Diagnostic point-of-care ultrasound (POCUS) has rapidly evolved into an indispensable bedside tool across critical care and perioperative medicine. Patients undergoing cardiothoracic (CT) surgery present with complex, multi-organ vulnerabilities ranging from neurologic injury and respiratory failure to hemodynamic instability, renal dysfunction, and thromboembolic disease. This review outlines how diagnostic POCUS, as described in the recent JoVE Methods Collection, can be applied in a structured "head-to-toe" framework to facilitate the care of CT surgery patients. At the cranial level, modalities such as optic nerve sheath diameter and transcranial Doppler offer non-invasive insights into intracranial pressure and cerebrovascular flow. Airway and lung ultrasound support rapid diagnosis of intubation complications, pneumothorax, effusion, and parenchymal edema, while cardiac ultrasound provides a stepwise approach to evaluating hypotension, ventricular dysfunction, and pericardial effusion. Diaphragmatic assessment further informs the management of phrenic nerve injury and postoperative respiratory compromise. Abdominal ultrasound, including renal evaluation and venous excess ultrasound (VExUS), aids in identifying obstructive causes of acute kidney injury and in assessing the systemic impact of right-sided congestion. Finally, lower-extremity vascular and musculoskeletal POCUS permits efficient screening for deep vein thrombosis and provides a window into frailty in the chronically ill. By synthesizing these diagnostic workflows, this review highlights how structured, multimodal POCUS can accelerate diagnosis, individualize therapy, and reduce reliance on resource-intensive testing in a vulnerable surgical population. Broader adoption and training may help standardize POCUS integration into cardiothoracic perioperative care and ultimately improve patient outcomes.

Pleural diseases pose a significant burden on healthcare systems due to diagnostic challenges and high costs. Artificial intelligence (AI) has the potential to provide faster, more accurate, and more reliable results in the diagnosis of these diseases. This review evaluates the current status of AI technologies in the diagnosis of pleural effusion (PE), malignant PE, tuberculosis pleurisy (TP), pneumothorax, and malignant pleural mesothelioma (MPM). Deep learning algorithms developed for radiological diagnosis provide high sensitivity and specificity in determining the presence and severity of PE. AI models that integrate clinical parameters such as chest computed tomography (CT), positron emission tomography (PET)-CT, and tumour markers in distinguishing between benign and malignant effusions have significantly improved diagnostic accuracy (area under the curve: >0.90). In cytological diagnosis, computer-assisted systems such as Aitrox have demonstrated performance comparable to that of expert cytopathologists in diagnosing malignant effusions. In the diagnosis of TP, AI models outperform conventional diagnostic methods, particularly when combined with laboratory parameters such as adenosine deaminase. Food and Drug Administration-approved AI models are effectively used for the rapid diagnosis of pneumothorax and for emergency interventions. In MPM diagnosis, AI models using PET-CT images and three-dimensional segmentation offer significant advantages in prognostic evaluation and treatment response monitoring. However, large-scale, multi-centre studies are needed to standardise and generalise AI models. In light of these developments, AI may fundamentally change the diagnostic management of pleural diseases.

Pneumothorax-induced electrocardiography (ECG) changes are common. However, the usage of ECG in diagnosing of pneumothorax remains limited. In addition, these changes could challenge the diagnosis of pneumothorax, particularly in patients with minimal symptoms. The objective of this study is to evaluate the ECG changes in patients with primary spontaneous pneumothorax (PSP).A retrospective observational study was conducted at King Abdullah University Hospital among patients diagnosed with PSP between January 2018 and December 2023. Patients who had PSP with ECG performed before chest tube insertion were included in the study. Data on patient demographics, ECG changes, and pneumothorax volume were collected.Eighteen patients with PSP met the inclusion criteria for this study. The most frequent ECG findings in all patients were incomplete right bundle branch block (RBBB) and S < 1.2mV in V2; both were observed in 17 out of the 18 ECGs (94%). For patients with left-sided PSP, the most frequent ECG changes were S < 1.2mV in V2 and S < 0.9mV in V3; both were observed in 6 out of 7 ECGs (85%). Among those with right-sided PSP, the most frequent findings were incomplete RBBB and S < 1.2mV in V2, occurring in 10 out of 11 ECGs (90.9%). Incomplete RBBB and S < 1.2mV in V2 were present in all cases of PSP with a size > 50%.The relationship between pneumothorax and ECG changes induced by pneumothorax is well established. However, these changes are not statistically significant. Despite this, they could play a role in supporting the diagnosis of pneumothorax in patients with clinical symptoms and signs of pneumothorax.

