Using data mining for prediction of hospital length of stay: an application of the CRISP-DM Methodology

Hospitals are nowadays collecting vast amounts of data related with patient records. All this data hold valuable knowledge that can be used to improve hospital decision making. Data mining techniques aim precisely at the extraction of useful knowledge from raw data. This work describes an implementation of a medical data mining project approach based on the CRISP-DM methodology. Recent real-world data, from 2000 to 2013, were collected from a Portuguese hospital and related with inpatient hospitalization. The goal was to predict generic hospital Length Of Stay based on indicators that are commonly available at the hospitalization process (e.g., gender, age, episode type, medical specialty). At the data preparation stage, the data were cleaned and variables were selected and transformed, leading to 14 inputs. Next, at the modeling stage, a regression approach was adopted, where six learning methods were compared: Average Prediction, Multiple Regression, Decision Tree, Artificial Neural Network ensemble, Support Vector Machine and Random Forest. The best learning model was obtained by the Random Forest method, which presents a high quality coefficient of determination value (0.81). This model was then opened by using a sensitivity analysis procedure that revealed three influential input attributes: the hospital episode type, the physical service where the patient is hospitalized and the associated medical specialty. Such extracted knowledge confirmed that the obtained predictive model is credible and with potential value for supporting decisions of hospital managers.

Tree species diversity in a seasonally-dry forest: the case of the Pinkaití site, in the Kayapó Indigenous Area, Southeastern limits of the Amazon

This study investigates patterns of forest structure and tree species diversity in an anthropogenic palm grove and undisturbed areas at the seasonally-dry Pinkaití research station, in the Kayapó Indigenous Area. This site, managed by the Conservation International do Brasil, is the most southeastern site floristically surveyed in the Amazon until now. The secondary and a nearby undisturbed forest were sampled in a group of 52 floristic plots of 0.0625-ha (25x25-m) where all trees with DBH > 10 cm were measured and identified. The analyses were complemented with other two floristic plots of 1-ha (10x1000-m). The present study has shown that the Pinkaití, like other seasonally-dry forests, have great heterogeneity in forest structure and composition, associated with biotic characteristics of the most important tree species, natural disturbance and history of land-use. The palm grove, moderately dominated by the arborescent palm Attalea maripa (Aubl.) Mart., presented high tree species diversity and was floristically similar to undisturbed forests at the study site. It is discussed the importance of large arborescent palms for the seasonally-dry Amazon forests regeneration.

Abundance of two Dendrocincla woodcreepers (aves: Dendrocolaptidae) in relation to forest structure in Central Amazonia

Few studies have been conducted to verify how the structure of the forest affects the occurence and abundance of neotropical birds. Our research was undertaken between January 2002 and July 2004 at the Reserva Ducke, near Manaus (02º55',03º01'S; 59º53',59º59'W) in central Amazonia, to verify how the forest structure affects the occurrence and abundance of two bird species: the Plain-brown Woodcreeper Dendrocincla fuliginosa and the White-chinned Woodcreeper Dendrocincla merula. Bird species occurrence was recorded using lines of 20 mist-nets (one sample unit), along 51 1-km transects distributed along 9 pararel 8 km trails covering an area of 6400 ha. Along these transects, we placed 50 x 50m plots where we recorded forest structure components (tree abundance, canopy openness, leaf litter, standing dead trees, logs, proximity to streams, and altitude). We then related these variables to bird occurence and abundance using multiple logistic and multiple linear regression models, respectively. We found that D. fuliginosa frequently used plateau areas; being more abundant in areas with more trees. On the other hand, D. merula occurred more frequently and was more abundant in areas with low tree abundance. Our results suggest that although both species overlap in the reserve (both were recorded in at least 68% of the sampled sites), they differ in the way they use the forest microhabitats. Therefore, local variation in the forest structure may contribute to the coexistence of congeneric species and may help to maintain local alpha diversity.

Extreme value Birnbaum-Saunders regression models applied to environmental data

Extreme value models are widely used in different areas. The Birnbaum–Saunders distribution is receiving considerable attention due to its physical arguments and its good properties. We propose a methodology based on extreme value Birnbaum–Saunders regression models, which includes model formulation, estimation, inference and checking. We further conduct a simulation study for evaluating its performance. A statistical analysis with real-world extreme value environmental data using the methodology is provided as illustration.

