Malignant Mediastinal Nodes Research Articles

Endobronchial Ultrasound-Based Support Vector Machine Model for Differentiating between Benign and Malignant Mediastinal and Hilar Lymph Nodes

Introduction: The aim of the study was to establish an ultrasonographic radiomics machine learning model based on endobronchial ultrasound (EBUS) to assist in diagnosing benign and malignant mediastinal and hilar lymph nodes (LNs). Methods: The clinical and ultrasonographic image data of 197 patients were retrospectively analyzed. The radiomics features extracted by EBUS-based radiomics were analyzed by the least absolute shrinkage and selection operator. Then, we used a support vector machine (SVM) algorithm to establish an EBUS-based radiomics model. A total of 205 lesions were randomly divided into training (n = 143) and validation (n = 62) groups. The diagnostic efficiency was evaluated by receiver operating characteristic (ROC) curve analysis. Results: A total of 13 stable radiomics features with non-zero coefficients were selected. The SVM model exhibited promising performance in both groups. In the training group, the SVM model achieved an ROC area under the curve (AUC) of 0.892 (95% CI: 0.885–0.899), with an accuracy of 85.3%, sensitivity of 93.2%, and specificity of 79.8%. In the validation group, the SVM model had an ROC AUC of 0.906 (95% CI: 0.890–0.923), an accuracy of 74.2%, a sensitivity of 70.3%, and a specificity of 74.1%. Conclusion: The EBUS-based radiomics model can be used to differentiate mediastinal and hilar benign and malignant LNs. The SVM model demonstrated excellent potential as a diagnostic tool in clinical practice.

Quantitative evaluation of endobronchial ultrasound elastography in the diagnosis of benign and malignant mediastinal and hilar lymph nodes

ObjectiveTo compare the diagnostic value of different quantitative methods of endobronchial ultrasound elastography in benign and malignant mediastinal and hilar lymph nodes. MethodsThis retrospective study included patients who underwent endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for mediastinal and hilar lymph node enlargement in our hospital between January 2019 and August 2022. We compared different quantitative elastography parameters [red area ratio (RAR, lymph node red area/lymph node area), green area ratio (GAR, lymph node green area/lymph node area), blue area ratio (SAR, lymph node blue area/lymph node area), mixed area ratio (MAR, lymph node green area/lymph node area), blue-green lymph node area/lymph node area), strain rate ratio (SR), strain rate in the target lymph node (LPA), ratio of blue area to total lymph node area outside the center of the target lymph node (PAR), and average grey value (MGV)], in order to find the best quantitative evaluation method. ResultsA total of 244 patients (346 lymph nodes) were included in this study. All quantitative elastography parameters were statistically significant for the differentiation of benign and malignant lesions except the average grey value of the target lymph nodes. The area under the receiver operating characteristic curve of SAR was 0.872 (95% confidence interval: 0.83−0.91), the cutoff value was 0.409, and the sensitivity, specificity, positive and negative predictive values were 85.4%, 78.0%, 80.4%, and 83.4%, respectively. ConclusionCompared with other types of quantitative analysis, SAR has a higher predictive significance for benign and malignant lymph nodes.

Dual-energy computed tomography for evaluating nodal staging in lung adenocarcinoma: correlation with surgical pathology.

To ascertain the performance of dual-energy CT (DECT) with iodine quantification in differentiating malignant mediastinal and hilar lymph nodes (LNs) from benign ones, focusing on patients with lung adenocarcinoma. In this study, patients with suspected lung cancer received a preoperative contrast-enhanced DECT scan from Jun 2018 to Dec 2020. Quantitative DECT parameters and the size were compared between metastatic and benign LNs. Their diagnostic performances were analyzed by the ROC curves and compared by using the two-sample t test. 72 patients (23 men, 49 women; mean age 62.5 ± 10.1 years) fulfilled the inclusion criteria. A total of 98 LNs (67 benign, 31 metastatic) were analyzed. The iodine concentration normalized by muscle (NICmuscle) was significantly higher (P < 0.001) in metastatic LNs (4.79 ± 1.70) than in benign ones (3.00 ± 1.45). The optimal threshold of NICmuscle was 3.44, which yielded AUC: 0.798, sensitivity: 83.9%, specificity: 73.1%, accuracy: 76.5%, respectively. Applying the established size parameters with 10mm as the threshold yielded AUC: 0.600, sensitivity: 29.0%, specificity: 91.0%, accuracy: 71.4%, respectively. The diagnostic performance of NICmuscle was significantly better (P = 0.007) than the performance obtained using the established size parameters. For lung adenocarcinoma, the quantitative measurement of NICmuscle derived from DECT is useful for differentiating benign and metastatic mediastinal and hilar LNs before surgical intervention.