Pneumothorax is a potentially life-threatening complication in critically ill patients. Lung ultrasound (LUS) is highly accurate for bedside diagnosis, with the lung point sign being specific for pneumothorax. A variant, the heart point sign, has been described in isolated case reports as a marker of left-sided pneumothorax, but its clinical relevance remains incompletely understood. We retrospectively reviewed intensive care unit cases at Sanatorio Los Arcos (Buenos Aires, Argentina) from January 2018 to July 2024. Patients were included if the heart point sign was documented during cardiac ultrasound performed as part of the hemodynamic assessment following the diagnosis of pneumothorax by lung ultrasound. The sign was evaluated in B-mode and, when available, M-mode, and defined as the visualization of the heart during diastole that disappeared during systole, coinciding with a lung pattern showing absent lung sliding and A-lines. Clinical data, confirmatory imaging, and management strategies were obtained from medical records. Ten patients (median age 56 years, IQR 45-65; 50% male) exhibited the heart point sign. Most patients (80%) were on invasive mechanical ventilation, and two required vasopressor support. The sign was visualized in parasternal (6/10) and apical (4/10) views. Pneumothorax was confirmed by CT (5/10), chest radiography (3/10), or ultrasound alone (2/10). Etiologies included central line placement (5/10), surgical complications (2/10), failure of lung re-expansion (1/10), spontaneous pneumothorax (1/10), and lung abscess rupture (1/10). Management consisted of chest tube insertion (6/10), video-assisted thoracoscopic surgery (3/10), and conservative observation (1/10). The heart point sign is an infrequent but highly specific sonographic marker of left-sided pneumothorax. It can be assessed in B-mode and M-mode, reflecting the dynamic interaction between cardiac motion and interposed pleural air. Recognition of this sign provides rapid bedside confirmation, complementing classical ultrasonographic findings and enhancing diagnostic confidence in critically ill patients.

Evaluation of the effectiveness of the ChatGPT artificial intelligence application in the diagnosis of spontaneous pneumothorax on chest radiograph interpretation

BackgroundBlunt chest trauma represents a significant cause of preventable mortality globally, accounting for approximately one-third of trauma-related deaths in India. Timely diagnosis of hemothorax and pneumothorax is essential to prevent clinical deterioration. While clinical examination and chest X-ray (CXR) are commonly employed, their diagnostic sensitivity is limited. The extended focused assessment with sonography in trauma (eFAST) offers a rapid, bedside, radiation-free alternative with demonstrated diagnostic utility.ObjectiveThe objective of this study was to evaluate and compare the diagnostic accuracy of eFAST, clinical examination, and CXR for the detection of hemothorax and pneumothorax, with CT chest serving as the reference standard.MethodsThis prospective diagnostic observational study was conducted at the Department of General Surgery, Pondicherry Institute of Medical Sciences, India, between January 2021 and June 2022. Eighty-one hemodynamically stable patients with blunt chest trauma underwent clinical examination, CXR, eFAST, and CT chest evaluation. Diagnostic performance metrics, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy, were calculated with CT chest as the gold standard.ResultsThe mean age was 49.2 ± 15.6 years, with 71.6% of participants being male. Road traffic accidents were the predominant mechanism (67.9%). CT chest revealed thoracic injuries in 64.2% of cases. Clinical examination showed a sensitivity of 34.6% and specificity of 100%. Chest X-ray had a sensitivity of 76.9%, a specificity of 85.2%, and an accuracy of 81.5%. eFAST achieved a sensitivity of 98.1%, specificity of 98.8%, PPV of 98%, NPV of 96.7%, and an overall accuracy of 98.5%.ConclusioneFAST demonstrated superior diagnostic accuracy compared to clinical examination and CXR, and closely approximated the diagnostic reliability of CT chest. Given its portability, absence of radiation, and rapid application, eFAST should be integrated as a frontline tool in the evaluation of blunt chest trauma.

