Abstract

Abstract
Objective: Electronic health records (EHR) offer medical and pharmacogenomics research unprecedented opportunities to identify and classify patients at risk. EHRs are collections of highly inter-dependent records that include biological, anatomical, physiological, and behavioral observations. They comprise a patient's clinical phenome, where each patient has thousands of date-stamped records distributed across many relational tables. Development of EHR computer-based phenotyping algorithms require time and medical insight from clinical experts, who most often can only review a small patient subset representative of the total EHR records, to identify phenotype features. In this research we evaluate whether relational machine learning (ML) using inductive logic programming (ILP) can contribute to addressing these issues as a viable approach for EHR-based phenotyping. Methods: Two relational learning ILP approaches and three well-known WEKA (Waikato Environment for Knowledge Analysis) implementations of non-relational approaches (PART, J48, and JRIP) were used to develop models for nine phenotypes. International Classification of Diseases, Ninth Revision (ICD-9) coded EHR data were used to select training cohorts for the development of each phenotypic model. Accuracy, precision, recall, F-Measure, and Area Under the Receiver Operating Characteristic (AUROC) curve statistics were measured for each phenotypic model based on independent manually verified test cohorts. A two-sided binomial distribution test (sign test) compared the five ML approaches across phenotypes for statistical significance. Results: We developed an approach to automatically label training examples using ICD-9 diagnosis codes for the ML approaches being evaluated. Nine phenotypic models for each ML approach were evaluated, resulting in better overall model performance in AUROC using ILP when compared to PART (p = 0.039), J48 (p = 0.003) and JRIP (p = 0.003). Discussion: ILP has the potential to improve phenotyping by independently delivering clinically expert interpretable rules for phenotype definitions, or intuitive phenotypes to assist experts. Conclusion: Relational learning using ILP offers a viable approach to EHR-driven phenotyping.

Abstract
A successful deployment of electrical energy storage (EES) in current electricity grid systems is a plausible episode in several power systems given the outstanding technical, economic and environmental benefits that EES provides. Also, more distributed resources are becoming key actors in remotely located renewable energy based and poorly interconnected island grid systems. Healing these fragile grid systems in islands requires devising and promoting a robust grid system operation. Such move should go hand in hand with an increasing distributed energy resource use to guarantee sustainable renewable energy integration into these systems. In the present document, EES technologies and applications have been compiled, where special emphasis has been given on islands, studying their particular requirements and technology appropriateness on operating project experiences around the world, from which some lessons can be learned. Conclusions about EES technologies' general suitability on island systems are duly drawn throughout the content of this paper.

Abstract
Mechanical maintenance of F-16 engines is carried out as a team effort involving 3-4 skilled engine technicians, but the details of its procedures and requisites change constantly, to improve safety, optimize resources, and respond to knowledge learned from field outcomes. This provides a challenge for development of training simulators, since simulated actions risk becoming obsolete rapidly and require costly reimplementation. This paper presents the development of a 3D mechanical maintenance training simulator for this context, using a low-cost simulation platform and a software architecture that separates simulation control from simulation visualization, in view of enabling more agile adaptation of simulators. This specific simulator aims to enable technician training to be enhanced with cooperation and context prior to the training phase with actual physical engines. We provide data in support of the feasibility of this approach, describing the requirements that were identified with the Portuguese Air Force, the overall software architecture of the system, the current stage of the prototype, and the outcomes of the first field tests with users.

Abstract
This work presents a fiber CRD configuration for the measurement of strain. An Optical Time-Domain Reflectometer was used to send impulses down into the fiber loop cavity, inside of which a chirped fiber Bragg grating was placed to act as a strain sensing element. This technique could provide strain results with both conventional CRD-based configuration and the OTDR.

Abstract
With advances in communication technology, future internet presents numerous opportunities to develop new systems designed to make day to day life easier and to enhance and prolong the life of people with disabilities. This motivation propels the development of new services that integrate the mobility of cloud systems and the diversity of IoT (Internet of Things). It will enable us to create new and more independent care systems for people with disabilities, enabling a certain degree of independence. This can have a psychological and social impact due to the better quality of life that enables. Other motivation is the versatility and mobility of services it can provide, making those services available. In this paper is explored and explained the different kinds of technologies that can be integrated to enable creation of future internet platforms. Also, an IoT Cloud platform will be analyzed and some tests will be made, ending with some conclusions and lessons learned in this work.