Time filter

Source Type

Surchev S.,University Burgas | Sotirov S.,University Burgas | Korneta W.,The College of Computer Science
International Journal Bioautomation | Year: 2013

Many objects and processes inspired by the nature have been recreated by the scientists. The inspiration to create a Multilayer Neural Network came from human brain as member of the group. It possesses complicated structure and it is difficult to recreate, because of the existence of too many processes that require different solving methods. The aim of the following paper is to describe one of the methods that improve learning process of Artificial Neural Network. The proposed generalized net method presents Regularization process in Multilayer Neural Network. The purpose of verification is to protect the neural network from overfitting. The regularization is commonly used in neural network training process. Many methods of verification are present, the subject of interest is the one known as Regularization. It contains function in order to set weights and biases with smaller values to protect from overfitting. Source

The paper presents the conductivity of the ground effect on the distribution of temperature field in three-phasehighvoltage cables, 64/110 kV, depending on the depth of their arrangement in the ground. The simulation uses the finite element method FEM. Source

Waszkiewicz N.,Medical University of Bialystok | Chojnowska S.,The College of Computer Science | Zalewska A.,Medical University of Bialystok | Szulc A.,Medical University of Bialystok | Szajda S.D.,Medical University of Bialystok
Drug and Alcohol Dependence | Year: 2013

Background: A sensitive alcohol marker, β-hexosaminidase (HEX), in the saliva of alcoholics, is investigated for the first time. Methods: The activity, specific-activity and output of total HEX and its isoenzymes HEX A and HEX B were measured in the saliva of healthy controls (C), alcohol-dependent non-smokers (ANS), and alcohol-dependent smokers (AS). Results: We observed a significantly increased activity/specific-activity and output of HEX A in the ANS and AS groups, due to the inflammatory state of the oral-cavity/salivary-glands. Significantly increased activity of HEX A contributed to an increase in the salivary activity of the total HEX in the ANS group. A significant decrease in the activity/specific-activity of HEX B in AS seemed to be due to HEX B inactivation by cigarette smoke. We noticed a tendency for deteriorated dental state (lower decayed-missing-filled-teeth index - DMFT), worse periodontal state (higher gingival index - GI and papilla-bleeding index - PBI) in AS, and worse periodontal state (higher GI) in ANS, as compared to the controls. We found no differences in the salivary protein concentrations between all groups and decreased salivary flow in both alcoholic groups as compared to the controls. In alcoholics, the area under the curve (AUC) for HEX A activity/specific-activity was significantly greater than for HEX and HEX B. The salivary HEX A activity/specific-activity had good/excellent sensitivity and specificity in smoking and non-smoking alcoholics, whereas salivary HEX and HEX B had poor/fair sensitivity and specificity. Conclusions: Salivary HEX A may be helpful in the diagnosis of chronic alcohol intoxication, even in smokers. © 2012 Elsevier Ireland Ltd. Source

Szweda P.,Technical University of Gdansk | Schielmann M.,Technical University of Gdansk | Milewski S.,Technical University of Gdansk | Frankowska A.,Siedlce University Of Natural Sciences And Humanities | Jakubczak A.,The College of Computer Science
Polish Journal of Microbiology | Year: 2012

The aim of the study was phenotypic and genotypic analysis of 132 S. aureus strains isolated from mastitis in eastern Poland in respect to their biofilm formation ability. The analysis of the size polymorphism of fragment X in the gene encoding protein A (spa) revealed high genetic differentiation of the analyzed group of isolates. The ability of biofilm formation by the isolates was tested using two phenotypic methods. The Congo Red plate assay was found to be irreproducible and very subjective. More objective results were obtained using the spectrophotometric, microtiter plate assay. Most of the isolates, namely 76/132 (57.6%) were classified as biofilm producers depending on the value of absorbance in the microtiter plate test. All of the isolates tested were found to possess both icaA and icaD genes, while the bap gene was absent in all strains. Source

Stachelska M.A.,The College of Computer Science
Acta Scientiarum Polonorum, Technologia Alimentaria | Year: 2015

Background: The aim of this paper was to assess the inhibitory properties of salts of phenolic acids against Escherichia coli O157:H7 ATCC 8739. Escherichia coli O157:H7 is a pathogen which is able to produce verotoxins provoking hemorrhagic diarrhea in humans. There is a strong need for the effective natural methods eliminating E. coli O157:H7 from food. Methodology: The following salts were tested: sodium, potassium and lithium salts of ortho-coumaric, meta-coumaric and para-coumaric acids. The 1%, 2%, 3%, 4% and 5% water solutions of each substance were prepared. Agar-well diffusion method was applied. Petri dishes were incubated at 35°C for 24 h. At the end of the incubation period, inhibition zones which appeared on the medium Petri dishes were calculated in millimeters. Results: It was found that lithium salt of o-coumaric acid, potassium salt of o-coumaric acid, lithium salt of m-coumaric acid and sodium salt of m-coumaric acid were most effective towards E. coli O157:H7, while potassium salt of m-coumaric acid, a sodium salt of p-coumaric acid were slightly less effective and lithium salt of p-coumaric acid did not possess any antimicrobial activity. Conclusion: The salts of phenolic acids having various structural features showed different characteristics towards foodborne pathogens. Such findings indicate that phenolic acids and their salts may be a potential bio-alternative for chemical food preservation. © by Wydawnictwo Uniwersytetu Przyrodniczego w Poznaniu. Source

Discover hidden collaborations