Tension-type pneumothorax in particular is a life-threatening emergency that requires prompt diagnosis and treatment. Although imaging modalities—a chest CT scan being the gold standard—are frequently employed for confirmation, clinical diagnosis is essential in cases when imaging delays could be lethal. A bronchopleural fistula (BPF) complicated spontaneous pneumothorax in an obese patient with a body mass index of 43.6 kg/m². Before the chest tube was inserted, the diagnosis was verified by a straightforward and repeatable bedside diagnostic technique: loss of syringe plunger recoil following pleural entrance with a 16G trocar cannula. The patient experienced a persistent air leak that was consistent with BPF after tube thoracostomy, and this was meticulously watched. After three weeks of conservative treatment, the air leak eventually closed on its own, negating the need for surgery. This case demonstrates how the syringe plunger recoil test can be used as a simple and trustworthy bedside tool to support imaging in the quick diagnosis of pneumothorax, enabling prompt treatment to start. It also implies that a cautious conservative approach may allow for the spontaneous closure of a bronchopleural fistula with careful monitoring and the right safeguards, saving some patients from the hazards associated with major surgery.

Abstract Introduction/Objective Spontaneous pneumothorax is a relatively rare presentation in otherwise healthy, pregnant or postpartum women, and it can mask other conditions. LAM is a relatively rare neoplastic disease which predominantly involves pulmonary and lymphatic systems in this group. It is often underdiagnosed until more advanced presentations, such as pneumothorax or chylous effusions. Concomitantly, dedifferentiated endometrial carcinoma is an uncommon yet aggressive gynecologic cancer, and it typically manifests with nonspecific symptoms. We describe a patient with spontaneous bilateral pneumothoraces with presenting feature of both pulmonary and nodal LAM in a female also with previously undiagnosed dedifferentiated endometrial adenocarcinoma. This case demonstrates the importance of pathology in the accurate discrimination between overlapping disease entities, accentuating the difficulty in diagnosing combined rare neoplasms. The aim of this article is to point out this diagnostic dilemma and to stress the necessity of keeping LAM in mind in female patients presenting with nonspecific pulmonary symptoms, especially in the presence of coexisting neoplasm that influences on the overall clinical presentation. This case underscores the need for histopathologic and immunohistochemical markers and the importance of multi-disciplinary approach in the management of complicated presentations. Methods/Case Report A 46-year-old nulligravid woman had acute dyspnea and a hemoglobin level of 2.6 g/dL. Chest X-ray showed bilateral pneumothoraces with subsequent emergent chest tubes. Pelvic imaging demonstrated thickened endometrium and bilateral adnexal masses, and retroperitoneal lympadenopathy. She had cis-video-assisted thoracoscopic surgery (VATS) on both the sides with wedge resection and pleurodesis. Histopathological sections of lung revealed nodular and cystic growth of spindle cells invading pulmonary parenchyma that were immunoreactive for HMB-45 and smooth muscle actin and consistent with pulmonary LAM. There were no features of tuberous sclerosis complex (TSC) favouring sporadic LAM. Other gynecologic surgery performed was hysterectomy, bilateral salpingo-oophorectomy and omentectomy and lymph node dissection. Pathology showed dedifferentiated endometrial adenocarcinoma with rhabdoid features with predominant involvement of ovaries and the omentum. Significantly, one external iliac lymph node had features of nodal LAM—spindle cell morphology and HMB-45 positivity—but no evidence for metastatic carcinoma. This uncommon dual pathology led to diagnostic dilemma. Given the histomorphologic and immunophenotypic overlap, a comprehensive immunohistochemical work up was a requisite to prevent misdiagnosis of LAM as metastatic disease. Definitive diagnosis was made by the cooperation with pathology, thoracic surgeon, gynecologic oncologist, and a pulmonologist departmentally. This case illustrates the diagnostic challenges of both rare diseases Results NA Conclusion This case illustrates an unusual association of pulmonary and nodal lymphangioleiomyomatosis and dedifferentiated endometrial carcinoma, which first presented with spontaneous bilateral pneumothoraces. The finding of LAM in the lung and lymph nodes in the setting of an aggressive gynecologic malignancy emphasizes the need for meticulous morphologic and immunohistochemical evaluation. HMB-45 staining differentiated nodal LAM from metastatic carcinoma. In the absence of thorough pathologic examination, nodal LAM can be mistaken for lymphatic metastasis from endometrial carcinoma, resulting in misdiagnosis and unnecessary treatment. This case highlights the importance of clinical suspicion and interdisciplinary communication when confronted with diagnostically difficult presentations. Our patient’s complicated clinical course illustrates the importance of considering rare diseases, such as LAM, as differential diagnoses of spontaneous pneumothorax, especially in women of childbearing age. Furthermore, concurrent occurrence of two hormonally responsive neoplasms in the breast triggers interesting queries relative to common molecular or environmental etiological factors. In the end, this case is a timely reminder of the central role of pathology in negotiating diagnostic uncertainty and in directing appropriate clinical care. Continued recognition and reporting of rare co-morbidites will contribute towards accruing the necessary knowledge base in order to refine diagnosis and patient management in such multifactorial presentations.

