A Feature Selection Algorithm for the Regularization of Neuron Models

This paper presents a novel training method for estimating the parameters of retina models, such as integrate-and-fire (IF) or Poisson based. The presented models are constructed using a set of linear and nonlinear filters, which are described by basis functions and Taylor polynomials, respectively. This approach allows for the identification of a set of features that can be used for reproducing retina responses. By using the Bayesian-Laplace feature selection algorithm herein proposed, an efficient model with a reduced set of parameters is achieved. Experimental results show that the proposed algorithm is able to remove unimportant features while still accurately reproducing retina responses. These results also show that the IF model is able to mimic the retina visual processing system using less parameters than the Poisson-based model.     

Apoferritin Modified Magnetic Particles as Doxorubicin Carriers for Anticancer Drug Delivery

Magnetic particle mediated transport in combination with nanomaterial based drug carrier has a great potential for targeted cancer therapy. In this study, doxorubicin encapsulation into the apoferritin and its conjugation with magnetic particles was investigated by capillary electrophoresis with laser-induced fluorescence detection (CE-LIF). The quantification of encapsulated doxorubicin was performed by fluorescence spectroscopy and compared to CE-LIF. Moreover, the significant enhancement of the doxorubicin signal was observed by addition of methanol into the sample solution.     

Optimization of a Continuous Precipitation Process to Produce Nanoscale BaSO4

The effect of supersaturation, reaction temperature, and mixing intensity on particle size was investigated. Sterical stabilization of barium sulfate suspensions was applied to prevent formation of agglomerates. This allowed a reactant ratio of 1:1, thus maximizing product yield. The local supersaturation is strongly affected by the mixing intensity that can be characterized by Reynolds numbers. The significant decrease in particle size was observed by increasing the Reynolds number from 600 to 8000. A higher reactant concentration leads to a higher degree of supersaturation, and finer particles are precipitated. The particle size can be reduced with increasing reactant concentration. The degree of supersaturation increases with temperature reduction, i.e., the particle size will be reduced at low temperature. In addition, nucleation and growth kinetics are changed in a way that reduces the particle size. The optimized lab‐scale process is capable of producing over 1 kg h^–1 of nanoscaled BaSO₄ with a median diameter of 75 nm.     

A Novel Insight into the Cardiotoxicity of Antineoplastic Drug Doxorubicin

Doxorubicin is a commonly used antineoplastic agent in the treatment of many types of cancer. Little is known about the interactions of doxorubicin with cardiac biomolecules. Serious cardiotoxicity including dilated cardiomyopathy often resulting in a fatal congestive heart failure may occur as a consequence of chemotherapy with doxorubicin. The purpose of this study was to determine the effect of exposure to doxorubicin on the changes in major amino acids in tissue of cardiac muscle (proline, taurine, glutamic acid, arginine, aspartic acid, leucine, glycine, valine, alanine, isoleucine, threonine, lysine and serine). An in vitro interaction study was performed as a comparison of amino acid profiles in heart tissue before and after application of doxorubicin. We found that doxorubicin directly influences myocardial amino acid representation even at low concentrations. In addition, we performed an interaction study that resulted in the determination of breaking points for each of analyzed amino acids. Lysine, arginine, β-alanine, valine and serine were determined as the most sensitive amino acids. Additionally we compared amino acid profiles of myocardium before and after exposure to doxorubicin. The amount of amino acids after interaction with doxorubicin was significantly reduced (p = 0.05). This fact points at an ability of doxorubicin to induce changes in quantitative composition of amino acids in myocardium. Moreover, this confirms that the interactions between doxorubicin and amino acids may act as another factor most likely responsible for adverse effects of doxorubicin on myocardium.     

On the stability of delamination growth at scratching of thin film structures

Scratching of thin film/substrate structures is studied theoretically and numerically. The results are discussed in connection to delamination initiation and in particular subsequent growth at scratching. The material behavior of the film is described by classical elastoplasticity accounting for large deformations. The deformation of the substrate is neglected indicating that the results are pertinent to soft thin films. The numerical investigation is performed using the finite element method (FEM) and the numerical strategy is discussed in some detail. The results from this study show that delamination growth at thin film scratching is a stable feature with crack arrest occurring at a decreasing load.     

