Transformation of PET raw data into images for event classification using convolutional neural networks.

In positron emission tomography (PET) studies, convolutional neural networks (CNNs) may be applied directly to the reconstructed distribution of radioactive tracers injected into the patient's body, as a pattern recognition tool. Nonetheless, unprocessed PET coincidence data exist in tabular format. This paper develops the transformation of tabular data into n-dimensional matrices, as a preparation stage for classification based on CNNs. This method explicitly introduces a nonlinear transformation at the feature engineering stage and then uses principal component analysis to create the images. We apply the proposed methodology to the classification of simulated PET coincidence events originating from NEMA IEC and anthropomorphic XCAT phantom. Comparative studies of neural network architectures, including multilayer perceptron and convolutional networks, were conducted. The developed method increased the initial number of features from 6 to 209 and gave the best precision results (79.8) for all tested neural network architectures; it also showed the smallest decrease when changing the test data to another phantom.

Low-Vacuum Filtration as an Alternative Extracellular Vesicle Concentration Method: A Comparison with Ultracentrifugation and Differential Centrifugation.

Recent years have brought great focus on the development of drug delivery systems based on extracellular vesicles (EVs). Considering the possible applications of EVs as drug carriers, the isolation process is a crucial step. To solve the problems involved in EV isolation, we developed and validated a new EV isolation method-low-vacuum filtration (LVF)-and compared it with two commonly applied procedures-differential centrifugation (DC) and ultracentrifugation (UC). EVs isolated from endothelial cell culture media were characterized by (a) Transmission Electron Microscopy (TEM), (b) Nanoparticle Tracking Analysis (NTA), (c) Western blot and (d) Attenuated Total Reflection Fourier-Transform Infrared Spectroscopy (ATR-FTIR). Additionally, the membrane surface was imaged with Environmental Scanning Electron Microscopy (ESEM). We found that LVF was a reproducible and efficient method for EV isolation from conditioned media. Additionally, we observed a correlation between ATR-FTIR spectra quality and EV and protein concentration. ESEM imaging confirmed that the actual pore diameter was close to the values calculated theoretically. LVF is an easy, fast and inexpensive EV isolation method that allows for the isolation of both ectosomes and exosomes from high-volume sources with good repeatability. We believe that it could be an efficient alternative to commonly applied methods.

Metformin attenuates adhesion between cancer and endothelial cells in chronic hyperglycemia by recovery of the endothelial glycocalyx barrier.

BACKGROUND: Epidemiologic studies suggest that diabetes is associated with an increased risk of cancer. Concurrently, clinical trials have shown that metformin, which is a first-line antidiabetic drug, displays anticancer activity. The underlying mechanisms for these effects are, however, still not well recognized. METHODS: Methods based on atomic force microscopy (AFM) were used to directly evaluate the influence of metformin on the nanomechanical and adhesive properties of endothelial and cancer cells in chronic hyperglycemia. AFM single-cell force spectroscopy (SCFS) was used to measure the total adhesion force and the work of detachment between EA.hy926 endothelial cells and A549 lung carcinoma cells. Nanoindentation with a spherical AFM probe provided information about the nanomechanical properties of cells, particularly the length and grafting density of the glycocalyx layer. Fluorescence imaging was used for glycocalyx visualization and monitoring of E-selectin and ICAM-1 expression. RESULTS: SCFS demonstrated that metformin attenuates adhesive interactions between EA.hy926 endothelial cells and A549 lung carcinoma cells in chronic hyperglycemia. Nanoindentation experiments, confirmed by confocal microscopy imaging, revealed metformin-induced recovery of endothelial glycocalyx length and density. The recovery of endothelial glycocalyx was correlated with a decrease in the surface expression of E-selectin and ICAM-1. CONCLUSION: Our results identify metformin-induced endothelial glycocalyx restoration as a key factor responsible for the attenuation of adhesion between EA.hy926 endothelial cells and A549 lung carcinoma cells. GENERAL SIGNIFICANCE: Metformin-induced glycocalyx restoration and the resulting attenuation of adhesive interactions between the endothelium and cancer cells may account for the antimetastatic properties of this drug.

Retinopathy severity correlates with RANTES concentrations and CCR 5-positive microvesicles in diabetes.

INTRODUCTION: RANTES regulates leukocyte recruitment to areas affected by the inflammatory process. Microvesicles (MVs) belong to a subpopulation of extracellular vesicles and show proangiogenic potential by transferring bioactive molecules to target cells. OBJECTIVES: the aim of this study was to determine the relationship between circulating proangiogenic factors (MVs and RANTES) and diabetes complications in patients with different severities of diabetic retinopathy (DR). CCR5 (CD195) receptors transported by annexin V-labeled MVs were also investigated. PATIENTS AND METHODS: Diabetic patients (n = 61), among whom 35 had confirmed DR classified according to guidelines, and controls (n = 25) were included. MVs were isolated by centrifugation and analyzed using flow cytometry, RANTES was assessed by ELISA. RESULTS: the study group differed from the control group with respect to BMI, age, heart rate and systolic blood pressure. Additionally, glucose and creatinine concentrations were significantly increased: 5.30 [5.09-5.62] vs. 9.38 [7.48-11.55] (p <0.0001) mmol/l and 74.59 [64-84] vs. 89.00 [77.11-105.44] µmol/l (p = 0.0005), respectively. RANTES concentrations were significantly increased in diabetic patients compared to those of controls (15.5 (9.7-18.1) vs. 8.9 (0.9-14.6) µg/ml (p = 0.011)), and RANTES concentration significantly increased with respect to nonproliferative DR progression. Moreover, the number of CCR5-positive MVs was significantly increased in patients with heavy nonproliferative diabetic retinopathy (HNPDR) compared to those with so nonproliferative DR (SNPDR): 1178 [836-2254] vs. 394 [275-799] counts/µl. CONCLUSIONS: Correlation of RANTES concentrations with the stage of nonproliferative DR and the statistically significant dependence of CCR5-positive MVs with disease progression suggest that MVs and RANTES can be considered new biomarkers.