Background: A significant portion of all traumas occur in the chest. Almost 10% of trauma patients end up in the hospital because of this. Different kinds and levels of chest trauma cause different kinds of injuries, which in turn cause different kinds of consequences, and those outcomes show up in the corresponding death rate. Aim and objectives: Assessment of the use of chest computed tomography (CT) and lung ultrasound (US) in the diagnosis of pneumothorax in patients with a history of trauma, as well as comparisons between the two methods for the purpose of early detection and treatment of this condition. Subjects and methods: From December 2023 through December 2024, thirty trauma patients presenting with respiratory distress were enrolled in this comparative study. The patients were admitted to the emergency and critical care departments of Al-Azhar University Hospitals in Cairo, Egypt. Results: Twenty-four (80%) patients were diagnosed with traumatic pneumothorax by lung ultrasound. 26(86.67%) patients diagnosed with traumatic pneumothorax by chest CT. Chest ultrasound can detect traumatic pneumothorax with (Kappa=0.524) 88.5% sensitivity, 75% specificity, 95.8% PPV, and 50% NPV. Conclusion: In trauma patients with polytrauma or direct chest trauma, chest ultrasound is a quick, easy, readily available, and useful diagnostic technique for pneumothorax (Kappa=0.524) with an 88.5% sensitivity and 75% specificity.