Outcrossing rate of Couratari multiflora (J.Smith) Eyma (Lecythidaceae), a low-density tropical tree species from a central Amazonian rainforest

A multilocus mixed-mating model was used to evaluate the mating system of a population of Couratari multiflora, an emergent tree species found in low densities (1 individual/10 ha) in lowland forests of central Amazonia. We surveyed and observed phenologically 41 trees in an area of 400 ha. From these, only four mother trees were analyzed here because few of them set fruits, which also suffered high predation. No difference was observed between the population multilocus outcrossing rate (t mp = 0.953 ± 0.040) and the average single locus rate (t sp = 0.968 ± 0.132). The four mother trees were highly outcrossed (t m ~ 1). Two out of five loci showed departures from the Hardy-Weinberg Equilibrium (HWE) expectations, and the same results occurred with the mixed-mating model. Besides the low number of trees analyzed, the proportion of loci in HWE suggests random mating in the population. However, the pollen pool was heterogeneous among families, probably due to both the small sample number and the flowering of trees at different times of the flowering season. Reproductive phenology of the population and the results presented here suggest, at least for part of the population, a long-distance pollen movement, around 1,000 m.

Density, size and distribution of stomata in 35 rainforest tree species in Central Amazonia

Stomata are turgor-operated valves that control water loss and CO2 uptake during photosynthesis, and thereby water relation and plant biomass accumulation is closely related to stomatal functioning. The aims of this work were to document how stomata are distributed on the leaf surface and to determine if there is any significant variation in stomatal characteristics among Amazonian tree species, and finally to study the relationship between stomatal density (S D) and tree height. Thirty five trees (>17 m tall) of different species were selected. Stomatal type, density (S D), size (S S) and stomatal distribution on the leaf surface were determined using nail polish imprints taken from both leaf surfaces. Irrespective of tree species, stomata were located only on the abaxial surface (hypostomaty), with large variation in both S D and S S among species. S D ranged from 110 mm-2 in Neea altissima to 846 mm-2 in Qualea acuminata. However, in most species S D ranges between 271 and 543 mm-2, with a negative relationship between S D and S S. We also found a positive relationship between S D and tree height (r² = 0.14, p < 0.01), but no correlation was found between S D and leaf thickness. The most common stomatal type was anomocytic (37%), followed by paracytic (26%) and anisocytic (11%). We conclude that in Amazonian tree species, stomatal distribution on the leaf surface is a response most likely dependent on the genetic background of every species, rather than a reaction to environmental changes, and that somehow S D is influenced by environmental factors dependent on tree height.

Tree communities of white-sand and terra-firme forests of the upper Rio Negro

The high tree diversity and vast extent of Amazonian forests challenge our understanding of how tree species abundance and composition varies across this region. Information about these parameters, usually obtained from tree inventories plots, is essential for revealing patterns of tree diversity. Numerous tree inventories plots have been established in Amazonia, yet, tree species composition and diversity of white-sand and terra-firme forests of the upper Rio Negro still remain poorly understood. Here, we present data from eight new one-hectare tree inventories plots established in the upper Rio Negro; four of which were located in white-sand forests and four in terra-firme forests. Overall, we registered 4703 trees > 10 cm of diameter at breast height. These trees belong to 49 families, 215 genera, and 603 species. We found that tree communities of terra-firme and white-sand forests in the upper Rio Negro significantly differ from each other in their species composition. Tree communities of white-sand forests show a higher floristic similarity and lower diversity than those of terra-firme forests. We argue that mechanisms driving differences between tree communities of white-sand and terra-firme forests are related to habitat size, which ultimately influences large-scale and long-term evolutionary processes.

Biological and morphometric aspects and rearing of Syssphinx molina (Cramer) (Lepidoptera: Saturniidae), a defoliator of parica tree

The objective of this research was to describe the biological and morphometric aspects of the parica tree defoliator, Syssphinx molina (Cramer), and make recommendations about the insect rearing. The life cycle was 62.9 days with mean periods for the egg, larval, pre-pupal and pupal stages of 5.6, 31.1, 2.2 and 16.6 days respectively. The pupal viability was 60.5% for females and 48.6% for males. The sexual ratio was 0.5 with mean production of 182.3 ± 2.2 eggs per female and egg viability of 24.3%. The mean longevity was 7.9 ± 2 and 8.1 ± 3 days for females and males respectively. Other parameters were also observed and compared with description of other Saturniidae species.

Updated List and identification key of 219 tree species of white sand forests of the Allpahuayo Mishana National Reserve, Loreto, Peru

White sand forests, although low in nutrients, are characterized not only by several endemic species of plants but also by several monodominant species. In general, plants in this forest have noticeably thin stems. The aim of this work was to elaborate a parallel dichotomous key for the identification of Angiosperm tree species occurring on white sand forests at the Allpahuayo Mishana National Reserve, Loreto, Peru. We compiled a list of species from several publications in order to have the most comprehensive list of species that occur on white sand forest. We found 219 species of Angiosperm, the more abundant species were Pachira brevipes (26.27%), Caraipa utilis (17.90%), Dicymbe uaiparuensis (13.27%), Dendropanax umbellatus (3.28%), Sloanea spathulata (2.52%), Ternstroemia klugiana (2.30%), Haploclathra cordata (2.28%), Parkia igneiflora (1.20%), Emmotum floribundum (1.06%), Ravenia biramosa (1.04%) among others. Most species of white sand forests can be distinguished using characteristics of stems, branches and leaves. This key is very useful for the development of floristic inventories and related projects on white sand forests from Allpahuayo Mishana National Reserve.

