Development of an intelligent earthwork optimization system

Tese de Doutoramento em Engenharia Civil.

Plataforma de apoio à decisão nos cuidados de ginecologia e obstetrícia

Dissertação de mestrado integrado em Engenharia Biomédica (área de especialização em Informática Médica)

Modelos de regressão multinível no estudo do desempenho escolar

Dissertação de mestrado em Estatística

Pervasive patient timeline for intensive care units

This research work explores a new way of presenting and representing information about patients in critical care, which is the use of a timeline to display information. This is accomplished with the development of an interactive Pervasive Patient Timeline able to give to the intensivists an access in real-time to an environment containing patients clinical information from the moment in which the patients are admitted in the Intensive Care Unit (ICU) until their discharge This solution allows the intensivists to analyse data regarding vital signs, medication, exams, data mining predictions, among others. Due to the pervasive features, intensivists can have access to the timeline anywhere and anytime, allowing them to make decisions when they need to be made. This platform is patient-centred and is prepared to support the decision process allowing the intensivists to provide better care to patients due the inclusion of clinical forecasts.

A clinical recommendation system to maternity care

Nowadays in healthcare, the Clinical Decision Support Systems are used in order to help health professionals to take an evidence-based decision. An example is the Clinical Recommendation Systems. In this sense, it was developed and implemented in Centro Hospitalar do Porto a pre-triage system in order to group the patients on two levels (urgent or outpatient). However, although this system is calibrated and specific to the urgency of obstetrics and gynaecology, it does not meet all clinical requirements by the general department of the Portuguese HealthCare (Direção Geral de Saúde). The main requirement is the need of having priority triage system characterized by five levels. Thus some studies have been conducted with the aim of presenting a methodology able to evolve the pre-triage system on a Clinical Recommendation System with five levels. After some tests (using data mining and simulation techniques), it has been validated the possibility of transformation the pre-triage system in a Clinical Recommendation System in the obstetric context. This paper presents an overview of the Clinical Recommendation System for obstetric triage, the model developed and the main results achieved.

Real-Time models to predict the use of vasopressors in monitored patients

The needs of reducing human error has been growing in every field of study, and medicine is one of those. Through the implementation of technologies is possible to help in the decision making process of clinics, therefore to reduce the difficulties that are typically faced. This study focuses on easing some of those difficulties by presenting real-time data mining models capable of predicting if a monitored patient, typically admitted in intensive care, will need to take vasopressors. Data Mining models were induced using clinical variables such as vital signs, laboratory analysis, among others. The best model presented a sensitivity of 94.94%. With this model it is possible reducing the misuse of vasopressors acting as prevention. At same time it is offered a better care to patients by anticipating their treatment with vasopressors.

Anfragetechniken für heterogene Datenbanksysteme

Magdeburg, Univ., Fak. für Informatik, Habil.-Schr., 2003

Partially supervised learning of fuzzy classification rules

Fuzzy classification, semi-supervised learning, data mining

On propositionalization for knowledge discovery in relational databases

Propositionalization, Inductive Logic Programming, Multi-Relational Data Mining

Häufige Muster in zeitbezogenen Daten

Magdeburg, Univ., Fak. für Informatik, Diss., 2008

Label monitoring on document streams

Magdeburg, Univ., Fak. für Informatik, Diss., 2012

Discovery and visualization of interesting patterns

Die Preise für Speicherplatz fallen stetig, da verwundert es nicht, dass Unternehmen riesige Datenmengen anhäufen und sammeln. Diese immensen Datenmengen müssen jedoch mit geeigneten Methoden analysiert werden, um für das Unternehmen überlebensnotwendige Muster zu identifizieren. Solche Muster können Probleme aber auch Chancen darstellen. In jedem Fall ist es von größter Bedeutung, rechtzeitig diese Muster zu entdecken, um zeitnah reagieren zu können. Um breite Nutzerschichten anzusprechen, müssen Analysemethoden ferner einfach zu bedienen sein, sofort Rückmeldungen liefern und intuitive Visualisierungen anbieten. Ich schlage in der vorliegenden Arbeit Methoden zur Visualisierung und Filterung von Assoziationsregeln basierend auf ihren zeitlichen Änderungen vor. Ich werde lingustische Terme (die durch Fuzzymengen modelliert werden) verwenden, um die Historien von Regelbewertungsmaßen zu charakterisieren und so eine Ordnung von relevanten Regeln zu generieren. Weiterhin werde ich die vorgeschlagenen Methoden auf weitereModellarten übertragen, die Software-Plattformvorstellen, die die Analysemethoden dem Nutzer zugänglich macht und schließlich empirische Auswertungen auf Echtdaten aus Unternehmenskooperationen vorstellen, die die Wirksamkeit meiner Vorschläge belegen.

Proceedings of the 1st International Conference on Applied Innovations in IT

This volume contains publications of the 1st International Conference on Applied Innovations in IT (ICAIIT), which took place in Koethen March 25th 2013. The conference is devoted to problems of applied research in the fields of mechanical and economical engineering, auotmation and communications as well as of data mining. The research results can be of interest for researchers and development engineers, who deal with theoretical base and the application of the knowledge in the respective areas.

Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Bioinformática: consultas cruzadas a bases de datos biomédicas remotas

En la presente memoria se detallan con exactitud los pasos y procesos realizados para construir una aplicación que posibilite el cruce de datos genéticos a partir de información contenida en bases de datos remotas. Desarrolla un estudio en profundidad del contenido y estructura de las bases de datos remotas del NCBI y del KEGG, documentando una minería de datos con el objetivo de extraer de ellas la información necesaria para desarrollar la aplicación de cruce de datos genéticos. Finalmente se establecen los programas, scripts y entornos gráficos que han sido implementados para la construcción y posterior puesta en marcha de la aplicación que proporciona la funcionalidad de cruce de la que es objeto este proyecto fin de carrera.