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.

Is multidimensional scaling suitable for mapping the input respiratory impedance in subjects and patients?

This paper presents the application of multidimensional scaling (MDS) analysis to data emerging from noninvasive lung function tests, namely the input respiratory impedance. The aim is to obtain a geometrical mapping of the diseases in a 3D space representation, allowing analysis of (dis)similarities between subjects within the same pathology groups, as well as between the various groups. The adult patient groups investigated were healthy, diagnosed chronic obstructive pulmonary disease (COPD) and diagnosed kyphoscoliosis, respectively. The children patient groups were healthy, asthma and cystic fibrosis. The results suggest that MDS can be successfully employed for mapping purposes of restrictive (kyphoscoliosis) and obstructive (COPD) pathologies. Hence, MDS tools can be further examined to define clear limits between pools of patients for clinical classification, and used as a training aid for medical traineeship.

Compact SMB chromatography for binary separation

A thesis submitted for the degree of Doctor of Philosophy

Time-domain optimization of amplifiers based on distributed genetic algorithms

Thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the subject of Electrical and Computer Engineering

Desenvolvimento de um Controlador de Boundary Scan (Ieee 1149.1)

Mestrado em Engenharia Electrotécnica e de Computadores - Área de Especialização em Automação e Sistemas

Avaliação e Otimização Energética das Fornalhas da Fábrica de Aromáticos

O consumo energético verificado nas refinarias petrolíferas é muito elevado, sendo as fornalhas os equipamentos que mais contribuem para esse consumo. Neste estudo foi efetuada uma avaliação e otimização energética às fornalhas da Fábrica de Aromáticos da Refinaria de Matosinhos. Numa primeira fase foi efetuado um levantamento exaustivo de dados de todas as correntes de entrada e saída dos equipamentos para posteriormente efetuar os balanços de massa e energia a cada uma das fornalhas. Os dados relativos ao levantamento compreenderam dois períodos de funcionamento distintos da unidade fabril, o período de funcionamento normal e o período relativo ao arranque. O período de funcionamento normal foi relativo ao ano de 2012 entre os meses de janeiro a setembro, por sua vez o período de arranque foi de dezembro de 2012 a março de 2013. Na segunda fase foram realizados os balanços de massa e energia quantificando todas as correntes de entrada e saída das fornalhas em termos mássicos e energéticos permitindo o cálculo do rendimento térmico das fornalhas para avaliar a sua performance. A avaliação energética permitiu concluir que existe um consumo maior de energia proveniente da combustão do Fuel Gás do que do Fuel Óleo, tanto no período de funcionamento normal como no arranque. As fornalhas H0101, H0301 e a H0471 possuem os consumos mais elevados, sendo responsáveis por mais de 70% do consumo da Fábrica de Aromáticos. Na terceira fase foram enunciadas duas medidas para a otimização energética das três fornalhas mais consumidoras de energia, a limpeza mensal e o uso exclusivo de Fuel Gás como combustível. As poupanças energéticas obtidas para uma limpeza mensal foram de 0,3% na fornalha H0101, 0,7% na fornalha H0301 e uma poupança de 0,9 % na fornalha H0471. Para o uso exclusivo de Fuel Gás obteve-se uma poupança de 0,9% na fornalha H0101 e uma poupança de 1,3% nas fornalhas H0301 e H0471. A análise económica efetuada à sugestão de alteração do combustível mostra que os custos de operação sofrerão um aumento anual de 621 679 €. Apesar do aumento dos custos, a redução na emissão de 24% de dióxido de carbono, poderá justificar este aumento na despesa.