Quantitative analysis of endobronchial elastography combined with serum tumour markers of lung cancer in the diagnosis of benign and malignant mediastinal and hilar lymph nodes

Purpose: In malignant tumours, elastography and serum tumour markers have shown high diagnostic efficacy. Therefore, we aimed to quantitatively analyse the results of endobronchial elastography combined with serum tumour markers of lung cancer to accurately distinguish benign and malignant mediastinal and hilar lymph nodes.Methods: Data of patients who underwent endobronchial ultrasound-guided transbronchial needle aspiration for mediastinal lymph node enlargement in our hospital between January 2018 and August 2022 were retrospectively collected. The characteristics of quantitative elastography and serum tumour markers were evaluated.Results: We enrolled 197 patients (273 lymph nodes). In the differential diagnosis of benign and malignant mediastinal and hilar lymph nodes, the stiffness area ratio (SAR), strain ratio (SR), and strain rate in lymph nodes were significant, among which SAR had the highest diagnostic value (cut-off value, 0.409). The combination of the four tumour markers had a high diagnostic value (AUC, 0.886). Three types of quantitative elastography indices combined with serum tumour markers for lung cancer showed a higher diagnostic value (AUC, 0.930; sensitivity, 83.5%; specificity, 89.3%; positive predictive value, 88.1%; negative predictive value, 85%) (p &amp;lt; 0.05). In the differential diagnosis of pathological types of lung cancer, different quantitative elastography indicators and serum tumour markers for lung cancer have different diagnostic significance for the differential diagnosis of lung cancer pathological types.Conclusion: The quantitative analysis of endobronchial ultrasound elastography combined with tumour markers can improve the diagnosis rate of benign and malignant mediastinal and hilar lymph nodes, help guide the puncture of false negative lymph nodes, and reduce the misdiagnosis rate.

Diagnostic value of endobronchial ultrasound elastography in differentiating between benign and malignant hilar and mediastinal lymph nodes: a retrospective study.

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a first-line approach for diagnosing hilar and mediastinal lymph node metastasis. Endobronchial ultrasound (EBUS) elastography is an imaging technique for describing the elasticity of intrathoracic lesions. However, the reported accuracy of EBUS elastography needs to be improved. In this study, we aimed to explore the diagnostic value of EBUS elastography for differentiating between benign and malignant hilar and mediastinal lymph nodes. We conducted a single-center, retrospective study enrolling consecutive patients who received EBUS elastography followed by EBUS-TBNA at the Chinese PLA General Hospital from October 2015 to October 2022. The pathological results of EBUS-TBNA confirmed by 6-month follow-up were used as the gold standard. The ultrasound elastography parameters of lymph nodes included strain rate, stiff area ratio, and elasticity score, along with the conventional ultrasound characteristics such as short axis diameter, shape, margin, echogenicity distribution and intensity, and blood flow. The diagnostic performance of these parameters was compared, and conjointly analyzed using multivariate logistic regression. Bootstrapping resampling was applied for internal validation of the regression model. A total of 83 patients were enrolled with an average age of 57 years, and 66.3% of patients were male. In total, 131 lymph nodes were punctured, among which 79 (60.3%) were malignant. All the conventional ultrasound characteristics were significantly different between benign and malignant lymph nodes. All the ultrasound elastography parameters of malignant lymph nodes were markedly higher than those of benign lymph nodes. Multivariate logistic regression analysis showed that the margin, echogenicity intensity, blood flow, short axis diameter, and stiff area ratio were the main factors affecting the lymph node property. The diagnostic accuracy, sensitivity, and specificity were 91.8% [95% confidence interval (CI): 85.4-96.0%], 94.4% (95% CI: 86.4-98.5%), and 88.0% (95% CI: 75.7-95.5%), respectively. Bootstrap resampling validation showed a concordance index (C-index) of 0.949. The calibration plot indicated good agreement between the predicted and observed results. EBUS elastography is a promising approach for differentiating between benign and malignant lymph nodes. The combination of conventional EBUS and elastography can improve diagnostic efficacy, provide reliable complementary information, and guide the implementation of EBUS-TBNA more accurately.

