Automatic tuning of PID controllers using a neuro-genetic system

Neural networks and genetic algorithms have been in the past successfully applied, separately, to controller turning problems. In this paper we propose to combine its joint use, by exploiting the nonlinear mapping capabilites of neural networks to model objective functions, and to use them to supply their values to a genetic algorithm which performs on-line minimization.

O estudo das funções com as F-Tools: o conceito F-Tool

O presente relatório, orientado pela Doutora Ana Conceição, foi elaborado no âmbito da Prática de Ensino Supervisionada (PES), integrada no plano de estudos do Mestrado em Ensino de Matemática no 3.º Ciclo do Ensino Básico e no Ensino Secundário, da Universidade do Algarve (UAlg). Apresento uma breve caraterização das escolas, dos orientadores cooperantes e das turmas. Menciono as aulas assistidas e as aulas lecionadas com a supervisão da orientadora e relato os trabalhos desenvolvidos em cada escola. Dedico um capítulo à importância das tecnologias no processo de ensino e aprendizagem pois, na minha opinião, na formação atual de qualquer professor deve ser tido em conta o uso de tecnologias em sala de aula. Partilhando a minha orientadora da mesma opinião, colocou-me um desafio: a criação de um software educacional com recurso ao sistema de álgebra computacional Mathematica, tendo em vista a melhoria do processo de ensino e aprendizagem da Matemática (em particular, no estudo das funções reais de variável real). Aceitei de imediato e, juntamente com a Dra. Cristina Simão e o Dr. José Pereira, iniciamos um projeto inovador na área de ensino, posteriormente distinguido com o prémio Timberlake1. Neste relatório apresento o conceito F-Tool2, isto é, ferramentas de ensino visuais, dinâmicas e interativas que permitem explorar de uma forma inovadora alguns dos principais conceitos nas áreas de Pré-Cálculo e Cálculo Diferencial, nos níveis básico, secundário e universitário. O conceito F-Tool foi utilizado nas aulas que lecionei em regime de par pedagógico com a Dra. Cristina Simão. Devido à sua importância na minha PES, ficou claro que o tema do meu relatório teria de ter uma referência explícita ao estudo das funções com as F-Tool. Reflito ainda sobre os seminários que apresentei na UAlg, no âmbito da PES, e apresento uma reflexão final de toda a Prática de Ensino Supervisionada.

Automatic processing of bidimensional images of ''omic'' expression blobs

In this work, a comprehensive review on automatic analysis of Proteomics and Genomics images is presented. Special emphasis is given to a particularly complex image produced by a technique called Two-Dimensional Gel Electrophoresis (2-DE), with thousands of spots (or blobs). Automatic methods for the detection, segmentation and matching of blob like features are discussed and proposed. In particular, a very robust procedure was achieved for processing 2-DE images, consisting mainly of two steps: a) A very trustworthy new approach for the automatic detection and segmentation of spots, based on the Watershed Transform, without any foreknowledge of spot shape or size, and without user intervention; b) A new method for spot matching, based on image registration, that performs well for either global or local distortions. The results of the proposed methods are compared to state-of-the-art academic and commercial products.

Application of molecular tools for detection of plant viruses

Grapevine leafroll disease (GLRD) is one of the most important virus diseases of grapevines worldwide, causing major economical impact. The disease has a complex aetiology and currently eleven phloem-limited viruses, termed in general Grapevine leafroll-associated virus (GLRaVs), have been identified. Two of the GLRaVs, GLRaV-1 and GLRaV-3, are included in the European certification scheme of propagation material. However, the flawed notion that GLRaV-3 is more frequent than GLRaV-1 and that all other GLRaVs are possibly not as relevant for GLRD, has until now precluded the development of specific serological and molecular detection assays and limited the scope of molecular characterization of the viruses known to be associated with the disease. Hence, few studies have addressed the phylodynamics of GLRaVs or even characterized the genetic structure of their natural populations. This generalized lack of molecular information, in turn underlie the deficient capacity to detect the viruses. The phylogenetic analyses were conducted on the basis of the heat shock protein 70 homologue (HSP70h) and the coat protein (CP) genes for GLRaV-1 and the HSP70h, the heat shock protein 90 homologue (HSP90h) and the CP genes for GLRaV-5. The data obtained for GLRaV-1 contributed 83 new CP sequences. This information was combined with previous analysis by other authors and used for the production of new polyclonal IgG, capable of detecting CP variants from all the phylogroups observed. Successful testing of this new tool included tissue print immunoblotting (TPIB) and in situ immunoassay (ISIA). The data obtained for GLRaV-5, contributed 61 new CP and 28 new HSP90h gene sequences. Eight phylogenetic groups were identified on the basis of the CP. Characterization of the genetic structure of the isolates revealed a higher diversity than previously reported and allowed the identification of dominant virus variants. For both GLRaV-1 and GLRaV-5, the effect of vegetative propagation on the virus transmission dynamics was addressed.

