Deep-MEG: spatiotemporal CNN features and multiband ensemble classification for predicting the early signs of Alzheimer’s disease with magnetoencephalography

Neural Computing and Applications - In this paper, we present the novel Deep-MEG approach in which image-based representations of magnetoencephalography (MEG) data are combined with ensemble...     

Sulfur containing polymers

Summary Copolymers containing -S-CO-S-and/or -O-CO-S-groups have been synthesized mainly by interfacial polycondensation.Different chemical structures were obtained by reacting 1,3-benzene dithiol (BDT) respectively with phosgene alone, phosgene and bisphenol-A (BPA), bischloroformate of BPA, BPA polycarbonate oligomers and by reacting phosgene with the products of BPA polycarbonate degraded with BDT. The chemical structures of the copolymers were investigated by IR, ¹H-NMR and ¹³C-NMR; molecular weights were determined by viscometry and vapor pressure osmometry. Although no attempt was made to find the optimum conditions for high molecular weight, some copolymers with fairly high mol. weight were obtained.     

Occurrence patterns of alien freshwater turtles in a large urban pond ‘Archipelago’ (Rome,Italy): Suggesting hypotheses on root causes

The patterns of the occurrence and distribution of alien freshwater turtle species in an urban pond archipelago (Rome, Italy) were analysed, with the aim of exploring the role of a set of factors (type of ponds, landscape context, size area, distance from the nearest road) with a generalized linear model approach. A total of 311 ponds subdivided in three types (fountains, small basins, lakes) embedded in different landscape contexts (public parks, private parks, urban areas) at differing distances from the nearest road were sampled. Six non‐native freshwater turtle species in 31 sites were recorded (9.97%). Lakes exhibited the highest occurrence rates of alien freshwater turtles, compared to small basins and fountains. Freshwater turtle species in urbanized areas were only observed in parks (both public and private). In both the public and private parks, the lakes exhibited the highest percentage of occupied sites, with fountains being the lowest. A direct and significant relationship was observed between pond size and species richness. The distance from the nearest road did not appear to affect species richness. A first interpretation of the data from this study facilitated the postulation of two a posteriori hypotheses that should be tested, as follows: (i) the causal process of turtle release is random, and the rate of extinction (and recapture) is higher in smaller ponds, thus producing the observed pattern; and (ii) the turtle release is not random, and people actively select the ponds they consider more suitable for their pet animals. In this study, it appears the lakes were perceived by those who abandon their pets as the most ecologically suitable habitats among other pond types to accommodate the different species of turtles. Knowledge of people's attitudes in regard to releasing pet animals also might assist managers of public green spaces to develop strategies aimed to preserve local biodiversity, and to educate the public about the conservation issue represented by the alien species.     

Thermodynamic interactions of branched polyethylene in polar and nonpolar systems

The thermodynamic properties of polymer solutions are frequently described in terms of the Flory-Huggins equation. This equation includes a parameter χ, which depends upon the intermolecular forces acting between the molecules in a solution. The experimental determination of χ was performed by an improved microtechnique and extended to a wide range of polar and nonpolar diluents of polyethylene. Careful correlations are prescribed for calculating χ from pure-component properties; they are based on an extension of the Hildebrand-Scatchard theory of solutions and on the theory of intermolecular forces. Polar (τ) and nonpolar (δ) solubility parameters are presented for a variety of solvents. For polyethylene—nonpolar solvent systems we have emphasized the factor deciding the sign of heat of mixing, while for polyethylene-polar solvent systems we have determined the contribution of dipole-induced dipole interactions ψ (δτ) in interchange-energy density B and, hence, χ.     

