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.

Secure multiParty computation protocols

Dissertação de Mestrado em Engenharia Informática

Search for the Standard Model Higgs boson produced in association with top quarks and decaying into bb in pp collisions at √s = 8 TeV with the ATLAS detector

A search for the Standard Model Higgs boson produced in association with a pair of top quarks, tt¯H, is presented. The analysis uses 20.3 fb−1 of pp collision data at s√ = 8 TeV, collected with the ATLAS detector at the Large Hadron Collider during 2012. The search is designed for the H to bb¯ decay mode and uses events containing one or two electrons or muons. In order to improve the sensitivity of the search, events are categorised according to their jet and b-tagged jet multiplicities. A neural network is used to discriminate between signal and background events, the latter being dominated by tt¯+jets production. In the single-lepton channel, variables calculated using a matrix element method are included as inputs to the neural network to improve discrimination of the irreducible tt¯+bb¯ background. No significant excess of events above the background expectation is found and an observed (expected) limit of 3.4 (2.2) times the Standard Model cross section is obtained at 95% confidence level. The ratio of the measured tt¯H signal cross section to the Standard Model expectation is found to be μ=1.5±1.1 assuming a Higgs boson mass of 125 GeV.

Redes de inovação em pequenas e médias empresas: um estudo de caso no setor têxtil

Dissertação de mestrado em Engenharia Industrial

Assessment of the mechanical performance of recycled unbound materials for pavement subbase layers

The cyclic load triaxial test is a laboratory test that allows studying the mechanical behaviour of unbound granular materials used in base/subbase layers of road pavements. The resilient modulus and permanent strains are required as inputs in structural pavement design. This paper presents some results obtained for recycled materials (crushed concrete aggregate and blended crushed waste aggregate), with a view to promoting their use in pavement structures. Results relating to a reference material (limestone) are also presented, for comparison. All the test results discussed in this paper were obtained in variable cyclic radial pressure (VCP) tests. The tests performed (VCP) aim to study the influence of water content on the resilient modulus of recycled materials, as well as on the resistance to permanent deformation. Using the experimental data as a basis, further modelling work was carried out to establish the stresses developing in base/capping layers in typical Belgian road pavements. These numerical results allow to propose some simplifications of the stress paths applied in the testing procedures and to establish a new test protocol that also considers compaction during construction works. The results of this research work provide an excellent set of findings for the mechanical characterization of unbound base materials through the cyclic triaxial test, and contribute to a better understanding and correct application of recycled materials under geotechnical engineering background

Cibercultura e Sub-culturas: Estética Digital

Introdução: informação, conteúdos, saberes: Os conteúdos digitais; Projeto 1: A vida quotidiana dos infonautas; Netiquette; Produção e consumo cultural na sociedade da informação; Cibercultura e vida académica; Media studies e new media studies; Diversas áreas de pesquisa; Proliferação de questões; Reflexividade dos especialistas do saber prático; A falência das antigas certezas. Estética digital: Sean Cubbit: Digital Aesthetics; Computação estética; Natureza da arte digital; Interfaces; o iPhone e iPad; o Windows 8; Design da interação digital; Luvas de realidade virtual; Inputs diretos da eletricidade cerebral; o Sistema de interação com múltiplas interfaces.

A integração de técnicas BIM (Building Information Modeling) no projeto e na construção de estruturas de madeira

Dissertação de mestrado integrado em Engenharia Civil

Seismic behavior of concrete block masonry buildings

Tese de Doutoramento em Engenharia Civil

Algorithms and computational tools for metabolic flux analysis

PhD thesis in Biomedical Engineering

A critical evaluation of methods for the reconstruction of tissue-specific models

Under the framework of constraint based modeling, genome-scale metabolic models (GSMMs) have been used for several tasks, such as metabolic engineering and phenotype prediction. More recently, their application in health related research has spanned drug discovery, biomarker identification and host-pathogen interactions, targeting diseases such as cancer, Alzheimer, obesity or diabetes. In the last years, the development of novel techniques for genome sequencing and other high-throughput methods, together with advances in Bioinformatics, allowed the reconstruction of GSMMs for human cells. Considering the diversity of cell types and tissues present in the human body, it is imperative to develop tissue-specific metabolic models. Methods to automatically generate these models, based on generic human metabolic models and a plethora of omics data, have been proposed. However, their results have not yet been adequately and critically evaluated and compared. This work presents a survey of the most important tissue or cell type specific metabolic model reconstruction methods, which use literature, transcriptomics, proteomics and metabolomics data, together with a global template model. As a case study, we analyzed the consistency between several omics data sources and reconstructed distinct metabolic models of hepatocytes using different methods and data sources as inputs. The results show that omics data sources have a poor overlapping and, in some cases, are even contradictory. Additionally, the hepatocyte metabolic models generated are in many cases not able to perform metabolic functions known to be present in the liver tissue. We conclude that reliable methods for a priori omics data integration are required to support the reconstruction of complex models of human cells.