Predicting the distribution of forest tree species using topographic variables and vegetation index in eastern Acre, Brazil

Species distribution modeling has relevant implications for the studies of biodiversity, decision making about conservation and knowledge about ecological requirements of the species. The aim of this study was to evaluate if the use of forest inventories can improve the estimation of occurrence probability, identify the limits of the potential distribution and habitat preference of a group of timber tree species. The environmental predictor variables were: elevation, slope, aspect, normalized difference vegetation index (NDVI) and height above the nearest drainage (HAND). To estimate the distribution of species we used the maximum entropy method (Maxent). In comparison with a random distribution, using topographic variables and vegetation index as features, the Maxent method predicted with an average accuracy of 86% the geographical distribution of studied species. The altitude and NDVI were the most important variables. There were limitations to the interpolation of the models for non-sampled locations and that are outside of the elevation gradient associated with the occurrence data in approximately 7% of the basin area. Ceiba pentandra (samaúma), Castilla ulei (caucho) and Hura crepitans (assacu) is more likely to occur in nearby water course areas. Clarisia racemosa (guariúba), Amburana acreana (cerejeira), Aspidosperma macrocarpon (pereiro), Apuleia leiocarpa (cumaru cetim), Aspidosperma parvifolium (amarelão) and Astronium lecointei (aroeira) can also occur in upland forest and well drained soils. This modeling approach has potential for application on other tropical species still less studied, especially those that are under pressure from logging.

Fine time scaling of purifying selection on human nonsynonymous mtDNA mutations based on the worldwide population tree and mother-child pairs

A high-resolution mtDNA phylogenetic tree allowed us to look backward in time to investigate purifying selection. Purifying selection was very strong in the last 2,500 years, continuously eliminating pathogenic mutations back until the end of the Younger Dryas (∼11,000 years ago), when a large population expansion likely relaxed selection pressure. This was preceded by a phase of stable selection until another relaxation occurred in the out-of-Africa migration. Demography and selection are closely related: expansions led to relaxation of selection and higher pathogenicity mutations significantly decreased the growth of descendants. The only detectible positive selection was the recurrence of highly pathogenic nonsynonymous mutations (m.3394T>C-m.3397A>G-m.3398T>C) at interior branches of the tree, preventing the formation of a dinucleotide STR (TATATA) in the MT-ND1 gene. At the most recent time scale in 124 mother-children transmissions, purifying selection was detectable through the loss of mtDNA variants with high predicted pathogenicity. A few haplogroup-defining sites were also heteroplasmic, agreeing with a significant propensity in 349 positions in the phylogenetic tree to revert back to the ancestral variant. This nonrandom mutation property explains the observation of heteroplasmic mutations at some haplogroup-defining sites in sequencing datasets, which may not indicate poor quality as has been claimed.

Walking Tree - Organisation neuer Forschung in der Wissenschaft : Zusammenarbeit zwischen Integration und Differenzierung

Magdeburg, Univ., Fak. für Geistes-, Sozial- und Erziehungswiss., Diss., 2011

Association between LDL-C, Non HDL-C, and Apolipoprotein B Levels with Coronary Plaque Regression

Background:Previous reports have inferred a linear relationship between LDL-C and changes in coronary plaque volume (CPV) measured by intravascular ultrasound. However, these publications included a small number of studies and did not explore other lipid markers.Objective:To assess the association between changes in lipid markers and regression of CPV using published data.Methods:We collected data from the control, placebo and intervention arms in studies that compared the effect of lipidlowering treatments on CPV, and from the placebo and control arms in studies that tested drugs that did not affect lipids. Baseline and final measurements of plaque volume, expressed in mm3, were extracted and the percentage changes after the interventions were calculated. Performing three linear regression analyses, we assessed the relationship between percentage and absolute changes in lipid markers and percentage variations in CPV.Results:Twenty-seven studies were selected. Correlations between percentage changes in LDL-C, non-HDL-C, and apolipoprotein B (ApoB) and percentage changes in CPV were moderate (r = 0.48, r = 0.47, and r = 0.44, respectively). Correlations between absolute differences in LDL-C, non‑HDL-C, and ApoB with percentage differences in CPV were stronger (r = 0.57, r = 0.52, and r = 0.79). The linear regression model showed a statistically significant association between a reduction in lipid markers and regression of plaque volume.Conclusion:A significant association between changes in different atherogenic particles and regression of CPV was observed. The absolute reduction in ApoB showed the strongest correlation with coronary plaque regression.

Optimal designs for mixed effects poisson regression models

Magdeburg, Univ., Fak. für Mathematik, Diss., 2010

Optimal designs for the prediction in hierarchical random coefficient regression models

Magdeburg, Univ., Fak. für Mathematik, Diss., 2015