Repositório IC-Online
Repositório Científico da Instituição
Entradas recentes
Cut-off for the legal ages in the Portuguese Population by Third Maturity Index: Measures of Accuracy
Publication . Pereira, Cristiana Palmela; Rodrigues, Ana; Santos, Adriana; Salvado, Francisco; Santos, Rui; Cameriere, Roberto
Objectives: The aim of this study is to estimate the cut-off points for the Portuguese legal ages through the I3M, and to compare them with the methods of Demirjian, Nolla and Moorrees. Design: The lower third molars were analyzed on 348 orthopantomography's aged between 12 and 23 years in a Portuguese Population. The images were analyzed by ImageJ and the cut-off points were calculated for the respective legal ages of 14, 16, 18 and 21 years old. Results: The correlation between age and the I3M was 0.862, whereas with Demirjian's stadiums, Nolla's stages, and Moorrees’ stages the correlation coefficients were 0.863, 0.842 and 0.844, respectively. For the cut-off point of 0.08 for the age of 18, a sensitivity of 78.99 %, specificity of 93.48 %, an accuracy of 88.54 %, a positive predictive value of 86.24 % and a posteriori Bayes probability of 92.82 %. The cut-off points established for the ages of 14, 16, 18 and 21 years for the Portuguese population achieved an accuracy of 83.67 %, 85.67 %, 88.54 % and 87.11 %, respectively. Conclusion: Demirjian's method and Cameriere's method attain quite similar results. In certain ages, the methods of Nolla and Moorrees show higher sensitivity or higher specificity when compared to Cameriere's method, however this last method is more stable in terms of reliability and more suitable to use in the Portuguese population.
eCrash: An Empirical Study on the Apache Ant Project
Publication . Nogueira, Ana Filipa; Ribeiro, José Carlos Bregieiro; de Vega, Francisco Fernández; Zenha-Rela, Mário Alberto
The eCrash tool employs Strongly-Typed Genetic Programming to automate the generation of test data for the structural unit testing of Object-Oriented Java programs. This paper depicts the results attained by utilising eCrash to generate test data for the classes of the Apache Ant project.
Sustainable energy for plastic industry plant
Publication . Galvão, João; Moreira, Licínio; Leitão, Sérgio; Silva, Élson; Neto, Mário
This research intends to reduce energy consumption in a plastic film industrial unit. Trough the development of a new energy model focused on self renewable energy production and energy management in some sectors, such as lighting and compressed air. It is expected that this model leads to energy savings with environmental advantages (lower greenhouse gases emission). This may grant access to eco-sustainable certifications/awards as a factor of competitiveness, as well as economic benefits. The methodology followed to design the new energy model was developed from an energy audit to characterize the existing consumptions and use of energy.
Optimization of Thermoplastic Pre-Pregs Overmoulding
Publication . Carreira, Pedro; Alves, Nuno; Carina Ramos; Bártolo, Paulo J.
The search for new technological concepts in the field of injection of thermoplastic polymers, in order to optimize the process and reduce cycle time, faced us with a new concept in moulding, where two different technologies are proposed to be unified. The fabric impregnated with thermoplastic is an example of a quite new product that combines the potential of polymers with long reinforcing fibres. In order to process this new product, injection of polymers and thermoforming will be applied in only one operation. To allow cycle time prediction for this new technique, changes will be made in the formulation of conventional injection cycle time, in order to comprise the new necessary stages. In this work a new approach is proposed to obtain moulded parts of thermoplastic polymers with functional fabric only in a single processing cycle. The description of the new stages and its mathematical formulation is made, in function of the time needed to complete each one. The results presented come from processing parameters optimization using the Particle Swarm Optimization (PSO) algorithm.
An interactive Web-GIS fluvial flood forecast and alert system in operation in Portugal
Publication . Mourato, Sandra; Fernandez, Paulo; Marques, Fábio; Rocha, Alfredo; Pereira, Luísa
Floods are one of the natural disasters not preventable, affecting people and causing significant damage to economic activities and infrastructures. Thus, it is of foremost importance to, within a disaster risk-reduction strategy, develop a useful flood forecast and alert system to prevent people from suffering flood disasters and mitigate its consequences. This article presents the Flood Forecast and Alert System in operational mode since 2019 for the Águeda river basin located in Portugal's centre region. This system is technologically advanced, differing from others since it uses a coupled real-time hydrologic and 2D hydrodynamic modelling supported on numerical weather prediction and a high-resolution digital terrain surface model. The system components are automatically activated and linked: i) a rainfall forecasting model (WRF), ii) a hydrological model (HEC-HMS), iii) a hydraulic model (HEC-RAS 2D), and a iv) Web-GIS platform. The hydrological model is forced with forecast precipitation for the next three days and updated every 6 h, which is crucial to generate pre-flood hazard maps. It also includes a Web GIS service for flood hazard dissemination available for civil authorities and citizens. A flood forecast and alert system is highly relevant to the community since, by enhancing knowledge, it provides the authorities responsible for assessing and managing the flood risk, responsiveness to disasters and timely decision-making, which is even more evident in the context of climate change.