Simple SummaryPneumothorax is a life-threatening condition that requires rapid and accurate diagnosis, especially in emergency and critical care settings. Although lung ultrasound (LUS) offers a fast and radiation-free diagnostic option, its accuracy can vary depending on the operator’s experience. This study evaluated the potential of artificial intelligence (AI) to assist clinicians by automatically detecting pneumothorax on LUS images and videos. Using transformer-based deep learning models, we compared the diagnostic performance of Vision Transformer (ViT), DINOv2, and Video Vision Transformer (ViViT) architectures. When tested on data from different patients, the DINOv2 model achieved 90% accuracy, demonstrating reliable generalization without overfitting. Furthermore, when video sequences were analyzed, both Random Forest and eXtreme Gradient Boosting classifiers trained on ViViT-derived features achieved 90% accuracy, showing that AI can effectively interpret dynamic pleural motion. These results indicate that transformer-based AI can enhance pneumothorax diagnosis by improving consistency and reducing operator dependence, supporting broader use of lung ultrasound in emergency and point-of-care environments.Background: Pneumothorax (PTX) requires rapid recognition in emergency and critical care. Lung ultrasound (LUS) offers a fast, radiation-free alternative to computed tomography (CT), but its accuracy is limited by operator dependence. Artificial intelligence (AI) may standardize interpretation and improve performance. Methods: This retrospective single-center study included 46 patients (23 with CT-confirmed PTX and 23 controls). Sixty B-mode and M-mode frames per patient were extracted using a Clarius C3 HD3 wireless device, yielding 2760 images. CT served as the diagnostic reference. Experimental studies were conducted within the framework of three scenarios. Transformer-based models, Vision Transformer (ViT) and DINOv2, were trained and tested under two scenarios: random frame split and patient-level split. Also, Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) classifiers were trained on the feature maps extracted by using Video Vision Transformer (ViViT) for ultrasound video sequences in Scenario 3. Model performance was evaluated using accuracy, sensitivity, specificity, F1-score, and area under the ROC curve (AUC). Results: Both transformers achieved high diagnostic accuracy, with B-mode images outperforming M-mode inputs in the first two scenarios. In Scenario 1, ViT reached 99.1% accuracy, while DINOv2 achieved 97.3%. In Scenario 2, which avoided data leakage, DINOv2 performed best in the B-mode region (90% accuracy, 80% sensitivity, 100% specificity, F1-score 88.9%). ROC analysis confirmed strong discriminative ability, with AUC values of 0.973 for DINOv2 and 0.964 for ViT on B-mode images. Also, both RF and XGBoost classifiers trained on the ViViT feature maps reached 90% accuracy on the video sequences. Conclusions: AI-assisted LUS substantially improves PTX detection, with transformers—particularly DINOv2—achieving near-expert accuracy. Larger multicenter datasets are required for validation and clinical integration.

Emergency Department Accuracy of Point-of-Care Ultrasound in Identifying Clinically Significant Pneumothorax in High-Severity Trauma Patients.

Introduction: Pneumothorax is a life-threatening condition resulting from the entry of air into the pleural cavity. The most popular imaging test used in its diagnosis is classical X-ray, and in doubtful cases, also computed tomography (CT). However, ultrasound (US) is becoming increasingly popular. Thanks to the development of diagnostic criteria, it is possible to quickly, safely, and extremely effectively detect pneumothorax using lung ultrasound. Despite its many advantages, this method has certain limitations, including conditions and diseases whose images may falsely suggest pneumothorax. Methods: The following paper is based on a compilation of our own experiences from approximately 350 lung ultrasound examinations and a literature review to identify as many clinical conditions as possible that may falsely suggest an ultrasound image of pneumothorax, along with potential causes of error, taking into account the sonomorphology and pathophysiology of the lesions. Additionally, criteria have been developed to differentiate specific conditions from pneumothorax, and a classification has been prepared based on the difficulty of differentiation, with particular emphasis on situations requiring conclusive additional testing. Conclusions: Ultrasound diagnosis of pneumothorax must be closely correlated with the clinical presentation to avoid errors. Adherence to diagnostic criteria will prevent errors in most cases. Less experienced examiners should pay particular attention to subtle differences in pleural sliding.