C11 LARGE PLEURAL AND PERICARDIAL EFFUSION: WHICH DRAIN FIRST?

Abstract Introduction Large pleural and pericardial effusion is a common finding in lung cancer patients. However, a chronic large pleural effusion and the “lung entrapment” fenomenon, due to the neoplastic lung infiltration, could precipitate the re–expansion pulmonary edema (REPO) after the thoracentesis procedure. REPO is a rare and potentially life–threatening complication after large volume thoracentesis. It is characterized by alveolar infiltration in the reexpanded lung.Indeed, in the presence of “lung entrapment”, REPO development could be due not only to and excessive fluid removal, but also to the development of an excessively negative intrapleural pressure (&amp;lt; – 20 mmHg). As the REPO is often associated to an hemodynamic impairment, in patients with concomitant severe pericardial effusion, this condition could cause cardiac tamponade and the optimal therapeutic approach could be challenging.In this perspective, the decision on the effusion to be drained first, in this population, is not always straightforward.Case report In a 62 years –old woman with severe respiratory failure, a chest CT scan showed a massive left –sided pleural effusion with a complete collapse of the left lung, a severe pericardial effusion (35 mm) and the presence of malignant mediastinal nodes. (Figure 1).In the ICU a left thoracentesis was performed, with an early drainage of 1400 ml, the drain was then clamped. 30 minutes later, the patient developed a severe acute respiratory failure. An urgent chest x ray revealed an ipsilateral pulmonary oedema (Figure2). The patient was then treated with respiratory support with cPAP and intravenous diuretic therapy. Afterward, a rapid hemodynamic derangement occurred, with cardiac tamponade. An emergency echoguided pericardiocentesis was then performed followed by an immediate recovery. A repeated thorax CT scan revealed an advanced pulmonary tumor (Figure 3). Discussion In this case all the risk factors for REPO were coexistent: chronic pleural effusion and lung cancer. In addition, the large pericardial effusion represented a complicating factor in this situation as both the cPAP and the REPO’s hypovolemia may have brought to a clinical tamponade. In this context, where a neoplastic severe pleural and pericardial effusions are coexistent, it is essential to know which possible complications may occur after a large volume thoracentesis in order to prevent a high risk urgent pericardial drainage in patient with severe respiratory distress.

Prediction of malignancy in mediastinal lymph nodes during endobronchial ultrasound: A comparative validation study.

In this study, we aimed to compare the diagnostic performances of three existing prediction tools in visually identifying a malignant lymph node. Between April 2016 and January 2021, a total of 827 lymph nodes of 259 patients (211 males, 48 females; mean age: 61.1±7.2 years; range, 41 to 79 years) who underwent endobronchial ultrasound procedure for diagnosis and/or staging of lung cancer and diagnosis of mediastinal lymphadenopathy of unknown origin were retrospectively analyzed. This external validation study was designed to compare the diagnostic yields of the prediction tools developed by Shafiek et al., Alici et al., and Canada Lymph Node Score (CLNS). Endobronchial ultrasoundguided transbronchial needle aspiration results and predictions were compared to gold-standard tool. Overall, endobronchial ultrasound-guided transbronchial needle aspiration had a sensitivity, specificity, positive and negative predictive value, and accuracy of 95.6%, 100%, 100%, 97.6%, and 98.4%, respectively. Diagnostic performances of proposed tools were quite remarkable. Among them, Alici algorithm had a higher sensitivity and negative predictive value, which were matched by excellent specificity and positive predictive value offered by CLNS ≥3 and Shafiek tool. The area under the curve value of CLNS ≥3 was higher than Shafiek tool and CLNS ≥2. Conventional prediction tools relying on simple real-time sonographic features were found to be consistent by the means of diagnostic performance in this external validation dataset. Despite being inferior to cytology, their superior performance was proven with defined individual strengths and weaknesses.

