Development of a hand-gesture recognition system for human-computer interaction

The aim of this Master Thesis is the analysis, design and development of a robust and reliable Human-Computer Interaction interface, based on visual hand-gesture recognition. The implementation of the required functions is oriented to the simulation of a classical hardware interaction device: the mouse, by recognizing a specific hand-gesture vocabulary in color video sequences. For this purpose, a prototype of a hand-gesture recognition system has been designed and implemented, which is composed of three stages: detection, tracking and recognition. This system is based on machine learning methods and pattern recognition techniques, which have been integrated together with other image processing approaches to get a high recognition accuracy and a low computational cost. Regarding pattern recongition techniques, several algorithms and strategies have been designed and implemented, which are applicable to color images and video sequences. The design of these algorithms has the purpose of extracting spatial and spatio-temporal features from static and dynamic hand gestures, in order to identify them in a robust and reliable way. Finally, a visual database containing the necessary vocabulary of gestures for interacting with the computer has been created.

Análise de grandezas cinemáticas e dinâmicas inerentes à hemiparesia através da descoberta de conhecimento em bases de dados

Em virtude de uma elevada expectativa de vida mundial, faz-se crescente a probabilidade de ocorrer acidentes naturais e traumas físicos no cotidiano, o que ocasiona um aumento na demanda por reabilitação. A terapia física, sob o paradigma da reabilitação robótica com serious games, oferece maior motivação e engajamento do paciente ao tratamento, cujo emprego foi recomendado pela American Heart Association (AHA), apontando a mais alta avaliação (Level A) para pacientes internados e ambulatoriais. No entanto, o potencial de análise dos dados coletados pelos dispositivos robóticos envolvidos é pouco explorado, deixando de extrair informações que podem ser de grande valia para os tratamentos. O foco deste trabalho consiste na aplicação de técnicas para descoberta de conhecimento, classificando o desempenho de pacientes diagnosticados com hemiparesia crônica. Os pacientes foram inseridos em um ambiente de reabilitação robótica, fazendo uso do InMotion ARM, um dispositivo robótico para reabilitação de membros superiores e coleta dos dados de desempenho. Foi aplicado sobre os dados um roteiro para descoberta de conhecimento em bases de dados, desempenhando pré-processamento, transformação (extração de características) e então a mineração de dados a partir de algoritmos de aprendizado de máquina. A estratégia do presente trabalho culminou em uma classificação de padrões com a capacidade de distinguir lados hemiparéticos sob uma precisão de 94%, havendo oito atributos alimentando a entrada do mecanismo obtido. Interpretando esta coleção de atributos, foi observado que dados de força são mais significativos, os quais abrangem metade da composição de uma amostra.

Inference of binding affinity from neuronal receptors in humans

Tese de mestrado, Bioinformática e Biologia Computacional (Bioinformática), Universidade de Lisboa, Faculdade de Ciências, 2016

Approaches and Strategy for Cancer Research and Surveillance Data: Integration, Information Pipeline, Data Models, and Informatics Opportunities

Thesis (Ph.D.)--University of Washington, 2016-04

Understanding human genome regulation through entropic graph-based regularization and submodular optimization

Thesis (Ph.D.)--University of Washington, 2016-06

Genes for the majority of group A streptococcal virulence factors and extracellular surface proteins do not confer an increased propensity to cause invasive disease

Background. The factors behind the reemergence of severe, invasive group A streptococcal (GAS) diseases are unclear, but it could be caused by altered genetic endowment in these organisms. However, data from previous studies assessing the association between single genetic factors and invasive disease are often conflicting, suggesting that other, as-yet unidentified factors are necessary for the development of this class of disease. Methods. In this study, we used a targeted GAS virulence microarray containing 226 GAS genes to determine the virulence gene repertoires of 68 GAS isolates (42 associated with invasive disease and 28 associated with noninvasive disease) collected in a defined geographic location during a contiguous time period. We then employed 3 advanced machine learning methods (genetic algorithm neural network, support vector machines, and classification trees) to identify genes with an increased association with invasive disease. Results. Virulence gene profiles of individual GAS isolates varied extensively among these geographically and temporally related strains. Using genetic algorithm neural network analysis, we identified 3 genes with a marginal overrepresentation in invasive disease isolates. Significantly, 2 of these genes, ssa and mf4, encoded superantigens but were only present in a restricted set of GAS M-types. The third gene, spa, was found in variable distributions in all M-types in the study. Conclusions. Our comprehensive analysis of GAS virulence profiles provides strong evidence for the incongruent relationships among any of the 226 genes represented on the array and the overall propensity of GAS to cause invasive disease, underscoring the pathogenic complexity of these diseases, as well as the importance of multiple bacteria and/ or host factors.