BackgroundThis study evaluates the diagnostic performance of three multimodal large language models (LLMs)—ChatGPT-4o, Gemini 2.0, and Claude 3.5—in identifying pneumothorax from chest radiographs.MethodsIn this retrospective analysis, 172 pneumothorax cases (148 patients aged >12 years, 24 patients aged ≤12 years) with both chest radiographs and confirmatory thoracic CT were included from a tertiary emergency department. Patients were categorized by age and pneumothorax size (small/large). Each radiograph was presented to all three LLMs accompanied by basic symptoms (dyspnea or chest pain), with each model analyzing each image three times. Diagnostic accuracy was evaluated using overall accuracy (all three responses correct), strict accuracy (≥2 responses correct), and ideal accuracy (≥1 response correct), alongside response consistency assessment using Fleiss’ Kappa.ResultsIn patients older than 12 years, ChatGPT-4o demonstrated the highest overall accuracy (69.6%), followed by Claude 3.5 (64.9%) and Gemini 2.0 (57.4%). Performance was significantly poorer in pediatric patients across all models (20.8%, 12.5%, and 20.8%, respectively). For large pneumothorax in adults, ChatGPT-4o showed significantly higher accuracy compared to small pneumothorax (81.6% vs. 42.2%; p < 0.001). Regarding consistency, Gemini 2.0 demonstrated excellent reliability for large pneumothorax (Kappa = 1.00), while Claude 3.5 showed moderate consistency across both pneumothorax sizes.ConclusionThis study, the first to evaluate these three current multimodal LLMs in pneumothorax identification across different age groups, demonstrates promising results for potential clinical applications, particularly for adult patients with large pneumothorax. However, performance limitations in pediatric cases and with small pneumothoraces highlight the need for further validation before clinical implementation.

Tension pneumothorax is the third leading cause of potentially survivable death in the prehospital, combat setting. Identification of the presence of a pneumothorax before tension physiology develops remains challenging in this setting. We conducted an early developmental pilot study to determine if unprocessed raw radio frequency (RF) data from a single-crystal ultrasound array could fill this gap. We prospectively enrolled sus scrofa models as part of a medical education training program with intentional induction of pneumothorax. We obtained thoracic imaging using Clarius (clinical) and Verisonics (research) devices, only the latter of which could provide RF data from the entire probe array. We assembled RF time histories into a feature vector and used principal components analysis to extract features with the greatest variance. We examined linear discriminant analysis (LDA) and logistic regression as classifiers. Six sus scofa were included in the final analysis. The Clarius system yielded single image-based RF-traces per acquisition, which did not prove useful for further analysis. From the Verisonics system, we obtained 49 acquisitions pre-pneumothorax and 41 acquisitions pneumothorax, each of which contained 20 image frames and raw RF data for all scanlines. A vast majority of the RF signal variance was contained in the first PC, although all but the last PC contained at least >0.3% of the total variance. Only PC0 mean is statistically significant between pre- and post- groups (P = .0472). A bivariate logistic model using PC0 and PC8 (P = .184) correctly predicted 5 of 6 animals in each condition (83.3%), with 1 animal misclassified from each condition. The LDA analysis yielded only 1 linear discriminator feature, which showed a difference in the means between groups (P = .0161). This single LD used as input to a univariate logistic model yielded equal prediction accuracy to the previous classifier (83%, 1 misclassified per group), with animal 3 pre and animal 1 post misclassified by this reduced feature, and animal 2 post being nearly misclassified. In this pilot study, we were able to determine a potential signal for the diagnosis of pneumothorax using RF data. Our findings will aid in the development of low-power devices to detect pneumothorax.

Background: Catamenial pneumothorax, a rare manifestation of thoracic endometriosis syndrome (TES), presents a significant diagnostic and therapeutic challenge. It is characterized by recurrent spontaneous pneumothorax occurring in temporal relation to menstruation in women of reproductive age. The underlying pathophysiology is complex, often involving the ectopic presence of endometrial tissue within the thoracic cavity. Coexisting pelvic pathologies, such as adenomyosis, may be associated, further complicating the clinical picture. Case presentation: We present the case of a 38-year-old woman with a four-month history of recurrent, right-sided pneumothorax, with symptoms consistently commencing 24 to 48 hours prior to the onset of her menstrual cycle. Initial investigations, including high-resolution computed tomography of the thorax performed between menstrual cycles and microbiological analysis for tuberculosis, were unremarkable. The diagnosis of catamenial pneumothorax was established based on the distinct cyclical pattern of her symptoms. A subsequent gynecological evaluation, prompted by a history of secondary dysmenorrhea and menorrhagia, revealed uterine adenomyosis via transvaginal ultrasonography. The patient was managed through a collaborative, multidisciplinary approach involving pulmonology, thoracic surgery, and gynecology. Treatment consisted of chemical pleurodesis with doxycycline, administered via a chest tube, followed by continuous hormonal suppression therapy with oral progestin (2 mg/day). Conclusion: This case highlights the critical importance of maintaining a high index of suspicion for catamenial pneumothorax in women of reproductive age presenting with recurrent pneumothorax. A successful outcome was achieved through a coordinated, multidisciplinary strategy combining definitive pleural symphysis via pleurodesis with systemic hormonal therapy to suppress the underlying endometriotic process. This dual approach effectively prevented pneumothorax recurrence over a 12-month follow-up period, underscoring its efficacy in managing this complex condition.

