Combining machine learning and texture analysis to differentiate mediastinal lymph nodes in lung cancer patients.

Evaluate whether texture analysis associated with machine learning approaches could differentiate between malignant and benign lymph nodes. A total 18 patients with lung cancer were selected, with 39 lymph nodes, being 15 malignant and 24 benign. Retrospective computed tomography scans were utilized both with and without contrast medium. The great differential of this work was the use of 15 textures from mediastinal lymph nodes, with five different physicians as operators. First and second order statistical textures such as gray level run length and co-occurrence matrix were extracted and applied to three different machine learning classifiers. The best machine learning classifier demonstrated a variability of less than 5% among operators. The support vector machine (SVM) classifier presented 95% of the area under the ROC curve (AUC) and 89% of sensitivity for sequences without contrast medium. SVM classifier presented 93% of AUC and 86% of sensitivity for sequences with contrast medium. Texture analysis and machine learning may be helpful in the differentiation between malign and benign lymph nodes. This study can aid the physician in diagnosis and staging of lymph nodes and potentially reduce the number of invasive analysis to histopathological confirmation.

Persistent pleuropulmonary air leak treated with autologous blood: results from a university hospital and review of literature.

BACKGROUND: Persistent air leak after pulmonary resection is a difficult complication for thoracic surgeons to manage. OBJECTIVES: To show the results of our experience treating persistent pleuropulmonary air leak with autologous blood and review the literature on this specific method of treatment. METHODS: Retrospective study of patients with persistent aerial pleuropulmonary fistula treated with autologous blood. The patient's own blood was collected from a peripheral vein and directly introduced through the pleural drain. An inverted siphon was located in the drainage system to avoid prolonged clamping of the drain. This siphon impeded blood return but not air escape. RESULTS: Between January 2001 and August 2008, 27 patients were treated by the above method. Patient age ranged from 2 to 74 years, and 78% were male. Each procedure used a mean quantity of 92 ml blood. Mean persistent air leak time before pleurodesis was 10.6 days and mean time to fistula resolution after pleurodesis was 1.5 days. Twenty-three (85%) patients had persistent pleuropulmonary air leak closed with the above procedure. CONCLUSION: Treating persistent pleuropulmonary air leak with autologous blood is promising, but further studies are required to quantify its real effectiveness.

Study of bronchoalveolar lavage fluid in paracoccidioidomycosis: cytopathology and alveolar macrophage function in response to gamma interferon; comparison with blood monocytes.

Patients with paracoccidioidomycosis (PCM) present marked involvement of the lungs during the course of the mycosis. The purpose of this work was to obtain bronchoalveolar lavage (BAL) fluid from these patients to study the cytopathology, TNF levels and the oxidative and fungicidal response of alveolar macrophages (AMs) to in vitro incubation with recombinant IFN-gamma. To compare the lung and blood compartments, these determinations were also made in plasma and blood monocytes (BMs) obtained from the same patients. The cytopathology of BAL fluid revealed a predominance of macrophages, but with the presence of neutrophil exudation, and rare lymphocytes and epithelioid and giant cells. Comparison of the oxidative status and fungicidal activity of AMs and circulating BMs demonstrated that both cell types are highly activated for these two functions when compared to control cells. However, TNF levels were higher in BAL fluid than in plasma. The possible mechanisms involved in the hyperresponsiveness of cells from PCM patients are discussed.