Prediction of protein continuum secondary structure with probabilistic models based on NMR solved structures

Background: The structure of proteins may change as a result of the inherent flexibility of some protein regions. We develop and explore probabilistic machine learning methods for predicting a continuum secondary structure, i.e. assigning probabilities to the conformational states of a residue. We train our methods using data derived from high-quality NMR models. Results: Several probabilistic models not only successfully estimate the continuum secondary structure, but also provide a categorical output on par with models directly trained on categorical data. Importantly, models trained on the continuum secondary structure are also better than their categorical counterparts at identifying the conformational state for structurally ambivalent residues. Conclusion: Cascaded probabilistic neural networks trained on the continuum secondary structure exhibit better accuracy in structurally ambivalent regions of proteins, while sustaining an overall classification accuracy on par with standard, categorical prediction methods.

Rule extraction from technology IPOs in the US stock market

Machine learning techniques for prediction and rule extraction from artificial neural network methods are used. The hypothesis that market sentiment and IPO specific attributes are equally responsible for first-day IPO returns in the US stock market is tested. Machine learning methods used are Bayesian classifications, support vector machines, decision tree techniques, rule learners and artificial neural networks. The outcomes of the research are predictions and rules associated With first-day returns of technology IPOs. The hypothesis that first-day returns of technology IPOs are equally determined by IPO specific and market sentiment is rejected. Instead lower yielding IPOs are determined by IPO specific and market sentiment attributes, while higher yielding IPOs are largely dependent on IPO specific attributes.

AllerTOP v.2 - a server for in silico prediction of allergens

Allergy is an overreaction by the immune system to a previously encountered, ordinarily harmless substance - typically proteins - resulting in skin rash, swelling of mucous membranes, sneezing or wheezing, or other abnormal conditions. The use of modified proteins is increasingly widespread: their presence in food, commercial products, such as washing powder, and medical therapeutics and diagnostics, makes predicting and identifying potential allergens a crucial societal issue. The prediction of allergens has been explored widely using bioinformatics, with many tools being developed in the last decade; many of these are freely available online. Here, we report a set of novel models for allergen prediction utilizing amino acid E-descriptors, auto- and cross-covariance transformation, and several machine learning methods for classification, including logistic regression (LR), decision tree (DT), naïve Bayes (NB), random forest (RF), multilayer perceptron (MLP) and k nearest neighbours (kNN). The best performing method was kNN with 85.3% accuracy at 5-fold cross-validation. The resulting model has been implemented in a revised version of the AllerTOP server (http://www.ddg-pharmfac.net/AllerTOP). © Springer-Verlag 2014.

AllerTOP - a server for in silico prediction of allergens

Background: Allergy is a form of hypersensitivity to normally innocuous substances, such as dust, pollen, foods or drugs. Allergens are small antigens that commonly provoke an IgE antibody response. There are two types of bioinformatics-based allergen prediction. The first approach follows FAO/WHO Codex alimentarius guidelines and searches for sequence similarity. The second approach is based on identifying conserved allergenicity-related linear motifs. Both approaches assume that allergenicity is a linearly coded property. In the present study, we applied ACC pre-processing to sets of known allergens, developing alignment-independent models for allergen recognition based on the main chemical properties of amino acid sequences.Results: A set of 684 food, 1,156 inhalant and 555 toxin allergens was collected from several databases. A set of non-allergens from the same species were selected to mirror the allergen set. The amino acids in the protein sequences were described by three z-descriptors (z1, z2 and z3) and by auto- and cross-covariance (ACC) transformation were converted into uniform vectors. Each protein was presented as a vector of 45 variables. Five machine learning methods for classification were applied in the study to derive models for allergen prediction. The methods were: discriminant analysis by partial least squares (DA-PLS), logistic regression (LR), decision tree (DT), naïve Bayes (NB) and k nearest neighbours (kNN). The best performing model was derived by kNN at k = 3. It was optimized, cross-validated and implemented in a server named AllerTOP, freely accessible at http://www.pharmfac.net/allertop. AllerTOP also predicts the most probable route of exposure. In comparison to other servers for allergen prediction, AllerTOP outperforms them with 94% sensitivity.Conclusions: AllerTOP is the first alignment-free server for in silico prediction of allergens based on the main physicochemical properties of proteins. Significantly, as well allergenicity AllerTOP is able to predict the route of allergen exposure: food, inhalant or toxin. © 2013 Dimitrov et al.; licensee BioMed Central Ltd.

