A Markov random field model for combining optimum-path forest classifiers using decision graphs and game strategy approach

The research on multiple classifiers systems includes the creation of an ensemble of classifiers and the proper combination of the decisions. In order to combine the decisions given by classifiers, methods related to fixed rules and decision templates are often used. Therefore, the influence and relationship between classifier decisions are often not considered in the combination schemes. In this paper we propose a framework to combine classifiers using a decision graph under a random field model and a game strategy approach to obtain the final decision. The results of combining Optimum-Path Forest (OPF) classifiers using the proposed model are reported, obtaining good performance in experiments using simulated and real data sets. The results encourage the combination of OPF ensembles and the framework to design multiple classifier systems. © 2011 Springer-Verlag.

Reproductive migration of fish and movement in a series of reservoirs in the Upper Parana River basin, Brazil

AbstractThis study evaluates the effectiveness of two fish passes at two hydropower dams (Canoas I and II) in the Upper Parana basin, which form part of a cascade of three reservoirs. Fish from 12 migratory species (3089 specimens) were captured during their ascending, reproductive migration and were tagged with hydrostatic tags. The recapture data (294 specimens over two consecutive years) showed that there is a strong tendency for the maintenance of ascending migration through reservoirs with fish passes but with differences in migratory activity within the same species. No eggs, larvae or juveniles of these species were found in samples collected over 5 years in the reservoirs above the fish passes. These data suggest that fish passes have contributed to the restoration of the migratory routes of adult fish but that in the absence of suitable spawning or nursery habitats for these species; they probably act as ecological traps and do not contribute to the recruitment of the species.

Evaluation of chromosomal abnormalities by cIg-FISH and association with proliferative and apoptotic indexes in multiple myeloma

Eighty-six newly diagnosed multiple myeloma (MM) patients from a public hospital of São Paulo (Brazil) were evaluated by cIg-FISH for the presence of del(13)(q14), t(4;14)(p16.3;q32) and del(17)(p13). These abnormalities were observed in 46.5, 9.3, and 7.0% of the patients, respectively. In order to identify the possible role of del(13)(q14) in the physiopathology of MM, we investigated the association between this abnormality and the proliferative and apoptotic indexes of plasma cells. When cases demonstrating t(4;14)(p16.3;q32) and del(17)(p13) were excluded from the analysis, we observed a trend towards a positive correlation between the proportion of cells carrying del(13)(q14) and plasma cell proliferation, determined by Ki-67 expression (r = 0.23, P = 0.06). On the other hand, no correlation between the proportion of cells carrying del(13)(q14) and apoptosis, determined by annexin-V staining, was detected (r = 0.05, P = 0.69). In general, patients carrying del(13)(q14) did not have lower survival than patients without del(13)(q14) (P = 0.15), but patients with more than 80% of cells carrying del(13)(q14) showed a lower overall survival (P = 0.033). These results suggest that, when del(13)(q14) is observed in a high proportion of malignant cells, it may have a role in determining MM prognosis. Another finding was a statistically significant lower overall survival of patients with t(4;14)(p16.3;q32) (P = 0.026). In the present study, almost half the patients with t(4;14)(p16.3;q32) died just after diagnosis, before starting treatment. This fact suggests that, in São Paulo, there may be even more patients with this chromosomal abnormality, but they probably die before being diagnosed due to unfavorable socioeconomic conditions. This could explain the low prevalence of this chromosomal abnormality observed in the present study.

Fidelity and game-based technology in management education

This study explores educational technology and management education by analyzing fidelity in game-based management education interventions. A sample of 31 MBA students was selected to help answer the research question: To what extent do MBA students tend to recognize specific game-based academic experiences, in terms of fidelity, as relevant to their managerial performance? Two distinct game-based interventions (BG1 and BG2) with key differences in fidelity levels were explored: BG1 presented higher physical and functional fidelity levels and lower psychological fidelity levels. Hypotheses were tested with data from the participants, collected shortly after their experiences, related to the overall perceived quality of game-based interventions. The findings reveal a higher overall perception of quality towards BG1: (a) better for testing strategies, (b) offering better business and market models, (c) based on a pace that better stimulates learning, and (d) presenting a fidelity level that better supports real world performance. This study fosters the conclusion that MBA students tend to recognize, to a large extent, that specific game-based academic experiences are relevant and meaningful to their managerial development, mostly with heightened fidelity levels of adopted artifacts. Agents must be ready and motivated to explore the new, to try and err, and to learn collaboratively in order to perform.