AimPneumothorax is a potentially life-threatening condition whose diagnosis can be challenging. Ultrasound chest examination is generally fast and user-friendly, but in non-expert hands or with uncooperative patients, it may still be difficult and time-consuming. Adding another tool to support the suspicion of pneumothorax might be useful, potentially enhancing the diagnostic accuracy of standard ultrasound chest examination. We evaluated the feasibility of standardized bilateral ultrasound image comparison as a potential new tool for pneumothorax diagnosis.Materials and methodsWe enrolled 60 subjects (30 with pneumothorax and 30 controls) and collected bilateral ultrasound images of their chests (each image contained one frame from the left lung and one from the right lung). Ten physicians (eight blinded to diagnosis) divided into five groups according to expertise evaluated the images for potential grayscale differences and/or horizontal artifacts between the two frames. All images were then analyzed with image analysis software for grayscale pixel assessment (one sub-analysis for the entire area under the pleural line, one for a 100-pixel-wide rectangle under the pleural line).ResultsAll clinicians achieved good results in terms of diagnostic accuracy and inter-operator reliability, even those unexperienced in ultrasound. Mean, range, and median grayscale pixel ratio between the pneumothorax side and the healthy side in a single patient proved to be the most reliable parameters, reaching excellent sensitivity and specificity. Combining these parameters proved to be an excellent diagnostic tool (ROC area under curve = 1.00, p-value = 0.02).ConclusionsStandardized bilateral thoracic ultrasound image comparison may be a potential new tool for the diagnosis of pneumothorax.

The efficacy of bedside chest ultrasonography for the detection and diagnosis of pneumothorax is under debate. We aimed to compare Emergency Healthcare Workers performed chest ultrasonography with chest X-ray in the detection and diagnosis of pneumothorax in the emergency department. We queried PubMed, Cochrane, ScienceDirect, Web of Science and ClinicalTrials.gov databases from 2000 through January 2024. We included all studies (both retrospective and prospective) that compared the diagnostic performance of chest ultrasonography with chest radiography, using chest computed tomography as the gold standard. Participants are patients consulting in the emergency department and physician that performed the chest ultrasound was an Emergency Healthcare Workers. Studies reporting the sensitivity and specificity for both chest ultrasonography and chest X-ray met inclusion criteria. We applied a random effects meta-analysis methodology. We then performed a meta-regression analysis to search for influencing variables such as technical parameters of echograph, patients and pneumothorax. 15 studies totaling 3,171 patients were analyzed. 71% of patients were male with a mean age of 40.2years. The mean prevalence of pneumothorax was 27.6% (95 CI 20.9 to 34.3). Chest ultrasonography had higher sensitivity (79.4%, 68.2 to 90.7) compared to chest X-ray (48.1%, 36.8 to 59.4), and a greater negative predictive value (chest ultrasonography = 94.3%, 91.2 to 97.3, and chest X-ray = 87.9%, 84.1 to 91.6). There was no statistical difference in specificity between the two modalities: chest ultrasonography 99.5%, 99 to 100 and chest X-ray 99.8%, 99.4 to 100) or in positive predictive value (chest ultrasonography 94.2%, 90.5 to 97.9 vs chest X-ray 96.7%,92 to 100). Characteristics of echograph or pneumothorax and patients sociodemographic did not influence results. In this systematic review and meta-analysis, chest ultrasonography performed by Emergency Healthcare Workers, had greater sensitivity and negative predictive value than chest radiography for the diagnosis of pneumothorax in emergency department patients.