Implementação de um processo industrial de coloração de granito

O presente estágio foi desenvolvido na Britafiel. Um dos projectos em que a Empresa se encontra envolvida é o STOCO, pretendendo implementar à escala industrial um processo de coloração de pedra granítica natural para fins decorativos. Foi neste projecto que se enquadrou o estágio. O tema do estágio centra-se no processo de coloração de granito tendo como principal foco a implementação de um processo industrial de produção de granito colorido. O projecto STOCO nasce da necessidade de complementar a actividade da empresa com produtos de maior valor acrescentado para valorização da matéria-prima de base, o granito. STOCO, resultante de Stone Color, é o nome dado ao projecto e ao novo produto que é granito colorido, sob a forma de brita. Pretende-se obter um produto amigo do ambiente e com boas características: manter a textura natural da pedra granítica e assegurar uma boa resistência a factores agressivos. Estudos prévios de qualidade e de toxicidade mostraram que o produto STOCO desenvolvido até então apresenta um bom comportamento face a agressões climatéricas e que não compromete a vida das espécies usadas nos testes (peixes). Em relação aos lixiviados e resíduos da pedra colorida STOCO, estes não apresentaram qualquer problema ambiental, sendo considerado um produto amigo do ambiente. À data de início do presente trabalho estava em funcionamento um equipamento protótipo de produção de granito colorido (100 kg/partida), sendo a instalação e o arranque da unidade industrial (3 ton/h) concretizados no início de 2014, já no decorrer deste trabalho. Os objectivos cumpridos no âmbito deste trabalho foram então a implementação de uma linha industrial de produção de brita colorida, avaliação técnica do processo e do custo industrial de produção associado às matérias-primas. Neste relatório é descrito o processo inicial adoptado e apresentam-se as alterações efectuadas para melhoria do processo produtivo. Resolveram-se problemas como: definição e instalação de equipamentos complementares para a entrada e a saída da brita no equipamento industrial; pó excessivo na brita; cheiro intenso a gás e elevado ruído; adequação do sistema de pintura; e secagem incompleta da brita. Alguns destes problemas não foram totalmente resolvidos, mas sim minimizados. O equipamento industrial necessita ainda de alterações em diversas áreas, que foram identificadas e para as quais são feitas sugestões de melhoria. Conseguiu-se ainda fazer alguns testes para uma possível substituição de alguns constituintes da tinta. Os componentes que entram na composição base da tinta aquosa, são de modo simplificado: ligante, pigmento, solvente e aditivos. Os constituintes que mais encarecem a tinta, e consequentemente o processo em causa, são o ligante e o pigmento. Os estudos efectuados precisam de ser aprofundados, na tentativa de melhorar o processo minimizando os custos de produção. Formalizaram-se os procedimentos escritos de produção STOCO tanto para o protótipo como para o processo industrial e elaborou-se uma ficha técnica de produto para a brita colorida STOCO.

Bridging present and future of brain-computer interfaces: an assessment of impacts

Based on the report for the “Project III” unit of the PhD programme on Technology Assessment under the supervision of Prof. António B. Moniz. This report was discussed also at the 2nd Winter School on Technology Assessment held at Universidade Nova de Lisboa, Caparica Campus, Portugal on December 2011.

Automatic Small Bowel Tumor Diagnosis by Using Multi-Scale Wavelet-Based Analysis in Wireless Capsule Endoscopy Images

BACKGROUND: Wireless capsule endoscopy has been introduced as an innovative, non-invasive diagnostic technique for evaluation of the gastrointestinal tract, reaching places where conventional endoscopy is unable to. However, the output of this technique is an 8 hours video, whose analysis by the expert physician is very time consuming. Thus, a computer assisted diagnosis tool to help the physicians to evaluate CE exams faster and more accurately is an important technical challenge and an excellent economical opportunity. METHOD: The set of features proposed in this paper to code textural information is based on statistical modeling of second order textural measures extracted from co-occurrence matrices. To cope with both joint and marginal non-Gaussianity of second order textural measures, higher order moments are used. These statistical moments are taken from the two-dimensional color-scale feature space, where two different scales are considered. Second and higher order moments of textural measures are computed from the co-occurrence matrices computed from images synthesized by the inverse wavelet transform of the wavelet transform containing only the selected scales for the three color channels. The dimensionality of the data is reduced by using Principal Component Analysis. RESULTS: The proposed textural features are then used as the input of a classifier based on artificial neural networks. Classification performances of 93.1% specificity and 93.9% sensitivity are achieved on real data. These promising results open the path towards a deeper study regarding the applicability of this algorithm in computer aided diagnosis systems to assist physicians in their clinical practice.

Brain-computer interface (BCI): a methodological proposal to assess the impacts of medical applications in 2022

Based on the report for “Project IV” unit of the PhD programme on Technology Assessment (Doctoral Conference) at Universidade Nova de Lisboa (December 2011). This thesis research has the supervision of António Moniz (FCT-UNL and ITAS-KIT) and Armin Grunwald (Karlsruhe Institute of Technology-ITAS, Germany). Other members of the thesis committee are Mário Forjaz Secca (FCT-UNL) and Femke Nijboer (University of Twente, Netherlands).

Ontology-based representation and generation of workflows for micro-task human-machine computation

Doctoral Program in Computer Science

Computer integrated testing for the transport refrigeration systems industry.

This is a study of a state of the art implementation of a new computer integrated testing (CIT) facility within a company that designs and manufactures transport refrigeration systems. The aim was to use state of the art hardware, software and planning procedures in the design and implementation of three CIT systems. Typical CIT system components include data acquisition (DAQ) equipment, application and analysis software, communication devices, computer-based instrumentation and computer technology. It is shown that the introduction of computer technology into the area of testing can have a major effect on such issues as efficiency, flexibility, data accuracy, test quality, data integrity and much more. Findings reaffirm how the overall area of computer integration continues to benefit any organisation, but with more recent advances in computer technology, communication methods and software capabilities, less expensive more sophisticated test solutions are now possible. This allows more organisations to benefit from the many advantages associated with CIT. Examples of computer integration test set-ups and the benefits associated with computer integration have been discussed.