Energy absorption during compression and impact of dry elastic-plastic spherical granules

The discrete modelling and understanding of the particle dynamics in fluidized bed apparatuses, mixers, mills and others are based on the knowledge about the physical properties of particles and their mechanical behaviour during slow, fast and repeated stressing. In this paper model parameters (modulus of elasticity, stiffness, yield pressure, restitution coefficient and strength) of spherical granules (γ-Al₂O₃, zeolites 4A and 13X, sodium benzoate) with different mechanical behaviour have been measured by single particle compression and impact tests. Starting with the elastic compression behaviour of granules as described by Hertz theory, a new contact model was developed to describe the force-displacement behaviour of elastic-plastic granules. The aim of this work is to understand the energy absorption during compression (slow stressing velocity of 0.02 mm/s) and impact (the impact velocity of 0.5–4.5 m/s) of granules. For all examined granules the estimated energy absorption during the impact is found to be far lower than that during compression. Moreover, the measured restitution coefficient is independent of the impact velocity in the examined range and independent of the load intensity by compression (i.e. maximum compressive load). In the case of repeated loading with a constant load amplitude, the granules show cyclic hardening with increasing restitution coefficient up to a certain saturation in the plastic deformation. A model was proposed to describe the increase of the contact stiffness with the number of cycles. When the load amplitude is subsequently increased, further plastic deformation takes place and the restitution coefficient strongly decreases.     

Development of a non-destructive micro-analytical method for stable carbon isotope analysis of transmission electron microscope (TEM) samples

The biogenicity of ancient morphological microfossil-like objects can be established by linking morphological (e.g. cell remnants and extracellular polymeric matrix) and chemical (e.g. isotopes, biomarkers and biominerals) evidence indicative of microorganisms or microbial activity. We have developed a non-destructive micro-analytical ion beam system capable of measuring with high spatial resolution the stable carbon isotope ratios of thin samples used for transmission electron microscopy. The technique is based on elastic scattering of alpha particles with an energy of 2.751 MeV. At this energy the ¹³C cross section is enhanced relative to the pure Rutherford cross section for ¹³C, whereas the ¹²C cross section is reduced relative to its pure Rutherford cross section. Here we report the initial results of this experimental approach used to characterize ultramicrotomed sections of sulfur-embedded graphite and microbial cells.     

High-resolution capillary zone and gel electrophoresis of structurally similar amphipathic glutathione conjugates based on interaction with beta-cyclodextrins

The tripeptide glutathione is a prominent intracellular constituent that provides protection against genotoxic and carcinogenic electrophiles and is also a component of several biological signal substances. Glutathione conjugates, free glutathione, and glutathione disulfide contain charged amino acid residues, which contribute to solubility in aqueous media. However, the amphipathic nature of glutathione conjugates and the small differences that may distinguish the S substituents, pose analytical problems in their resolution. The present study demonstrates how homologous S-alkyl and S-benzyl conjugates of high structural similarity can be efficiently resolved by capillary electrophoresis. Inclusion of beta-cyclodextrins in the buffer or in a polyacrylamide gel affords baseline separation of the analytes. The separation methods described are applicable to enzyme assays in vitro and to the identification and quantification of glutathione conjugates of importance in toxicology and physiology. The contribution of beta-cyclodextrin to the separation is primarily based on interactions between its hydrophobic cavity and the S-alkyl and S-benzyl groups of the analytes.     

A multilevel relaxation algorithm for simultaneous localization and mapping

This paper addresses the problem of simultaneous localization and mapping (SLAM) by a mobile robot. An incremental SLAM algorithm is introduced that is derived from multigrid methods used for solving partial differential equations. The approach improves on the performance of previous relaxation methods for robot mapping, because it optimizes the map at multiple levels of resolution. The resulting algorithm has an update time that is linear in the number of estimated features for typical indoor environments, even when closing very large loops, and offers advantages in handling nonlinearities compared with other SLAM algorithms. Experimental comparisons with alternative algorithms using two well-known data sets and mapping results on a real robot are also presented.