QuMinS: fast and scalable querying, mining and summarizing multi-modal databases

Given a large image set, in which very few images have labels, how to guess labels for the remaining majority? How to spot images that need brand new labels different from the predefined ones? How to summarize these data to route the user’s attention to what really matters? Here we answer all these questions. Specifically, we propose QuMinS, a fast, scalable solution to two problems: (i) Low-labor labeling (LLL) – given an image set, very few images have labels, find the most appropriate labels for the rest; and (ii) Mining and attention routing – in the same setting, find clusters, the top-'N IND.O' outlier images, and the 'N IND.R' images that best represent the data. Experiments on satellite images spanning up to 2.25 GB show that, contrasting to the state-of-the-art labeling techniques, QuMinS scales linearly on the data size, being up to 40 times faster than top competitors (GCap), still achieving better or equal accuracy, it spots images that potentially require unpredicted labels, and it works even with tiny initial label sets, i.e., nearly five examples. We also report a case study of our method’s practical usage to show that QuMinS is a viable tool for automatic coffee crop detection from remote sensing images.

Percolated non-Newtonian flow for silicone obtained from exfoliated bioinorganic layered double hydroxide intercalated with amino acid

A simple and scalable procedure was used to obtain thin, stable, homogeneous, and easy-to-handle films composed of silicone derived from dimethicones containing dispersed hydrotalcite-type materials previously organo-modified with amino acids. The absence of the typical X-ray pattern of the bioinorganic LDH filler suggested an exfoliation process that was further indirectly evidenced by a drastic change in the rheological behavior, which turned from a quasi-Newtonian behavior for the silicone free of LDH filler to an extensive developed gel-like structure for the nanocomposite derivatives. Visualized by the shear-thinning exponent of the complex viscosity in the low-frequency range, the percolation threshold was evident for filler loading as low as <5 w/W%, suggesting the presence of a largely developed interface between the filler and the polymer. The increase of more than one order of magnitude in viscosity was explained by the rather strong attrition phenomenon between the tethered amino acid anions and the silicone chains. UVB radiation absorption profiles make such bioinorganic polymer nanocomposites potentially applicable in skin protection. Thermo-gravimetric analysis revealed significant improvement in the thermal stability, especially in the final step of the polymer combustion, thus underlining the role of the hybrid material as a thermal retardant agent. (C) 2011 Elsevier B.V. All rights reserved.

Tailored SERS substrates obtained with cathodic arc plasma ion implantation of gold nanoparticles into a polymer matrix

This manuscript reports on the fabrication of plasmonic substrates using cathodic arc plasma ion implantation, in addition to their performance as SERS substrates. The technique allows for the incorporation of a wide layer of metallic nanoparticles into a polymer matrix, such as PMMA. The ability to pattern different structures using the PMMA matrix is one of the main advantages of the fabrication method. This opens up new possibilities for obtaining tailored substrates with enhanced performance for SERS and other surface-enhanced spectroscopies, as well as for exploring the basic physics of patterned metal nanostructures. The architecture of the SERS-active substrate was varied using three adsorption strategies for incorporating a laser dye (rhodamine): alongside the nanoparticles into the polymer matrix, during the polymer cure and within nanoholes lithographed on the polymer. As a proof-of-concept, we obtained the SERS spectra of rhodamine for the three types of substrates. The hypothesis of incorporation of rhodamine molecules into the polymer matrix during the cathodic arc plasma ion implantation was supported by FDTD (Finite-Difference Time-Domain) simulations. In the case of arrays of nanoholes, rhodamine molecules could be adsorbed directly on the gold surface, then yielding a well-resolved SERS spectrum for a small amount of analyte owing to the short-range interactions and the large longitudinal field component inside the nanoholes. The results shown here demonstrate that the approach based on ion implantation can be adapted to produce reproducible tailored substrates for SERS and other surface-enhanced spectroscopies.

SK-HEP cells and lentiviral vector for production of human recombinant factor VIII

Hemophilia A is caused by a deficiency in coagulation factor VIII. Recombinant factor VIII can be used as an alternative although it is unavailable for most patients. Here, we describe the production of a human recombinant B-domain-deleted FVIII (rBDDFVIII) by the human cell line SK-HEP-1, modified by a lentiviral vector rBDDFVIII was produced by recombinant SK-HEP cells (rSK-HEP) at 1.5-2.1 IU/10(6) in 24 h. The recombinant factor had increased in vitro stability when compared to commercial pdFVIII. The functionality of rBDDFVIII was shown by its biological activity and by tail-clip challenge in hemophilia A mice. The rSK-HEP cells grew in a scalable system and produced active rBDDFVIII, indicating that this platform production can be optimized to meet the commercial production scale needs.