Improvements on the KNN classifier

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Haptic robot-environment interaction for self-supervised learning in ground mobility

Dissertação para obtenção do Grau de Mestre em Engenharia Eletrotécnica e de Computadores

Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

Estratégia multi-temporal para produção automática de cartografia de ocupação do solo com imagens AWiFS

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Ciência e Sistemas de Informação Geográfica

Spatio-temporal land use/land cover changes analysis and monitoring in the Valencia Municipality, Spain

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies

Wetland Habitat Studies using various Classification Techniques on Multi-Spectral Landsat Imagery: Case study: Tram chim National Park, Dong Thap Vietnam

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies

Urban change monitoring using GIS and remote sensing tools in Kathmandu valley (Nepal)

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies

Spatial metrics and Landsat data for urban landuse change detection: case of Addis Ababa, Ethiopia.

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Vision based context categorization for all-terrain robot

Dissertation presented at the Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa to obtain the Master degree in Electrical and Computer Engineering.

Automated image tagging through tag propagation

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial Para obtenção do grau de Mestre em Engenharia Informática

Programa de exercício e promoção da actividade física após cirurgia por cancro da mama

RESUMO: Objectivo: O exercício tem sido identificado como um factor de promoção da qualidade de vida em mulheres submetidas a cirurgia por cancro de mama. Mas os níveis de actividade física tendem a reduzir após o diagnóstico de cancro de mama e a manterem-se baixos após fim dos tratamentos. O objectivo deste estudo é verificar se um programa de exercício físico supervisionado, associado a estratégias motivacionais em mulheres submetidas a cirurgia por cancro da mama e que mantêm a intervenção usual praticada, é mais efectivo no aumento dos níveis de actividade física, na redução do índice de massa corporal e na melhoria da qualidade de vida, do que o tratamento usualmente praticado. Metodologia: Trata-se de um estudo piloto experimental, aleatorizado e controlado. Cumpriram os critérios de inclusão no estudo 37 utentes submetidas a cirurgia por cancro de mama no Hospital Fernando Fonseca, Amadora, e submetidos à intervenção usual da fisioterapia. Foram aleatoriamente distribuídas pelos grupos experimental (n=19) e de controlo (n=18). O contacto telefónico foi feito por um elemento externo, cego em relação à distribuição dos sujeitos. Alguns elementos desistiram ou não puderam participar no estudo. O número de sujeitos final foi de 11 para o grupo experimental e de 10 para o grupo de controlo, idênticos na média de idades. Todos os sujeitos mantiveram a intervenção usual da fisioterapia. Os sujeitos do grupo experimental foram ainda submetidos ao programa de exercício e promoção da actividade física. Todo o grupo foi avaliado no início do programa, no final do mesmo (12 semanas) e após 6 meses de follow-up, nos seguintes indicadores: nível de actividade física, pelo International Physical Activity Questionnaire (IPAQ), índice de massa corporal e qualidade de vida, pelos questionários: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) e Supplementary Questionnaire Breast Cancer Module (QLQ-BR23). Resultados: O programa implementado demonstrou-se efectivo no aumento dos níveis de actividade física (p=0,017) e na qualidade de vida associada ao estado funcional (p=0,016) e ao estado de saúde global auto-percepcionado (p=0,030) no final do programa. Foi ainda possível identificar resultados positivos noutras variáveis estudadas, como por exemplo a fadiga auto-reportada, mas que não se demonstraram estatisticamente significativos, facto que pode atribuir-se à reduzida dimensão da amostra. Conclusões: Apesar de algumas limitações, este estudo é de grande a importância para a comunidade científica preocupada com crescente problemática do cancro de mama. Mais uma vez se demonstra que o exercício físico é uma estratégia importante para a melhoria do estado de saúde de pacientes confrontados com doença. Esta comprovação é de grande interesse para os profissionais de saúde e particularmente para os fisioterapeutas dedicados à promoção da saúde nesta população.---------------------ABSTRACT: Purpose: Exercise has been identified as a quality of life promoting factor, amongst women after breast cancer surgery. But the physical activity levels reduce significantly after a diagnosis of breast cancer and remain low after treatment is completed. The aim of this study is to determine whether a supervised group exercise program allied to motivating strategies in women after breast cancer surgery in conjunction with standard treatment, is more effective in improving physical activity levels, reducing body mass index and promoting quality of life, than standard treatment on its own. Methods: This is a pilot randomized controlled trial. Thirty-seven women, submitted to breast cancer surgery in the Hospital Fernando Fonseca, Amadora, and submitted to standard physiotherapy intervention, completed the inclusion criteria. They were randomly allocated into intervention group (n=19) and control group (n=18). The telephone contact was done by an external element, blind to the subjects’ allocation. Some elements declined or could not participate. The final intervention group was n=11 and the control n=10, similar in age. All subjects maintained the standard physiotherapy intervention. The experimental group was submitted to the exercise and health promotion program. The complete group was evaluated in the beginning of the program, at the final (after 12 weeks) and after 6 months follow-up, using as outcomes: physical activity index with the international Physical Activity Questionnaire (IPAQ), the body mass index and quality of life using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) and Supplementary Questionnaire Breast Cancer Module (QLQBR23). Results: The studied program demonstrated to be effective in improving the physical activity index (p=0,017) and quality of live related to physical functioning (p=0,016) and to global health status (p=0,030) at the end of the program. It was possible to find positive results in some other outcomes, such as fatigue, although the differences were not statistically significant. We believe that these results can be attributed to the small sample size. Conclusions: Although we can identify some methodological constrains, this is a very important study for scientific community working on the breast cancer subject. Once more, exercise was identified as an important strategy to improve global heath status in breast cancer patients. This represents an important contribution to the health professionals and mostly for physiotherapists working on health promotion subject.

