy data are heterogeneous and biased by insufficient sonographic technique and patient selection. Reported sensitivities range from 0.21 to 1.0. We evaluated the sensitivity of LU on the first day after thoracic surgery under routine conditions. Methods We performed a prospective observational study (trial-ID DRKS00014557). Consecutive patients undergoing lung resection received standardized LU in addition to routine CXR on the first postoperative day. Ultrasound examiner and radiologist were blinded to corresponding X‑ray and ultrasound findings. CXR was used as reference to determine diagnostic test performance of ultrasound. The conformity of sonography- and routine-based therapeutic decisions was evaluated. Results A total of 68 patients were examined. The mean duration of ultrasound was 145 ± 64 s. CXR identified 23 patients with pneumothorax with a mean apex-to-cupola size of 1.5 ± 1.0 cm. Ultrasound detected 18 patients with pneumothorax. The computed sensitivity of LU was 0.48 (95% confidence interval [0.36; 0.60]). Specificity was between 0.81 and 1.0, the negative predictive value 0.76 [0.66; 0.86]. The sensitivity of CXR was 0.56 [0.44; 0.68]. Air leakage via chest tube correlated weakly with CXR (spearman’s rho = 0.26) and moderately with LU (rho = 0.43). The conformity between sonographically based recommendations and the actual therapy based on routine diagnostics was 96%. Conclusions Sensitivity of ultrasound for pneumothorax detection nearly reached CXR and resulted in equally safe patient management. Our data can serve as a pilot study for upcoming larger-scaled controlled trials.