Incremental learning of chunk data for online pattern classification systems

This paper presents a pattern classification system in which feature extraction and classifier learning are simultaneously carried out not only online but also in one pass where training samples are presented only once. For this purpose, we have extended incremental principal component analysis (IPCA) and some classifier models were effectively combined with it. However, there was a drawback in this approach that training samples must be learned one by one due to the limitation of IPCA. To overcome this problem, we propose another extension of IPCA called chunk IPCA in which a chunk of training samples is processed at a time. In the experiments, we evaluate the classification performance for several large-scale data sets to discuss the scalability of chunk IPCA under one-pass incremental learning environments. The experimental results suggest that chunk IPCA can reduce the training time effectively as compared with IPCA unless the number of input attributes is too large. We study the influence of the size of initial training data and the size of given chunk data on classification accuracy and learning time. We also show that chunk IPCA can obtain major eigenvectors with fairly good approximation.     

Electrospun PEO/rGO Scaffolds: The Influence of the Concentration of rGO on Overall Properties and Cytotoxicity

Reduced graphene oxide (rGO) is one of the graphene derivatives that can be employed to engineer bioactive and/or electroactive scaffolds. However, the influence of its low and especially high concentrations on scaffolds’ overall properties and cytotoxicity has yet to be explored. In this study, polyethylene oxide (PEO)-based scaffolds containing from 0.1 to 20 wt% rGO were obtained by electrospinning. Morphological, thermal and electrical properties of the scaffolds were characterized by SEM, Raman spectroscopy, XRD, DSC and electrical measurements. The diameter of the fibers decreased from 0.52 to 0.19 µm as the concentration of rGO increased from 0.1 wt% to 20 wt%. The presence of rGO above the percolation threshold (5.7 wt%) resulted in a significantly reduced electrical resistivity of the scaffolds. XRD and Raman analysis revealed delamination of the graphene layers (interlayer spacing increased from 0.36 nm to 0.40–0.41 nm), and exfoliation of rGO was detected for the samples with an rGO concentration lower than 1 wt%. In addition, an evident trend of increasing cell viability as a function of the rGO concentration was evidenced. The obtained results can serve as further guidance for the judicious selection of the rGO content incorporated into the PEO matrix for constructing electroactive scaffolds.     

Functionality of chitosan-halloysite nanocomposite films for sustained delivery of antibiotics: The effect of chitosan molar mass

This study was designed to investigate functionality of tetracycline-loaded chitosan-halloysite nanocomposite films, with focus on evaluating the influence of chitosan molar mass on films applicability for sustained local antibiotic delivery. The films were prepared by casting and solvent evaporation using low, medium, and high molar mass chitosan. SEM analysis revealed compact, nonporous and rough surface of the nanocomposite films due to the presence of halloysite agglomerates and tetracycline crystals. Increasing chitosan molar mass led to higher values of elongation at break (from 21.65 ± 2.65 to 34.48 ± 2.34%), tensile strength (from 134.8 ± 13.21 to 246.36 ± 14.69 MPa), and elastic modulus (from 633.79 ± 128.37 to 716.55 ± 60.76 MPa) of the nanocomposite films. FT-IR, XRPD, and thermal analyses confirmed molar mass dependent chitosan-halloysite interactions and improved thermal stability of the nanocomposite films in comparison with chitosan films. The nanocomposite films released tetracycline in a sustained manner, with the slowest release achieved from the films consisting of low molar mass chitosan. Chitosan molar mass was confirmed to be a functionality-related characteristic of chitosan-halloysite nanocomposite films as potential sustained-release carriers for topical delivery of antibiotics. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2020 , 137, 48406.     

Atrazine in groundwater of Vojvodina Province

The objective of this study was to investigate concentrations of atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-triazine), deethylatrazine (DEA) (2-amino-4-chloro-6-isopropylamino-1,3,5-triazine), deisopropylatrazine (DIA) (2-amino-4-chloro-6-ethylamino-1,3,5-triazine) and deethyldeisopropylatrazine (DEIA) (6-chloro-2,4-diamino-1,3,5-triazine) in groundwaters of Vojvodina Province. A study was conducted during April 2001. Some 110 samples of groundwater were taken from near surface aquifers. The water samples were first passed through a disk containing solid matrix coated with a chemically bonded C-18 organic phase. The disk was then eluted with supercritical carbon dioxide to remove compounds from the sorbent. Finally the extract was injected into capillary gas chromatograph. Average concentrations were 0.198 μg L⁻¹ for atrazine, 0.116 μg L⁻¹ for DEA, 0.043 μg L⁻¹ for DIA and 0.077 μg L⁻¹ for DEIA.     

Pilot Studies of Ozonation for Inactivation of Artemia salina Nauplii in Ballast Water

A pilot-plant study was conducted in the Republic of Croatia to determine the applicability of ozonation for inactivation of non-indigenous species and to provide necessary information regarding use of ozone as a ballast water treatment option. Nauplii of the brine shrimp Artemia salina were used as model organisms to investigate the efficacy of ozonation at three different ozone dosages (2.4, 3.7 and 10.9 mg L^?1). Mortality of Artemia nauplii at 98.6%, was achieved after 3 h of exposure in ozone-treated water with the highest ozone dosage. Our results indicated that ozonation is a promising treatment for controlling non-indigenous and potentially invasive species; however, to draw more general conclusions, several species with higher level of resistance to ozone are required and will be studied in the future.     

