Combining machine and deep transfer learning for mediastinal lymph node evaluation in patients with lung cancer

Background The prognosis and survival of patients with lung cancer are likely to deteriorate with metastasis.Using deep-learning in the detection of lymph node metastasis can facilitate the noninvasive calculation of the likelihood of such metastasis,thereby providing clinicians with crucial information to enhance diagnostic precision and ultimately improve patient survival and prognosis.Methods In total,623 eligible patients were recruited from two medical institutions.Seven deep learning models,namely Alex,GoogLeNet,Resnet18,Resnet101,Vgg16,Vgg19,and MobileNetv3(small),were utilized to extract deep image histological features.The dimensionality of the extracted features was then reduced using the Spearman correlation coefficient(r≥0.9)and Least Absolute Shrinkage and Selection Operator.Eleven machine learning methods,namely Support Vector Machine,K-nearest neighbor,Random Forest,Extra Trees,XGBoost,LightGBM,Naive Bayes,AdaBoost,Gradient Boosting Decision Tree,Linear Regression,and Multilayer Perceptron,were employed to construct classification prediction models for the filtered final features.The diagnostic performances of the models were assessed using various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value.Calibration and decision-curve analyses were also performed.Results The present study demonstrated that using deep radiomic features extracted from Vgg16,in conjunction with a prediction model constructed via a linear regression algorithm,effectively distinguished the status of mediastinal lymph nodes in patients with lung cancer.The performance of the model was evaluated based on various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value,which yielded values of 0.808,0.834,0.851,0.745,0.829,and 0.776,respectively.The validation set of the model was assessed using clinical decision curves,calibration curves,and confusion matrices,which collectively demonstrated the model's stability and accuracy.Conclusion In this study,information on the deep radiomics of Vgg16 was obtained from computed tomography images,and the linear regression method was able to accurately diagnose mediastinal lymph node metastases in patients with lung cancer.

Exploration and optimization of surgical techniques for laparoscopic transhiatal lower mediastinal lymph node dissection for adenocarcinoma of esophagogastric junction: A prospective IDEAL 2a study with qualitative design

Objective: To explore the change and feasibility of surgical techniques of laparoscopic transhiatal(TH)-lower mediastinal lymph node dissection(LMLND) for adenocarcinoma of the esophagogastric junction(AEG)according to Idea, Development, Exploration, Assessment, and Long-term follow-up(IDEAL) 2a standards.Methods: Patients diagnosed with AEG who underwent laparoscopic TH-LMLND were prospectively included from April 14, 2020, to March 26, 2021. Clinical and pathological information as well as surgical outcomes were quantitatively analyzed. Semistructured interviews with the surgeon after each operation were qualitatively analyzed.Results: Thirty-five patients were included. There were no cases of transition to open surgery, but three cases involved combination with transthoracic surgery. In qualitative analysis, 108 items under three main themes were detected: explosion, dissection, and reconstruction. Revised instruction was subsequently designed according to the change in surgical technique and the cognitive process behind it. Three patients had anastomotic leaks postoperatively, with one classified as Clavien-Dindo Ⅲa.Conclusions: The surgical technique of laparoscopic TH-LMLND is stable and feasible;further IDEAL 2b research is warranted.

A CLINICOPATHOLOGICAL STUDY OF MEDIASTINAL LYMPH NODE METASTASIS OF LUNG CANCER

Objective: To investigate pathologically the characteristics of lung cancer metastasis by mediastinal lymph node ways (N2). Methods: Of 398 lung cancer patients who underwent radical pulmonectomy and extensive lymph node dissection, 160 patients were diagnosed as with N2 metastasis, their 352 groups of mediastinal lymph nodes invaded were subject to the pathological study. Results: Evidences showed that the N2 metastasis of lung cancer was very active. It appears as single group or multi-group or jumping-form metastasis, rating 41.2%, 58.8% and 29.3% respectively. In addition, the extension of N2 metastasis was large, the most concentrated site was the 7th group lymph node (48.8%), then the 4th, 3rd and 5th group, rating 45.6%, 31.3% and 25.6% respectively. The occurrence of N2 metastasis was highly correlated with the site, size, histopathological type and the grade of differentiation of the cancer. An another feature of N2 metastasis was the invasion of metastasized lymph node into the bronchial wall, especially in adenocarcinoma. Conclusion: In order to achieve the radical removal of tumor, it is necessary to dissect the lymph nodes of the hilar and upper and lower mediastinum at the homolateral thoracic cavity actively and completely; beside, attention may be paid to the bronchial wall invasion caused by the lymph nodes metastasized.

