Abstract
Recent studies have shown that the early detection of neurodegenerative diseases (such as Parkinson) can significantly improve the effectiveness of treatments that increase quality of life, reducing the costs associated with the disease. In this paper, the proposed methodology consists in detecting early signs of Parkinson's disease through speech, with the presence of background noise. The approach uses machine learning algorithms and signal processing techniques to correctly distinguish between healthy controls and Parkinson's disease patients. In order to detect early signs of the disease, a database with patients at different stages of the Parkinson's disease is used. The learning algorithms were optimized for generalization and accuracy. An analysis of the results obtained from the proposed methodology show potential uses of machine learning algorithms in biomedical applications to detect early signs of Parkinson's disease.

Abstract

Abstract
This paper draws on an innovative methodological framework to assess the energy performance of a set of urban development alternatives, using the city of Porto (Portugal) as a case study. The methodology combines the advantages of a spatially-explicit analysis with the prediction accuracy of neural networks to estimate the energy demand (for space heating, space cooling and mobility) resulting from the physical configuration of urban areas. The urban alternatives under assessment reflect a number of development strategies taking place in different locations within the city. These correspond to well-known urban development approaches (infill development, consolidated development, modern development, multi-family housing, transit-oriented development and green infrastructure). The results for the city of Porto show that the transit-oriented development, the urban infill and the consolidated development are the urban alternatives yielding the most relevant energy savings, especially regarding mobility needs. This study makes evident that planning for more efficient urban forms potentially brings about more efficient urban settings and reinforces the relevance of ex ante appraisals of urban projects and plans.

Abstract
In this paper, the unrelated parallel machine scheduling problem considering machine-dependent and job sequence-dependent setup times is addressed. This problem involves the scheduling of n jobs on m unrelated machines with setup times in order to minimize the makespan. The Simulated Annealing algorithm is used to solve four sets of small scheduling problems, from the literature, on two unrelated machines: the first one has six jobs, the second has seven jobs and the third and fourth has eight and nine jobs, respectively. The results seem promising when compared with other methods referred in literature.

Abstract
This paper investigates the coordination problem of overcurrent relays (OCRs) in presence of wind power generation and electrical energy storage (EES) systems. As the injected short-circuit current of inverter-based devices connected to the electrical grid is a function of the power electronic withstand capacity, the short-circuit level would be limited for these types of devices. Furthermore, since the short-circuit current is a function of the pre-fault current, it is highly needed to take different conditions into account to accurately evaluate the injected current by such devices. This would mainly matter for the EES system operating in either charging or discharging modes, as well. This paper evaluates different operation strategies considering the variations of the load demand and the presence of large-scale wind farms as well as an EES system, while validating the suggested method for coordinating the directional OCRs (DOCRs). © 2018 IEEE.

Abstract
Expression Atlas (http://www.ebi.ac.uk/gxa) is an added value database that provides information about gene and protein expression in different species and contexts, such as tissue, developmental stage, disease or cell type. The available public and controlled access data sets from different sources are curated and re-analysed using standardized, open source pipelines and made available for queries, download and visualization. As of August 2017, Expression Atlas holds data from 3,126 studies across 33 different species, including 731 from plants. Data from large-scale RNA sequencing studies including Blueprint, PCAWG, ENCODE, GTEx and HipSci can be visualized next to each other. In Expression Atlas, users can query genes or gene-sets of interest and explore their expression across or within species, tissues, developmental stages in a constitutive or differential context, representing the effects of diseases, conditions or experimental interventions. All processed data matrices are available for direct download in tab-delimited format or as R-data. In addition to the web interface, data sets can now be searched and downloaded through the Expression Atlas R package. Novel features and visualizations include the on-the-fly analysis of gene set overlaps and the option to view gene co-expression in experiments investigating constitutive gene expression across tissues or other conditions.