CT texture analysis and node-RADS CT score of mediastinal lymph nodes - diagnostic performance in lung cancer patients

BackgroundTexture analysis derived from computed tomography (CT) can provide clinically relevant imaging biomarkers. Node-RADS is a recently proposed classification to categorize lymph nodes in radiological images. The present study sought to investigate the diagnostic abilities of CT texture analysis and Node-RADS to discriminate benign from malignant mediastinal lymph nodes in patients with lung cancer.MethodsNinety-one patients (n = 32 females, 35%) with a mean age of 64.8 ± 10.8 years were included in this retrospective study. Texture analysis was performed using the free available Mazda software. All lymph nodes were scored accordingly to the Node-RADS classification. All primary tumors and all investigated mediastinal lymph nodes were histopathologically confirmed during clinical workup.ResultsIn discrimination analysis, Node-RADS score showed statistically significant differences between N0 and N1-3 (p < 0.001). Multiple texture features were different between benign and malignant lymph nodes: S(1,0)AngScMom, S(1,0)SumEntrp, S(1,0)Entropy, S(0,1)SumAverg. Correlation analysis revealed positive associations between the texture features with Node-RADS score: S(4,0)Entropy (r = 0.72, p < 0.001), S(3,0) Entropy (r = 0.72, p < 0.001), S(2,2)Entropy (r = 0.72, p < 0.001).ConclusionsSeveral texture features and Node-RADS derived from CT were associated with the malignancy of mediastinal lymph nodes and might therefore be helpful for discrimination purposes. Both of the two quantitative assessments could be translated and used in clinical routine.

Endobronchial ultrasound-guided transbronchial needle aspiration in patients with previously treated malignancies: diagnostic performance and predictive value.

Endobronchial ultrasound-guided transbronchial needle aspiration is a minimally invasive and effective sampling approach for patients with mediastinal or hilar lymphadenopathy. Increased recognition of the ultrasonographic features revealed the value of its images in predicting mediastinal lymph node malignancy. However, its diagnostic validity and the predictive value of its ultrasonographic features have not been demonstrated well in patients after systemic anti-tumor therapy. This study aimed to evaluate the efficiency of endobronchial ultrasound-guided transbronchial needle aspiration in patients with suspicious lymph nodes after anti-tumor therapy. We retrospectively reviewed cases of endobronchial ultrasound-guided transbronchial needle aspiration performed between January 2019 and August 2021 at a single tertiary hospital center. Patients with suspected mediastinal or hilar lymph nodes within 5years of systemic anti-tumor therapy were enrolled. Final diagnoses were determined by pathologic diagnoses of samples from transbronchial needle aspiration, surgery, or follow-up for at least 6months. Ultrasonographic features were analyzed to assess the predictive value of malignant lymph nodes after treatment. Overall, 168 lymph nodes of 138 patients were analyzed. Among 110 (65.5%) malignant lymph nodes, 75 originated from lung cancers; the other 35 were from other malignancies. No complications related to endobronchial ultrasound-guided transbronchial needle aspiration were observed. Of 58 negative results of transbronchial needle aspiration, 51 were proven to be true negatives; 7 were false. The overall sensitivity and the negative predictive value were 94.02% and 87.93%, respectively. Univariate and multivariate analysis revealed the absence of central hilar structure and short axis > 10mm as independent predictive factors for malignancy. Endobronchial ultrasound-guided transbronchial needle aspiration performs satisfactorily in diagnosing mediastinal and hilar lymphadenopathy even after anti-tumor treatment.

Contribution of F-18 fluorodeoxyglucose PET/CT and contrast-enhanced thoracic CT texture analyses to the differentiation of benign and malignant mediastinal lymph nodes.

Texture analysis and machine learning methods are useful in distinguishing between benign and malignant tissues. To discriminate benign from malignant or metastatic mediastinal lymph nodes using F-18 fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) and contrast-enhanced computed tomography (CT) texture analyses with machine learning and determine lung cancer subtypes based on the analysis of lymph nodes. Suitable texture features were entered into the algorithms. Features that statistically significantly differed between the lymph nodes with small cell lung cancer (SCLC), adenocarcinoma (ADC), and squamous cell carcinoma (SCC) were determined. The most successful algorithms were decision tree with the sensitivity, specificity, and area under the curve (AUC) values of 89%, 50%, and 0.692, respectively, and naive Bayes (NB) with the sensitivity, specificity, and AUC values of 50%, 81%, and 0.756, respectively, for PET/CT, and NB with the sensitivity, specificity, and AUC values of 10%, 96%, and 0.515, respectively, and logistic regression with the sensitivity, specificity, and AUC values of 21%, 83%, and 0.631, respectively, for CT. In total, 13 features were able to differentiate SCLC and ADC, two features SCLC and SCC, and 33 features ADC and SCC lymph node metastases in PET/CT. One feature differed between SCLC and ADC metastases in CT. Texture analysis is beneficial to discriminate between benign and malignant lymph nodes and differentiate lung cancer subtypes based on the analysis of lymph nodes.