Padrões temporais e espaciais na detecção de resíduos e contaminantes na produção de aves e suínos no Brasil

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Agronomia e Medicina Veterinária, 2016.

Neural states in parietal areas during arm reaching

Since the first subdivisions of the brain into macro regions, it has always been thought a priori that, given the heterogeneity of neurons, different areas host specific functions and process unique information in order to generate a behaviour. Moreover, the various sensory inputs coming from different sources (eye, skin, proprioception) flow from one macro area to another, being constantly computed and updated. Therefore, especially for non-contiguous cortical areas, it is not expected to find the same information. From this point of view, it would be inconceivable that the motor and the parietal cortices, diversified by the information encoded and by the anatomical position in the brain, could show very similar neural dynamics. With the present thesis, by analyzing the population activity of parietal areas V6A and PEc with machine learning methods, we argue that a simplified view of the brain organization do not reflect the actual neural processes. We reliably detected a number of neural states that were tightly linked to distinct periods of the task sequence, i.e. the planning and execution of movement and the holding of target as already observed in motor cortices. The states before and after the movement could be further segmented into two states related to different stages of movement planning and arm posture processing. Rather unexpectedly, we found that activity during the movement could be parsed into two states of equal duration temporally linked to the acceleration and deceleration phases of the arm. Our findings suggest that, at least during arm reaching in 3D space, the posterior parietal cortex (PPC) shows low-level population neural dynamics remarkably similar to those found in the motor cortices. In addition, the present findings suggest that computational processes in PPC could be better understood if studied using a dynamical system approach rather than studying a mosaic of single units.

Addressing challenges in heterogeneous embedded systems programming: security and performance

Embedded systems are increasingly integral to daily life, improving and facilitating the efficiency of modern Cyber-Physical Systems which provide access to sensor data, and actuators. As modern architectures become increasingly complex and heterogeneous, their optimization becomes a challenging task. Additionally, ensuring platform security is important to avoid harm to individuals and assets. This study primarily addresses challenges in contemporary Embedded Systems, focusing on platform optimization and security enforcement. The initial section of this study delves into the application of machine learning methods to efficiently determine the optimal number of cores for a parallel RISC-V cluster to minimize energy consumption using static source code analysis. Results demonstrate that automated platform configuration is not only viable but also that there is a moderate performance trade-off when relying solely on static features. The second part focuses on addressing the problem of heterogeneous device mapping, which involves assigning tasks to the most suitable computational device in a heterogeneous platform for optimal runtime. The contribution of this section lies in the introduction of novel pre-processing techniques, along with a training framework called Siamese Networks, that enhances the classification performance of DeepLLVM, an advanced approach for task mapping. Importantly, these proposed approaches are independent from the specific deep-learning model used. Finally, this research work focuses on addressing issues concerning the binary exploitation of software running in modern Embedded Systems. It proposes an architecture to implement Control-Flow Integrity in embedded platforms with a Root-of-Trust, aiming to enhance security guarantees with limited hardware modifications. The approach involves enhancing the architecture of a modern RISC-V platform for autonomous vehicles by implementing a side-channel communication mechanism that relays control-flow changes executed by the process running on the host core to the Root-of-Trust. This approach has limited impact on performance and it is effective in enhancing the security of embedded platforms.