Perioperative considerations in patients with non small cell lung cancer and metastases in mediastinal lymph nodes

Since the latest revision of the TNM system reclassified T3N0 tumours into the ⅡB stage, N2 lesions became the major determinant of the ⅢA stage. Concerning the minority of patients with T3N1 tumours in this stage,

Video-assisted Thoracoscopic Surgery for the Treatment of Mediastinal Lymph Node Tuberculous Abscesses

The mediastinal lymph node tuberculous abscesses（MLNTAs） are secondary to mediastinal tuberculous lymphadenitis. Surgical excision is often required when cold abscesses form. This study was aimed to examine video-assisted thoracoscopic surgery（VATS） for the treatment of MLNTA. Clinical data of 16 MLNTA patients who were treated in our hospital between December 1, 2013 and December 1, 2015 were retrospectively analyzed. All of the patients underwent the radical debridement and drainage of abscesses, and intrathoracic lesions were removed by VATS. They were also administered the intensified anti-tuberculosis treatment（ATT）, and engaged in normal physical activity and follow-up for 3 to 6 months. The results showed that VATS was successfully attempted in all of the 16 MLNTA patients and they all had good recovery. Two patients developed complications after surgery, with one patient developing recurrent laryngeal nerve injury, and the other reporting poor wound healing. It was concluded that VATS is easy to perform, and safe, and has high rates of success and relatively few side-effects when used to treat MLNTA.

Endobronchial Ultrasound Elastography for Diagnosing Mediastinal and Hilar Lymph Nodes

Background： Pathophysiological processes, such as malignancy, can lead to the formation of stiffer tissue in lung cancers. Endobronchial ultrasound rEBUS） elastography is a novel technique for measuring tissue stiffness during EBUS-guided transbronchial needle aspiration （EBUS-TBNA）. The current study was conducted to investigate the diagnostic value of EBUS elastography tbr mediastinal and hilar lymph node metastasis in lung cancers. Methods： From January 2014 to January 2015, 40 patients suspected of lung cancer were enrolled, and a total of 68 lymph nodes were evaluated by EBUS-TBNA. EBUS-guided elastography of lymph nodes was perfornaed prior to EBUS-TBNA. Standard EBUS characteristics were also described. Pathological determination of malignant or benign lymph nodes was used as the gold standard for this study. If EBUS-TBNA did not result in a tbrmal pathological diagnosis of malignancy, patients were referred for a surgical procedure. Comparisons of elastography and standard EBUS characteristics were made between benign and malignant lymph nodes. Results： Elastography grading scores and strain ratios showed significant differences between benign and malignant lymph nodes （P = 0.000）. The elastography strain ratio was more sensitive and specific for determining malignant lymph nodes than elastography grading score or standard EBUS criteria. The receiver operating characteristic curve for the elastography strain ratio showed an area under the curve of 0.933. The best cut-off point of the strain ratio for differentiating malignant from benign lymph nodes was 32.07. The elastography strain ratio had a sensitivity of 88.1%, the specificity of 80.8%, positive predictive value of 88.1%, and negative predictive value of 80.8% for distinguishing malignant from benign nodes. The overall accuracy of elastography strain ratio was 85.3%. The strain ratio of malignant and benign lymph nodes positively correlated with the elastography grading score （r = 0.561, P = 0.000）. Conclusions： EBUS elastography can be effectively used to predict mediastinal and hilar lymph node metastases in lung cancer. This noninvasive technique may thus complement standard EBUS and help guide EBUS-TBNA procedures.