High SUVmax Is an Independent Predictor of Higher Diagnostic Accuracy of ROSE in EBUS-TBNA for Patients with NSCLC.

Introduction: This study aimed to verify the predictors of the diagnostic accuracy of rapid on-site evaluation (ROSE) in endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) among patients with non-small cell lung cancer (NSCLC). Methods: We retrospectively reviewed consecutive patients with NSCLC who underwent EBUS-TBNA for staging or diagnosis at our hospital from June 2016 to June 2018. The patients were divided into two groups—those with a correct diagnosis and an incorrect diagnosis after ROSE. Kaplan–Meier plots and log-rank tests were used to estimate outcomes. Results: A total of 84 patients underwent EBUS-TBNA for staging and diagnosis. Sixty patients with demonstrated malignant mediastinal lymph nodes were enrolled. In the univariate analysis, lymph nodes < 1.5 cm (HR = 3.667, p = 0.031) and a SUVmax > 5 (HR = 41, p = 0.001) were statistically significant for diagnostic accuracy of ROSE. In the multivariate Cox regression analysis, only a SUVmax > 5 (HR = 20.258, p = 0.016) was statistically significant. Conclusions: A SUVmax > 5 is an independent predictor of higher diagnostic accuracy of ROSE in EBUS-TBNA in patients with NSCLC with malignant mediastinal lymph nodes. Therefore, ROSE in patients with a SUVmax < 5 might not be reliable and requires further prudent assessment (more shots or repeated biopsies at mediastinal LNs) in clinical practice.

Diagnostic value of endobronchial ultrasound elastography for differentiating benign and malignant hilar and mediastinal lymph nodes: a systematic review and meta-analysis.

In the present study, a meta-analysis was performed to evaluate the diagnostic value of endobronchial ultrasound (EBUS) elastography for differentiating benign and malignant hilar and mediastinal lymph nodes (LNs). A comprehensive literature search was carried out through PubMed, Embase, and Cochrane Library. Two authors screened the papers and extracted the data independently and any discrepancies were resolved by discussion. The methodolog-ical quality of each included study was assessed by the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and area under the curve were calculated to evaluate the value of EBUS elastography for hilar and mediastinal LNs. Seventeen studies with the number of 2307 LNs were included. There was significant heterogeneity across the included studies. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio for the diagnosis of hilar and mediastinal LNs by EBUS elastography were 0.90 (95% confidence interval [CI], 0.84-0.94), 0.78 (95% CI, 0.74-0.81), 4.1 (95% CI, 3.4-4.9), 0.12 (95% CI, 0.07-0.21) and 33 (95% CI, 17-64), respectively. Furthermore, area under the curve was calculated to be 0.86 (95% CI, 0.82-0.88). EBUS elastography is a valuable technology in the differentiation of benign and malignant hilar and mediastinal LNs and could provide supplementary diagnostic information during endobronchial ultrasound-guided transbronchial needle aspiration. The combination of EBUS elastography and B-mode EBUS could improve the diagnostic accuracy for hilar and mediastinal LNs.

Diffusion Tensor Imaging in Characterization of Mediastinal Lymphadenopathy

To determine the efficacy of diffusion-weighted MRI (DWI) and diffusion tensor imaging (DTI) in the characterization of mediastinal lymphadenopathy and the differentiation between malignant and benign lymph nodes (LNs). a retrospective evaluation of 58 patients with mediastinal lymphadenopathy that underwent DWI and DTI with calculation of apparent diffusion coefficient (ADC), fractional anisotropy (FA), and mean diffusivity (MD) values of LNs. Final diagnosis was made by the histopathology and proved metastatic (n=21), lymphomatous (n=14), granulomatous (n=11) and reactive (n=12) LNs. Malignant mediastinal LNs had remarkably lower ADC and MD; (p=0.001) and higher FA; (p=0.001) than in benign LNs. The threshold of ADC, MD, and FA at (1.48, 1.32×10-3 mm2/s), (1.31, 1.33×10-3 mm2/s), (0.62, 0.52) to differentiate malignant from benign LNs has AUC of (0.89, 0.94), (0.96, 0.95), (0.72, 0.82), accuracy of (87%, 86%), (89%, 86%), (70%, 72%) by both observers respectively. The threshold of ADC, MD, and FA at (1.47, 1.32×10-3 mm2/s), (1.31, 1.3×10-3 mm2/s), (0.62, 0.67) used to differentiate metastatic from reactive LNs revealed AUC of (0.90, 0.94), (0.96, 0.96), (0.73, 0.77), accuracy of (87%, 81%), (87%, 81%), (72%, 66%) by both observers respectively. The mean ADC and MD values of metastatic LNs were statistically significant (p=0.001) and (p=0.002, 0.02) respectively when compared with that of lymphoma. The threshold of ADC, and MD (0.94, 0.97×10-3 mm2/s) and (0.87, 0.91×10-3 mm2/s) used to differentiates metastatic from lymphomatous nodes revealed AUC of (0.90, 0.91), (0.81, 0.74), an accuracy of (85%, 91%), (71%, 71%), by both observers respectively. The inter-class correlation between two observers for all nodes for ADC, MD and FA was r= 0.931, 0.956 and 0.885 respectively. Using ADC, MD, and FA can help in the characterization of mediastinal lymphadenopathy noninvasively.