Memento de imagens em dois actos. A imagem na aula de filosofia

Relatório de Estágio apresentado para cumprimento dos requisitos necessários à obtenção do grau de Mestre em Ensino da Filosofia no Ensino Secundário

Promoção da saúde oral: uma experiência de prevenção de cárie num concelho rural

Resumo: A cárie dentária, um problema que tem atingido populações em grande parte do mundo, é a doença de maior prevalência da cavidade oral, gerando graves consequências económicas e sociais. Esta doença tem sido estudada ao longo do tempo em diferentes países com o emprego de diversos índices, geralmente para o estudo da sua prevalência, a avaliação de medidas preventivas e o adequado planeamento das acções e serviços de saúde oral. O objectivo deste Projecto foi determinar se a escovagem quando realizada na escola com pasta fluoretada, e supervisionada pelos professores, duas vezes por dia, seria ou não eficaz na diminuição das populações microbianas de Streptococcus mutans e Lactobacillus e na consequente diminuição da incidência de cárie dentária. Material e métodos: Foram seleccionadas todas as crianças (universo = 178), com idades compreendidas entre os 5, 6 e 7 anos, residentes no Concelho de Aljustrel e a frequentar o pré-escolar e o primeiro ciclo do ensino básico oficial. Foi realizada a escovagem bi-diária com pasta fluoretada a 500 ppm F-, na escola, segundo o método de Bass modificado, supervisionado pelos professores titulares de turma, que tiveram formação da técnica de escovagem utilizada. Durante os 3 anos de estudo foram realizadas 6 observações dentárias e recolhas salivares para contagem de Streptococcus mutans e Lactobacillus e avaliação da capacidade tampão da saliva. Resultados: O grupo de estudo no início da intervenção apresentava valores dos índices de cárie dentária mais elevados do que os do grupo de controlo (mais 0,109 no CPO-S, 0,0749 no CPO-D, 1,505 no cpo-s e 0,831 no cpo-d), porém sem diferenças de significância estatística. A análise estatística dos resultados não veio confirmar este pressuposto uma vez que o grupo de estudo apresentou um aumento percentual ligeiramente maior do índice CPO-D (12,5%) do que o grupo de controlo (11,6%). Para além deste aspecto, ao contrário do que seria de esperar, não foi possível detectar nenhuma diferença estatisticamente significativa em nenhum dos índices de cárie dentária (cpo-s, cpo-d, CPO-S e CPO-D) entre o grupo de estudo e grupo de controlo entre a 1ª e última observação.Ainda que os resultados do estudo aqui apresentado tenham ficado aquém do esperado, deveria ser efectuada a escovagem diária na escola, uma vez por dia, com pasta fluoretada a 1000 ppm F-, atendendo a que esta medida contribui para a promoção da saúde e prevenção da doença e é facilitadora da construção de estilos de vida saudáveis.-------ABSTRACT: Dental caries, a problem that has affected populations worldwide, is one of the most prevalent diseases of the oral cavity, causing severe economic and social consequences. This disease has been studied over time in different countries with the use of various indices, usually for the knowledge of its prevalence, evaluation of preventive measures and appropriate planning of actions and oral services. The aim of this study was to determine whether toothbrushing when performed in schools, with fluoride toothpaste, and supervised by teachers twice a day, was effective in reducing microbial populations of Streptococcus mutans and Lactobacillus with consequent reduction in the incidence of dental caries. Material and Methods: All children aged 5, 6, and 7 years, from Aljustrel County, attending official pre-school and first cycle of basic educatio were selected. Toothbrushing was performed twice a day with toothpaste with 500 ppm F-,in the school, according to the modified Bass method, supervised by professors in the class, who were trained in the brushing technique used. During the study period were performed 6 observations of the dental status, and were also collected saliva for the count of Streptococcus mutans and Lactobacillus, and assessment of buffering capacity of saliva. Results: The study group at the beginning of the intervention had higher values of dental caries than the control group (more than 0,109 in DMF-S, 0, 0749 in DMF, dmf-s 1,505 and 0,831 in dmf-t) although without statistical significance. The expected results were not confirmed, since the study group had a slightly higher percentage increase of the DMF-T (12,5%) than the control group (11,6%). Apart from that, contrary to what one would expect, we could not detect any statistical significant difference in any of the indices of dental caries (dmf-s, dmf-t and DMF-S, DMF-T) between the study and the control group in all study periods. Although the study results were not has expected, toothbrushing should be performed daily at school, once a day with fluoride toothpaste with 1000 ppm F-, since this measure contributes to health promotion and disease prevention and encourages healthy lifestyles.

Technology governance in radiology: the example of magnetic resonance imaging

Based on the report for the unit “Project III” of the PhD programme on Technology Assessment in 2011. The unit was supervised by Prof. António B. Moniz (FCT-UNL).

Heuristics and policies for online pickup and delivery problems

Master Thesis