N-acetylcysteine Can Induce Massive Oxidative Stress,Resulting in Cell Death with Apoptotic Features in Human Leukemia Cells

N-acetylcysteine (NAC), often used as an antioxidant-scavenging reactive oxygen species (ROS) in vitro, was recently shown to increase the cytotoxicity of other compounds through ROS-dependent and ROS-independent mechanisms. In this study, NAC itself was found to induce extensive ROS production in human leukemia HL-60 and U937 cells. The cytotoxicity depends on ROS-modulating enzyme expression. In HL-60 cells, NAC activated NOX2 to produce superoxide (O₂•⁻). Its subsequent conversion into H₂O₂ by superoxide dismutase 1 and 3 (SOD1, SOD3) and production of ClO⁻ from H₂O₂ by myeloperoxidase (MPO) was necessary for cell death induction. While the addition of extracellular SOD potentiated NAC-induced cell death, extracellular catalase (CAT) prevented cell death in HL-60 cells. The MPO inhibitor partially reduced the number of dying HL-60 cells. In U937 cells, the weak cytotoxicity of NAC is probably caused by lower expression of NOX2, SOD1, SOD3, and by the absence of MOP expression. However, even here, the addition of extracellular SOD induced cell death in U937 cells, and this effect could be reversed by extracellular CAT. NAC-induced cell death exhibited predominantly apoptotic features in both cell lines. Conclusions: NAC itself can induce extensive production of O₂•⁻ in HL-60 and U937 cell lines. The fate of the cells then depends on the expression of enzymes that control the formation and conversion of ROS: NOX, SOD, and MPO. The mode of cell death in response to NAC treatment bears apoptotic and apoptotic-like features in both cell lines.     

Numerical simulation of combined screw compressor–expander machines for use in high pressure refrigeration systems

Recent interest in natural refrigerants has created a new impetus for studies of CO₂ as a working fluid in vapour compression systems for refrigeration and air conditioning. Two major drawbacks to its use are the very high pressure differences required across the compressor and the large efficiency losses associated with the throttling process in the refrigeration cycle. It is shown how these disadvantages can be minimised by the use of a screw machine both to compress the gas and use the expansion process to recover power. Both these functions can be performed simultaneously, using only one pair of rotors, in a configuration that partially balances out the forces induced by the pressure difference and hence, reduces the bearing loads to an acceptable level. A further feature is the use of rotors, which seal on both contacting surfaces so that the same profile may be used for the expander and the compressor sections. This enables the rotors performing both these functions to be machined or ground in the same cutting operation and then separated by machining a parting slot in them. Computational Continuum Mechanics comprising both, fluid flow and structural analysis is used in this paper for the investigation of fluid-solid interaction in such machines.     

Solution properties of poly(fluorostyrene-co-chlorostyrene) copolymers. I. Light scattering,differential refractometry and viscometry

The weight-average molecular weights (M?_w) of nominally random, unfractionated copolymers of ortho- and para-fluorostyrene and ortho- and para-chlorostyrene were determined by light scattering measurements in tetrahydro-furan, toluene, carbon tetrachloride and chloroform. It was shown that there is no significant variation in measured M?_w in the various solvents, a finding indicating that the copolymers are not compositionally heterogeneous. Intrinsic viscosity measurements in the same solvents established a consistent relationship between [n] and M?_w despite the differences in copolymer compositions. It was not possible to establish a similar relationship between the second virial coefficient A₂ and M?_w. It was concluded that measurements of the specific refractive index increment could be used for determining copolymer composition if the measurements were performed in thermodynamically poorer solvents. The results established for the Mark-Houwink constant α, intrinsic viscosity, and A₂ values indicated that these solvated copolymer molecules are in a less expanded conformation than are polystyrene molecules of similar molecular weights in a given solvent.     

Effect of SEM electron beam on the hydrogen desorption of pre-charged amorphous Cu33Ti67 alloys

The crystallization and hydrogen desorption of pre-charged amorphous Cu₃₃Ti₆₇ alloy have been studied by differential scanning calorimetry (DSC), thermogravimetry (TG) and scanning electron microscopy (SEM). The effect of the SEM electron beam on the desorption process was analyzed. It was found that small amounts of absorbed hydrogen do not influence on the thermal stability of the glass, but significantly reduce its crystallization enthalpy. At higher hydrogen concentrations the thermal stability of the glass was strongly reduced (with about 250 K), as the low-temperature crystallization product was mainly extremely fine nanocrystalline Cu₂Ti, formed as a result of the Cu enrichment of the amorphous phase after hydriding. Combining DSC, TG and SEM it was proved that during the electron beam exposure the weaker bonded hydrogen, which desorbs thermally at 430–500 K, releases. Annealing at higher temperatures (> 500 K) showed clearly the existence of nanocrystalline Cu₂Ti and TiH₂ phases. Only after the TiH₂ decomposition (at ~ 680 K) the equilibrium crystalline phase (Ti₂Cu) was formed, but with finer microstructure.     

Classification consistency analysis for bootstrapping gene selection

Consistency modelling for gene selection is a new topic emerging from recent cancer bioinformatics research. The result of operations such as classification, clustering, or gene selection on a training set is often found to be very different from the same operations on a testing set, presenting a serious consistency problem. In practice, the inconsistency of microarray datasets prevents many typical gene selection methods working properly for cancer diagnosis and prognosis. In an attempt to deal with this problem, this paper proposes a new concept of classification consistency and applies it for microarray gene selection problem using a bootstrapping approach, with encouraging results.