Conventional Transbronchial Needle Aspiration (cTBNA) and EBUS-Guided Transbronchial Needle Aspiration (EBUS-TBNA): A Retrospective Study on the Comparison of the Two Methods for Diagnostic Adequacy in Molecular Analysis

Introduction: In recent years, there has been a growing development of molecularly targeted therapies for various types of solid tumors—in particular, in non-small-cell lung cancer (NSCLC). This has required the need for greater quantities of tissue that is able to support ancillary studies, alongside cyto-histological diagnoses for the assessment of molecular targets. Conventional TBNA (cTBNA) and EBUS-guided TBNA (EBUS-TBNA) have shown a high diagnostic yield for malignant mediastinal and/or hilar lymph node enlargement and peribronchial masses; however, few studies have compared these two procedures. We retrospectively compared TBNA patients (EBUS-TBNA and cTBNA) in order to determine the diagnostic yield and material adequacy for subsequent ancillary analyses. Materials and Methods: We retrospectively evaluated 318 patients with clinical suspicion of lung cancer or with disease recurrence. All of the patients underwent TBNA (either EBUS-TBNA or cTBNA) on enlarged mediastinal and/or hilar lymph nodes and peribronchial masses between January 2017 and June 2021 at the University Hospital of Pisa, Italy. After a definitive diagnosis, molecular analyses and an evaluation of PD-L1 expression were performed in the cases of adenocarcinoma, squamous cell carcinoma, and NSCLC, not otherwise specified (NOS). Results: EBUS-TBNA was performed in 199 patients and cTBNA was performed in 119 patients with 374 and 142 lymph nodes, respectively. The overall diagnostic yield for positive diagnoses was 59% (diagnostic rate of 61% in EBUS-TBNA, and 55% in cTBNA). Adenocarcinoma (ADC) was the most frequent diagnosis in both methods. EBUS-TBNA diagnostic adequacy was 72% for molecular analysis, while it was 55.5% for cTBNA, showing a statistical trend (p = 0.08) towards the significance of EBUS. The average percentage of neoplastic cells was also statistically different between the two methods (p = 0.05), reaching 51.19 ± 22.14 in EBUS-TBNA and 45.25 ± 22.84 in cTBNA. With regard to the PD-L1 protein expression, the percentage of positivity was similar in both procedures (86% in EBUS-TBNA, 85% in cTBNA). Conclusions: Conventional TBNA (cTBNA) and EBUS-guided TBNA (EBUS-TBNA) are minimally invasive diagnostic methods that are associated with a high diagnostic yield. However, EBUS-TBNA has an improved diagnostic adequacy for molecular analysis compared to cTBNA, and is associated with a higher average percentage of neoplastic cells.

From Raw Pixels to Recurrence Image for Deep Learning of Benign and Malignant Mediastinal Lymph Nodes on Computed Tomography