In vivo characterization of renal auto-antigens involved in human auto-immune diseases: the case of membranous glomerulonephritis

Renal auto-immune diseases represent a major source of morbidity in humans. For many years the knowledge on mechanisms of auto-immunity involving the kidney has been uniquely based on animal models. However, these findings often could not be readily translated to humans owing to notably difference in antigen expression by human podocytes. One example is Heymann nephritis (HN), the experimental model of human membranous glomerulonephritis (MGN), which is obtained in rats by injecting antibodies against megalin, a protein that is not present in human glomeruli. Human studies could not be done in the past since sequencing required too much material exceeding what obtainable from tissue biopsies in vivo. Research is now on the way to identify auto-antigens and isolate specific auto-antibodies in humans. New technology developments based on tissue microdissection and proteomical analysis have facilitated the recent discoveries, allowing direct analysis of human tissue in vivo. Major advances on the pathogenesis of MGN, the prototype for the formation and glomerular deposition of auto-antibodies, are now in progress. Two independent groups have, in fact, demonstrated the existence of specific IgG(4) against phospholipase A2 receptor, aldose reductase and Mn-superoxide dismutase in glomerular eluates and in plasma of a prominent part of patients with MGN, suggesting a major role of these proteins as auto-antigens in human MGN. This review will focalize these aspects outlining the contribution of proteomics in most recent developments.     

PET-Derived Radiomics and Artificial Intelligence in Breast Cancer: A Systematic Review

Breast cancer (BC) is a heterogeneous malignancy that still represents the second cause of cancer-related death among women worldwide. Due to the heterogeneity of BC, the correct identification of valuable biomarkers able to predict tumor biology and the best treatment approaches are still far from clear. Although molecular imaging with positron emission tomography/computed tomography (PET/CT) has improved the characterization of BC, these methods are not free from drawbacks. In recent years, radiomics and artificial intelligence (AI) have been playing an important role in the detection of several features normally unseen by the human eye in medical images. The present review provides a summary of the current status of radiomics and AI in different clinical settings of BC. A systematic search of PubMed, Web of Science and Scopus was conducted, including all articles published in English that explored radiomics and AI analyses of PET/CT images in BC. Several studies have demonstrated the potential role of such new features for the staging and prognosis as well as the assessment of biological characteristics. Radiomics and AI features appear to be promising in different clinical settings of BC, although larger prospective trials are needed to confirm and to standardize this evidence.     

Effectiveness of Semi-Supervised Active Learning in Automated Wound Image Segmentation

Appropriate wound management shortens the healing times and reduces the management costs, benefiting the patient in physical terms and potentially reducing the healthcare system’s economic burden. Among the instrumental measurement methods, the image analysis of a wound area is becoming one of the cornerstones of chronic ulcer management. Our study aim is to develop a solid AI method based on a convolutional neural network to segment the wounds efficiently to make the work of the physician more efficient, and subsequently, to lay the foundations for the further development of more in-depth analyses of ulcer characteristics. In this work, we introduce a fully automated model for identifying and segmenting wound areas which can completely automatize the clinical wound severity assessment starting from images acquired from smartphones. This method is based on an active semi-supervised learning training of a convolutional neural network model. In our work, we tested the robustness of our method against a wide range of natural images acquired in different light conditions and image expositions. We collected the images using an ad hoc developed app and saved them in a database which we then used for AI training. We then tested different CNN architectures to develop a balanced model, which we finally validated with a public dataset. We used a dataset of images acquired during clinical practice and built an annotated wound image dataset consisting of 1564 ulcer images from 474 patients. Only a small part of this large amount of data was manually annotated by experts (ground truth). A multi-step, active, semi-supervised training procedure was applied to improve the segmentation performances of the model. The developed training strategy mimics a continuous learning approach and provides a viable alternative for further medical applications. We tested the efficiency of our model against other public datasets, proving its robustness. The efficiency of the transfer learning showed that after less than 50 epochs, the model achieved a stable DSC that was greater than 0.95. The proposed active semi-supervised learning strategy could allow us to obtain an efficient segmentation method, thereby facilitating the work of the clinician by reducing their working times to achieve the measurements. Finally, the robustness of our pipeline confirms its possible usage in clinical practice as a reliable decision support system for clinicians.