Learning Preferences with Kernel-Based Methods

Learning of preference relations has recently received significant attention in machine learning community. It is closely related to the classification and regression analysis and can be reduced to these tasks. However, preference learning involves prediction of ordering of the data points rather than prediction of a single numerical value as in case of regression or a class label as in case of classification. Therefore, studying preference relations within a separate framework facilitates not only better theoretical understanding of the problem, but also motivates development of the efficient algorithms for the task. Preference learning has many applications in domains such as information retrieval, bioinformatics, natural language processing, etc. For example, algorithms that learn to rank are frequently used in search engines for ordering documents retrieved by the query. Preference learning methods have been also applied to collaborative filtering problems for predicting individual customer choices from the vast amount of user generated feedback. In this thesis we propose several algorithms for learning preference relations. These algorithms stem from well founded and robust class of regularized least-squares methods and have many attractive computational properties. In order to improve the performance of our methods, we introduce several non-linear kernel functions. Thus, contribution of this thesis is twofold: kernel functions for structured data that are used to take advantage of various non-vectorial data representations and the preference learning algorithms that are suitable for different tasks, namely efficient learning of preference relations, learning with large amount of training data, and semi-supervised preference learning. Proposed kernel-based algorithms and kernels are applied to the parse ranking task in natural language processing, document ranking in information retrieval, and remote homology detection in bioinformatics domain. Training of kernel-based ranking algorithms can be infeasible when the size of the training set is large. This problem is addressed by proposing a preference learning algorithm whose computation complexity scales linearly with the number of training data points. We also introduce sparse approximation of the algorithm that can be efficiently trained with large amount of data. For situations when small amount of labeled data but a large amount of unlabeled data is available, we propose a co-regularized preference learning algorithm. To conclude, the methods presented in this thesis address not only the problem of the efficient training of the algorithms but also fast regularization parameter selection, multiple output prediction, and cross-validation. Furthermore, proposed algorithms lead to notably better performance in many preference learning tasks considered.

Kernel-Based Ranking. Methods for Learning and Performance Estimation

Machine learning provides tools for automated construction of predictive models in data intensive areas of engineering and science. The family of regularized kernel methods have in the recent years become one of the mainstream approaches to machine learning, due to a number of advantages the methods share. The approach provides theoretically well-founded solutions to the problems of under- and overfitting, allows learning from structured data, and has been empirically demonstrated to yield high predictive performance on a wide range of application domains. Historically, the problems of classification and regression have gained the majority of attention in the field. In this thesis we focus on another type of learning problem, that of learning to rank. In learning to rank, the aim is from a set of past observations to learn a ranking function that can order new objects according to how well they match some underlying criterion of goodness. As an important special case of the setting, we can recover the bipartite ranking problem, corresponding to maximizing the area under the ROC curve (AUC) in binary classification. Ranking applications appear in a large variety of settings, examples encountered in this thesis include document retrieval in web search, recommender systems, information extraction and automated parsing of natural language. We consider the pairwise approach to learning to rank, where ranking models are learned by minimizing the expected probability of ranking any two randomly drawn test examples incorrectly. The development of computationally efficient kernel methods, based on this approach, has in the past proven to be challenging. Moreover, it is not clear what techniques for estimating the predictive performance of learned models are the most reliable in the ranking setting, and how the techniques can be implemented efficiently. The contributions of this thesis are as follows. First, we develop RankRLS, a computationally efficient kernel method for learning to rank, that is based on minimizing a regularized pairwise least-squares loss. In addition to training methods, we introduce a variety of algorithms for tasks such as model selection, multi-output learning, and cross-validation, based on computational shortcuts from matrix algebra. Second, we improve the fastest known training method for the linear version of the RankSVM algorithm, which is one of the most well established methods for learning to rank. Third, we study the combination of the empirical kernel map and reduced set approximation, which allows the large-scale training of kernel machines using linear solvers, and propose computationally efficient solutions to cross-validation when using the approach. Next, we explore the problem of reliable cross-validation when using AUC as a performance criterion, through an extensive simulation study. We demonstrate that the proposed leave-pair-out cross-validation approach leads to more reliable performance estimation than commonly used alternative approaches. Finally, we present a case study on applying machine learning to information extraction from biomedical literature, which combines several of the approaches considered in the thesis. The thesis is divided into two parts. Part I provides the background for the research work and summarizes the most central results, Part II consists of the five original research articles that are the main contribution of this thesis.