Liver-like biosensor: immobilized CYP3A4 sensors as tools for ligand binding and enzyme activity measurements

Cytochromes P450 constitute a super-family of enzymes involved in the metabolism of Xenobiotics, where human cytochrome P450 3A4 is the most abundant of all P450s, accounting for about 50% of all human liver cytochromes. This membrane anchored protein is responsible for the metabolization of a wide array of environmental drugs and intoxicants, mainly due to its haem domain properties, and active site cavity volume. These properties make this protein an excellent subject for biosensor application, although CYO3A4 enzyme is also famous for its instability. Enzyme inactivation at room temperature is a normal conversion process that this enzyme undergoes, that may hamper any biosensing approach.

Automatic classification of scientific records using the German Subject Heading Authority File (SWD)

The following paper deals with an automatic text classification method which does not require training documents. For this method the German Subject Heading Authority File (SWD), provided by the linked data service of the German National Library is used. Recently the SWD was enriched with notations of the Dewey Decimal Classification (DDC). In consequence it became possible to utilize the subject headings as textual representations for the notations of the DDC. Basically, we we derive the classification of a text from the classification of the words in the text given by the thesaurus. The method was tested by classifying 3826 OAI-Records from 7 different repositories. Mean reciprocal rank and recall were chosen as evaluation measure. Direct comparison to a machine learning method has shown that this method is definitely competitive. Thus we can conclude that the enriched version of the SWD provides high quality information with a broad coverage for classification of German scientific articles.

Automatic extraction of definitions

Tese de doutoramento, Informática (Engenharia Informática), Universidade de Lisboa, Faculdade de Ciências, 2014

New tools for chemical modification of proteins

Tese de doutoramento, Farmácia (Química Farmacêutica e Terapêutica), Universidade de Lisboa, Faculdade de Farmácia, 2014

Effectiveness of resilience assessment tools in different case studies

Tese de doutoramento, Biologia (Biologia Marinha e Aquacultura), Universidade de Lisboa, Faculdade de Ciências, 2015

Enhancing extractive summarization with automatic post-processing

Tese de doutoramento, Informática (Ciência da Computação), Universidade de Lisboa, Faculdade de Ciências, 2015

Using Workload Measurement Tools in Diverse Care Contexts: the Experience of Staff in Mental Health and Learning Disability Inpatient Settings

Introduction The critical challenge of determining the correct level and skill-mix of nursing staff required to deliver safe and effective healthcare has become an international concern. It is recommended that evidence-based staffing decisions are central to the development of future workforce plans. Workforce planning in mental health and learning disability nursing is largely under-researched with few tools available to aid the development of evidence-based staffing levels in these environments. Aim It was the aim of this study to explore the experience of staff using the Safer Nursing Care Tool (SNCT) and the Mental Health and Learning Disability Workload Tool (MHLDWT) in mental health and learning disability environments. Method Following a 4-week trial period of both tools a survey was distributed via Qualtrics on-line survey software to staff members who used the tools during this time. Results The results of the survey revealed that the tools were considered a useful resource to aid staffing decisions; however specific criticisms were highlighted regarding their suitability to psychiatric intensive care units (PICU) and learning disability wards. Discussion This study highlights that further development of workload measurement tools is required to support the implementation of effective workforce planning strategies within mental health and learning disability services. Implications for Practice With increasing fiscal pressures the need to provide cost-effective care is paramount within NHS services. Evidence-based workforce planning is therefore necessary to ensure that appropriate levels of staff are determined. This is of particular importance within mental health and learning disability services due to the reduction in the number of available beds and an increasing focus on purposeful admission and discharge.

The Development of Environmental Assessment Tools to Support the Creation of Dementia Friendly Care Environments: Innovative Practice

The need for more dementia friendly design in hospitals and other care settings is now widely acknowledged. Working with 26 NHS Trusts in England as part of a Department of Health commissioned programme, The King’s Fund developed a set of overarching design principles and an environmental assessment tool for hospital wards in 2012. Following requests from other sectors, additional tools were developed for hospitals, care homes, health centres and housing with care. The tools have proven to be effective in both disseminating the principles of dementia friendly design and in enabling the case to be made for improvements that have a positive effect on patient outcomes and staff morale. This paper reports on the development,use and review of the environmental assessment tools, including further work that is now being taken forward by The Association for Dementia Studies, University of Worcester.

Automatic Video Analysis for Fisheries Survey Systems

Thesis (Ph.D.)--University of Washington, 2015