Assessment of estrogenic exposure in brown trout (Salmo trutta) in a Swiss midland river: integrated analysis of passive samplers, wild and caged fish, and vitellogenin mRNA and protein

This field study examined the vitellogenin (VTG) biomarker response under conditions of low and fluctuating activities of environmental estrogenicity. The present study was performed on immature brown trout (Salmo trutta) exposed to the small river Luetzelmurg, which is located in the prealpine Swiss midland region and receives effluents from a single sewage treatment plant (STP). To understand better factors influencing the relationship between estrogenic exposure and VTG induction, we compared VTG levels in caged (stationary) and feral (free-ranging) fish, VTG levels in fish from up- and downstream of the STP, and two different methods for quantifying VTG (enzyme-linked immunosorbent assay vs real-time reverse transcription-polymerase chain reaction), and we used passive samplers (polar organic chemical integrative sampler [POCIS]) to integrate the variable, bioaccumulative estrogenic load in the river water over time. The POCIS from the downstream site contained approximately 20-fold higher levels of bioassay-derived estrogen equivalents than the POCIS from the upstream site. In feral fish, this site difference in estrogenic exposure was reflected in VTG protein levels but not in VTG mRNA. In contrast, in caged fish, the site difference was evident only for VTG mRNA but not for VTG protein. Thus, the outcome of VTG biomarker measurements varied with the analytical detection method (protein vs mRNA) and with the exposure modus (caged vs feral). Our findings suggest that for environmental situations with low and variable estrogenic contamination, a multiple-assessment approach may be necessary for the assessment of estrogenic exposure in fish.

Monitoring pollution in river Mure , Romania, Part III: biochemical effect markers in fish and integrative reflection

Along a downstream stretch of River Mure , Romania, adult males of two feral fish species, European chub (Leuciscus cephalus) and sneep (Chondrostoma nasus) were sampled at four sites with different levels of contamination. Fish were analysed for the biochemical markers hsp70 (in liver and gills) and hepatic EROD activity, as well as several biometrical parameters (age, length, wet weight, condition factor). None of the biochemical markers correlated with any biometrical parameter, thus biomarker reactions were related to site-specific criteria. While the hepatic hsp70 level did not differ among the sites, significant elevation of the hsp70 level in the gills revealed proteotoxic damage in chub at the most upstream site, where we recorded the highest heavy metal contamination of the investigated stretch, and in both chub and sneep at the site right downstream of the city of Arad. In both species, significantly elevated hepatic EROD activity downstream of Arad indicated that fish from these sites are also exposed to organic chemicals. The results were indicative of impaired fish health at least at three of the four investigated sites. The approach to relate biomarker responses to analytical data on pollution was shown to fit well the recent EU demands on further enhanced efforts in the monitoring of Romanian water quality.

Persistent organic pollutants, d13C and d15N in zooplankton, fish and bird species in Kongsfjorden according to season

Seasonality in biomagnification of persistent organic pollutants (POPs; polychlorinated biphenyls, chlorinated pesticides, and brominated flame retardants) in Arctic marine pelagic food webs was investigated in Kongsfjorden, Svalbard, Norway. Trophic magnification factors (TMFs; average factor change in concentration between two trophic levels) were used to measure food web biomagnification in biota in May, July, and October 2007. Pelagic zooplankton (seven species), fish (five species), and seabirds (two species) were included in the study. For most POP compounds, highest TMFs were found in July and lowest were in May. Seasonally changing TMFs were a result of seasonally changing POP concentrations and the d15N-derived trophic positions of the species included in the food web. These seasonal differences in TMFs were independent of inclusion/exclusion of organisms based on physiology (i.e., warm- versus cold-blooded organisms) in the food web. The higher TMFs in July, when the food web consisted of a higher degree of boreal species, suggest that future warming of the Arctic and increased invasion by boreal species can result in increased food web magnification. Knowledge of the seasonal variation in POP biomagnification is a prerequisite for understanding changes in POP biomagnification caused by climate change.