Lung cancer causes the most cancer deaths worldwide and has one of the lowest five-year survival rates of all cancer types. It is reported that more than half of patients with lung cancer die within one year of being diagnosed. Because mediastinal lymph node status is the most important factor for the treatment and prognosis of lung cancer, the aim of this study is to improve the predictive value in assessing the computed tomography (CT) of mediastinal lymph-node malignancy in patients with primary lung cancer. This paper introduces a new method for creating pseudo-labeled images of CT regions of mediastinal lymph nodes by using the concept of recurrence analysis in nonlinear dynamics for the transfer learning. Pseudo-labeled images of original CT images are used as input into deep-learning models. Three popular pretrained convolutional neural networks (AlexNet, SqueezeNet, and DenseNet-201) were used for the implementation of the proposed concept for the classification of benign and malignant mediastinal lymph nodes using a public CT database. In comparison with the use of the original CT data, the results show the high performance of the transformed images for the task of classification. Three pretrained convolutional neural networks that are AlexNet, SqueezeNet, and DenseNet201 were trained and tested with the transformed images. Classification accuracies and areas under the receiver operating characteristic curve obtained from the ten-fold cross-validation are 93% and 0.97, 96% and 0.99, and 100% and 1 for the SqueezeNet, AlexNet, and DenseNet201, respectively. The proposed method has the potential for differentiating benign from malignant mediastinal lymph nodes on CT, and may provide a new way for studying lung cancer using radiology imaging.

Mediastinal staging for non-small cell lung cancer

The staging of mediastinal lymph nodes for lung cancer is crucial for planning treatments or reinterventions. In potentially curable patients the aim of mediastinal staging is to exclude the presence of malignancy in mediastinal lymph nodes with a high level of accuracy while also considering clinical factors and the balance of the benefits and risks of tissue sampling techniques. Mediastinal staging is based on computed tomography (CT) and positron emission tomography (PET) and can be sufficient when no mediastinal abnormalities are present and the probability of unforeseen N2 disease is low. In the case of bulky lymph nodes with a high probability of malignancy in PET-CT, tissue confirmation is not normally required. If mediastinal sampling is needed it can be achieved by endosonographic techniques, including endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) and endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) or a combination of the two. Positive results do not need further confirmation. In the case of negative results, surgical techniques still play a role in the selected cases discussed by multidisciplinary lung cancer committees. New mediastinal surgical techniques including video-assisted cervical mediastinoscopy (VACM), video-assisted mediastinoscopic lymphadenectomy (VAMLA), and transcervical extended mediastinal lymphadenectomy (TEMLA) have been shown to be useful in selected patients. Final pathological staging is based on lymph node removal during surgery and can be achieved by taking one of two approaches: lymph node sampling or systematic lymph node sampling. The accuracy of PET-CT and mediastinal endosonography is lower for mediastinal restaging than it is for surgical techniques; their false positive and false negative (FN) rate is high and so, they require histological confirmation. Here we explain and revise the results from the most recent studies and current international guidelines.

Can artificial intelligence distinguish between malignant and benign mediastinal lymph nodes using sonographic features on EBUS images?

Aims This study aimed to develop a new intelligent diagnostic approach using an artificial neural network (ANN). Moreover, we investigated whether the learning-method-guided quantitative analysis approach adequately described mediastinal lymphadenopathies on endobronchial ultrasound (EBUS) images. Methods In total, 345 lymph nodes (LNs) from 345 EBUS images were used as source input datasets for the application group. The group consisted of 300 and 45 textural patterns as input and output variables, respectively. The input and output datasets were processed using MATLAB. All these datasets were utilized for the training and testing of the ANN. Results The best diagnostic accuracy was 82% of that obtained from the textural patterns of the LNs pattern (89% sensitivity, 72% specificity, and 78.2% area under the curve). The negative predictive values were 81% compared to the corresponding positive predictive values of 83%. Due to the application group’s pattern-based evaluation, the LN pattern was statistically significant (p = .002). Conclusions The proposed intelligent approach could be useful in making diagnoses. Further development is required to improve the diagnostic accuracy of the visual interpretation.

High Diagnostic Accuracy and Safety of Endoscopic Ultrasound-Guided Fine-Needle Aspiration in Malignant Lymph Nodes: A Systematic Review and Meta-Analysis

Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) is increasingly being used for diagnosing lymphadenopathy. We aim to systematically review the accuracy of EUS-FNA in differentiating benign and malignant mediastinal and abdominal lymph nodes (LNs). A comprehensive literature search was performed on multiple electronic databases through February 2020. A random or fixed effect model generated the pooled sensitivity, specificity, likelihood ratio (LR), and diagnostic odds ratio (DOR) of EUS-FNA. Subgroup analyses and meta-regression were used to explore sources of heterogeneity. Twenty-six studies involving 2753 patients with 2833 LNs were included. In the differential diagnosis of benign and malignant LNs, EUS-FNA had a pooled sensitivity, specificity, positive LR, and negative LR of 87% (95% confidence interval [CI] 86-90%), 100% (95% CI 99-100%), 68.98 (95% CI 42.10-113.02), and 0.14 (95% CI 0.11-0.17), respectively. The pooled rate of adverse events associated with EUS-FNA was 1.57% (95% CI 1.06-2.24%). The summary receiver operating characteristic (SROC) yielded an area under the curve (AUC) of 0.9912. EUS-FNA performed in mediastinal LNs gained a sensitivity of 85% (95% CI 81-88%), while in abdominal LNs, it reached 87% (95% CI 82-91%). The sensitivity of the subgroup with rapid on-site evaluation (ROSE) was 91% (95% CI 89-93%), while non-ROSE was 85% (95% CI 82-87%). EUS-FNA is a sensitive, highly specific, and safe method for distinguishing benign and malignant mediastinal or abdominal LNs. However, the sensitivity of EUS-FNA still varies significantly among different centers.

Performance improvement of mediastinal lymph node severity detection using GAN and Inception network

Background and objectiveIn lung cancer, the determination of mediastinal lymph node (MLN) status as benign or malignant influence treatment planning and survival rate. Invasive pathological tests for the classification of MLNs into benign and malignant have various shortcomings like painfulness, the risk associated with anesthesia, and depends to a large extent on skillset and preferences of the surgeon performing the test. Hence, computer-aided system for MLNs severity detection has been explored widely by the researchers. Very recently, in our earlier concluded work on non-invasive method for MLNs differential diagnosis in computed tomography (CT) images, combination of different data augmentation approaches and state-of-art fully convolutional network (FCN) were implemented to enhance the performance of malignancy detection. However, the performance of FCN network were highly depended on the selection of appropriate data augmentation approach and control of their hyperparameters. Moreover, a standard practice to get hierarchical features in convolutional neural network (CNN) models requires deeper stacking of layers. This leads to an increase in number of trainable parameters which prone to overfitting of the network. MethodsIn view of the above mention limitations, in this paper, authors have proposed an approach that includes: 1) Generative Adversarial Network (GAN) for data augmentation, and 2) Inception network for malignancy detection. Unlike conventional data augmentation strategy, GAN based augmentation approach generates data that correlates to original data distribution. In the case of Inception based model, it uses multiple size kernels with factorized convolution for hierarchical feature extraction. This helps to a significant reduction in trainable parameters and the problem of overfitting. ResultsIn this paper, experiments with different GAN approaches, as well as with different Inception architectures, are conducted to evaluate and justify the selection of appropriate GAN and Inception architecture, respectively for MLNs severity detection. The proposed approach achieves superior results with an average accuracy, sensitivity, specificity, and area under curve of 94.95%, 93.65%, 96.67%, and 95%, respectively. ConclusionThe obtained results validate the usefulness of GANs for data augmentation in the differential diagnosis of benign and malignant MLNs. The proposed Inception network based classifier for malignancy detection shows promising results compared to all investigated methods presented in various literature.

A novel online education module to teach clinicians how to correctly identify ultrasonographic features of mediastinal lymph nodes during endobronchial ultrasound

Ultrasonographic features can be used to predict mediastinal lymph node malignancy during endobronchial ultrasonography. Despite the validity of using these features for this purpose, the features are not being widely used in clinical practice. This may be attributable to the absence of educational programs that teach clinicians how to identify the features. To address this knowledge gap, we developed an online educational module to teach clinicians how to correctly interpret ultrasonographic features. The module was designed using corrective feedback and test-enhanced learning theories and distributed to clinicians in relevant specialties. The efficacy of the program was determined by comparing the percentages of correctly identified ultrasonographic features as each clinician progressed through the module. Participants were also asked to self-rate their confidence during the module. Analysis of variance was conducted, and a learning curve and descriptive statistics were generated. Twenty-two of the 29 participants (76%) completed the module. Analysis of variance indicated that the percentage of correctly identified features increased significantly as clinicians completed the module (p = 0.004); this finding is supported by the positive slope of the learning curve. Even though they initially reported some difficulty with identifying certain features, their confidence increased as they progressed through the module. When asked, 86% of participants reported that they found the educational module helpful and 90% reported that they would recommend it to others. Participating clinicians were receptive to the interactive educational module. It enhances clinician skill and confidence in interpreting ultrasonographic features. The results of this study provide the foundation needed to test the validity of the educational module in clinical settings and to further explore clinician preferences for educational programs.