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.

Transimpedance amplifier for integrated SpO2 optic sensor

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

Integrated ecological-economic modelling and assessment aproach for coastal ecosystem management

Thesis submitted to the Faculty of Sciences and Technology, New University of Lisbon, for the degree of Doctor of Philosophy in Environmental Sciences

Heterogeneous sensor database framework for the sensor observation service: integrating remote and in-situ sensor observations at the database backend

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

Integrated study by NMR and X-ray Crystallography on the analysis of the molecular interactions in heme-binding proteins

Dissertação para obtenção do Grau de Doutor em Bioquímica, Especialidade Bioquímica Estrutural

Integrated care approaches and skills mix policies in Europe

ABSTRACT - Background: Integration of health care services is emerging as a central challenge of health care delivery, particularly for patients with elderly and complex chronic conditions. In 2003, the World Health Organization (WHO) already began to identify it as one of the key pathways to improve primary care. In 2005, the European Commission declared integrated care as vital for the sustainability of social protection systems in Europe. Nowadays, it is recognized as a core component of health and social care reforms across European countries. Implementing integrated care requires coordination between settings, organizations, providers and professionals. In order to address the challenge of integration in such complex scenario, an effective workforce is required capable of working across interdependent settings. The World Health Report 2006 noted that governments should prepare their workforce and explore what tasks the different levels of health workers are trained to do and are capable of performing (skills mix). Comparatively to other European countries, Portugal is at an early stage in what integrated care is concerned facing a growing elderly population and the subsequent increase in the pressure on institutions and professionals to provide social and medical care in the most cost-effective way. In 2006 the Portuguese government created the Portuguese Network for Integrated Care Development (PNICD) to solve the existing long-term gap in social support and healthcare. On what concerns health workforce, the Portuguese government already recognized the importance of redefine careers keeping professional motivation and satisfaction. Aim of the study: This study aims to contribute new evidence to the debate surrounding integrated care and skills mix policies in Europe. It also seeks to provide the first evidence that incorporates both the current dynamics of implementing integrated care in Portugal and the developments of international literature. The first ambition of our study is to contribute to the growing interest in integrated care and to the ongoing research in this area by identifying its different approaches and retrieve a number of experiences in some European countries. Our second goal of this research is to produce an update on the knowledge developed on skills mix to the international healthcare management community and to policy makers involved in reforming healthcare systems and organizations. To better inform Portuguese health policies makers in a third stage we explore the current dynamics of implementing integrated care in Portugal and contextualize them with the developments reported in the international literature. Methodology: This is essentially an exploratory and descriptive study using qualitative methodology. In order to identify integrated care approaches in Europe, a systematic literature review was undertaken which resulted in a paper published in the Journal of Management and Marketing in Health care titled: Approaches to developing integrated care in Europe: a systematic literature review. This article was recommended and included into a list of references identified by The King's Fund Library. A second systematic literature review was undertaken which resulted in a paper published in the International Journal of Healthcare Management titled: Skills mix in healthcare: An international update for the management debate. Semi-structured interviews were performed on experts representing the regional coordination teams of the Portuguese Network for Integrated Care Development. In a last stage a questionnaire survey was developed based on the findings of both systematic literature reviews and semi-structured interviews. Conclusions: Even though integrated care is a worldwide trend in health care reforms, there is no unique definition. Definitions can be grouped according to their sectorial focus: community-based care, combined health and social care, combined acute and primary care, the integration of providers, and in a more comprehensive approach the whole health system. Indeed, models that seek to apply the principles of integrated care have a similar background and are continually evolving and depend on the different initiatives taken at national level. . Despite the fact that we cannot argue that there is one single set typology of models for integrated care, it is possible to identify and categorize some of the basic approaches that have been taken in attempts to implement integrated care according to: changes in organizational structure, workforce reconfiguring, and changes in the financing system. The systematic literature review on skills mix showed that despite the widely acknowledged interest on skills mix initiatives there is a lack of evidence on skills mix implications, constraints, outcomes, and quality impact that would allow policy makers to take sustained and evidence-based decisions. Within the Portuguese health system, the integrated care approach is rather organizational and financial, whereas little attention is given to workforce integration. On what concerns workforce planning Portugal it is still in the stage of analyzing the acceptability of health workforce skills mix. In line with the international approaches, integration of health and social services and bridging primary and acute care are the main goals of the national government strategy. The findings from our interviews clarify perceptions which show no discrepancy with the related literature but are rather scarce comparing to international experience. Informants hold a realistic but narrow view of integrated care related issues. They seem to be limited to the regional context, requiring a more comprehensive perspective. The questionnaire developed in this thesis is an instrument which, when applied, will allow policy makers to understand the basic set of concepts and managerial motivations behind national and regional integrated care programs. The instrument developed can foster evidence on the three essential components of integrated care policies: organizational, financial, and human resources development, and can give additional input on the context in which integrated care is being developed, the type of providers and organizations involved, barriers and constraints, and the workforce skills mix planning related strategies. The thesis was successful in recognizing differences between countries and interventions and the instrument developed will allow a better comprehension of the international options available and how to address the vital components of integrated care programs.

Exporting a model of integrated disease management: A cross-case study analysis and application to Associação Protectora dos Diabéticos de Portugal

Although literature is lacking in the topic of internationalization of services, we manage to apply both the Uppsala model and the Eclectic Theory to the healthcare service. A cross-case study analysis with three international hospitals is done in order to define an internationalization pattern and conditions for a successful process. This is then applied to Associação Protectora dos Diabéticos de Portugal with the purpose of defining an internationalization strategy to the Association.

Using an LED as a sensor and visible light communication device in a smart illumination system

The need for more efficient illumination systems has led to the proliferation of Solid-State Lighting (SSL) systems, which offer optimized power consumption. SSL systems are comprised of LED devices which are intrinsically fast devices and permit very fast light modulation. This, along with the congestion of the radio frequency spectrum has paved the path for the emergence of Visible Light Communication (VLC) systems. VLC uses free space to convey information by using light modulation. Notwithstanding, as VLC systems proliferate and cost competitiveness ensues, there are two important aspects to be considered. State-of-the-art VLC implementations use power demanding PAs, and thus it is important to investigate if regular, existent Switched-Mode Power Supply (SMPS) circuits can be adapted for VLC use. A 28 W buck regulator was implemented using a off-the-shelf LED Driver integrated circuit, using both series and parallel dimming techniques. Results show that optical clock frequencies up to 500 kHz are achievable without any major modification besides adequate component sizing. The use of an LED as a sensor was investigated, in a short-range, low-data-rate perspective. Results show successful communication in an LED-to-LED configuration, with enhanced range when using LED strings as sensors. Besides, LEDs present spectral selective sensitivity, which makes them good contenders for a multi-colour LED-to-LED system, such as in the use of RGB displays and lamps. Ultimately, the present work shows evidence that LEDs can be used as a dual-purpose device, enabling not only illumination, but also bi-directional data communication.