Image processing is a solution in the development of chest X-ray technology, starting from the image segmentation process as a preprocessing stage to separate the image object from the original background. Spontaneous pneumothorax (SP) is a type of air collection in the pleural cavity that develops without trauma. The diagnosis of pneumothorax has a sensitivity of approximately 25 to 75% using an anteroposterior chest x-ray, which still provides a dubious picture of pneumothorax. However, the development of the Region Feature algorithm with a new algorithm, namely RM Multy, has improved the accuracy. The RM Multy algorithm can calculate the area of the object, allowing it to produce the area of infiltration in the right lung, left lung, and the lung as a whole. The Region Feature results of the Pneumothorax obtained with the detected image area as many as 19 areas, for the pixel size of each area are 145, 355, 110, 31, 31, 52, 30, 36, 54, 122, 58, 23, 476, 77, 192, 24, 168, 263, 41 and 44. So the total pixels for 19 areas is 2301. The area converted to mm2 is 2301 x 0.04 mm2 = 92.04 mm2. Classification results on lungs with Pneumothorax and Normal by detection process with RM Multy using the CNN algorithm with an accuracy of 96.43%. This accuracy confirms the success of the system, which has been processed using a new algorithm. Therefore, further development is needed to improve detection accuracy in pneumothorax cases with smaller area sizes.

Aim: This study aimed to retrospectively evaluate the patients who were followed up in the neonatal intensive care unit of our hospital with the diagnosis of pneumothorax. Methods: The records of patients who were followed up with the diagnosis of pneumothorax in our neonatal intensive care unit between September 1, 2016 and December 31, 2022 were retrospectively reviewed. Birth weight, sex, gestational week, mode of delivery, localization of pneumothorax, presence of underlying primary lung disease, and mortality were evaluated. Results: The mean birth weight of 35 patients (19 girls, 16 boys) who developed pneumothorax was 2200±1050 g and the mean gestational age was 33.2±5.1 weeks. Twenty-seven of the patients were delivered by cesarean section and 8 by normal spontaneous vaginal delivery. Pneumothorax was most common on the right side (n:19) and no patient had bilateral pneumothorax. 13 patients had received surfactant treatment before pneumothorax. The primary diagnoses were respiratory distress syndrome (RDS) in 17 patients and transient tachypnea of the newborn (TTN) in 11 patients. 14 patients were resuscitated at birth. A thoracic tube was inserted in 22 patients, while 13 patients were followed up conservatively. Conclusions: The most common predisposing causes in patients with pneumothorax are RDS and TTN. Early diagnosis and treatment of pneumothorax is life-saving. It should be kept in mind that pneumothorax may develop in patients who are followed up in the neonatal intensive care unit due to respiratory distress.

Pneumothorax is a potentially life-threatening condition defined as the collapse of the lung due to air leakage into the chest cavity. Delays in the diagnosis of pneumothorax can lead to severe complications and even mortality. A significant challenge in pneumothorax diagnosis is the shortage of radiologists, resulting in the absence of written reports in plain X-rays and, consequently, impacting patient care. In this paper, we propose an automatic triage system for pneumothorax detection in X-ray images based on deep learning. We address this problem from the perspective of multi-source domain adaptation where different datasets available on the Internet are used for training and testing. In particular, we use datasets which contain chest X-ray images corresponding to different conditions (including pneumothorax). A convolutional neural network (CNN) with an EfficientNet architecture is trained and optimized to identify radiographic signs of pneumothorax using those public datasets. We present the results using cross-dataset validation, demonstrating the robustness and generalization capabilities of our multi-source solution across different datasets. The experimental results demonstrate the model’s potential to assist clinicians in prioritizing and correctly detecting urgent cases of pneumothorax using different integrated deployment strategies.