Straightening out the concept of direct and indirect input requirements

Gim & Kim (1998) proposed a generalization of Jeong (1982, 1984) reinterpretation of the Hawkins-Simon condition for macroeconomic stability to off-diagonal matrix elements. This generalization is conceptually relevant for it offers a complementary view of interindustry linkages beyond final or net output influence. The extension is completely similar to the 'total flow' idea introduced by Szyrmer (1992) or the 'output-to-output' multiplier of Miller & Blair (2009). However the practical implementation of Gim & Kim is actually faulty since it confuses the appropriate order of output normalization. We provide a new and elementary solution for the correct formalization using standard interindustry accounting concepts.

Human-Computer Interaction with older people through ethnographical lenses: everyday interactions with ICT

Report for the scientific sojourn carried out at the School of Computing of the University of Dundee, United Kingdom, from 2010 to 2012. This document is a scientific report of the work done, main results, publications and accomplishment of the objectives of the 2-year post-doctoral research project with reference number BP-A 00239. The project has addressed the topic of older people (60+) and Information and Communication Technologies (ICT), which is a topic of growing social and research interest, from a Human-Computer Interaction perspective. Over a 2-year period (June 2010-June 2012), we have conducted classical ethnography of ICT use in a computer clubhouse in Scotland, addressing interaction barriers and strategies, social sharing practices in Social Network Sites, and ICT learning, and carried out rapid ethnographical studies related to geo-enabled ICT and e-government services towards supporting independent living and active ageing. The main results have provided a much deeper understanding of (i) the everyday use of Computer-Mediated Communication tools, such as video-chats and blogs, and its evolution as older people’s experience with ICT increases over time, (ii) cross-cultural aspects of ICT use in the north and south of Europe, (iii) the relevance of cognition over vision in interacting with geographical information and a wide range of ICT tools, despite common stereotypes (e.g. make things bigger), (iv) the important relationship offline-online to provide older people with socially inclusive and meaningful eservices for independent living and active ageing, (v) how older people carry out social sharing practices in the popular YouTube, (vi) their user experiences and (vii) the challenges they face in ICT learning and the strategies they use to become successful ICT learners over time. The research conducted in this project has been published in 17 papers, 4 in journals – two of which in JCR, 5 in conferences, 4 in workshops and 4 in magazines. Other public output consists of 10 invited talks and seminars.

Developing a Simple and Rapid Test for Monitoring the Heat Evolution of Concrete Mixtures for Both Laboratory and Field Applications, January 2006

Various test methods exist for measuring heat of cement hydration; however, most current methods require expensive equipment, complex testing procedures, and/or extensive time, thus not being suitable for field application. The objectives of this research are to identify, develop, and evaluate a standard test procedure for characterization and quality control of pavement concrete mixtures using a calorimetry technique. This research project has three phases. Phase I was designed to identify the user needs, including performance requirements and precision and bias limits, and to synthesize existing test methods for monitoring the heat of hydration, including device types, configurations, test procedures, measurements, advantages, disadvantages, applications, and accuracy. Phase II was designed to conduct experimental work to evaluate the calorimetry equipment recommended from the Phase I study and to develop a standard test procedure for using the equipment and interpreting the test results. Phase II also includes the development of models and computer programs for prediction of concrete pavement performance based on the characteristics of heat evolution curves. Phase III was designed to study for further development of a much simpler, inexpensive calorimeter for field concrete. In this report, the results from the Phase I study are presented, the plan for the Phase II study is described, and the recommendations for Phase III study are outlined. Phase I has been completed through three major activities: (1) collecting input and advice from the members of the project Technical Working Group (TWG), (2) conducting a literature survey, and (3) performing trials at the CP Tech Center’s research lab. The research results indicate that in addition to predicting maturity/strength, concrete heat evolution test results can also be used for (1) forecasting concrete setting time, (2) specifying curing period, (3) estimating risk of thermal cracking, (4) assessing pavement sawing/finishing time, (5) characterizing cement features, (6) identifying incompatibility of cementitious materials, (7) verifying concrete mix proportions, and (8) selecting materials and/or mix designs for given environmental conditions. Besides concrete materials and mix proportions, the configuration of the calorimeter device, sample size, mixing procedure, and testing environment (temperature) also have significant influences on features of concrete heat evolution process. The research team has found that although various calorimeter tests have been conducted for assorted purposes and the potential uses of calorimeter tests are clear, there is no consensus on how to utilize the heat evolution curves to characterize concrete materials and how to effectively relate the characteristics of heat evolution curves to concrete pavement performance. The goal of the Phase II study is to close these gaps.