First-principle data-driven models for assessment of motor disorders in Parkinson’s disease

Parkinson’s disease (PD) is an increasing neurological disorder in an aging society. The motor and non-motor symptoms of PD advance with the disease progression and occur in varying frequency and duration. In order to affirm the full extent of a patient’s condition, repeated assessments are necessary to adjust medical prescription. In clinical studies, symptoms are assessed using the unified Parkinson’s disease rating scale (UPDRS). On one hand, the subjective rating using UPDRS relies on clinical expertise. On the other hand, it requires the physical presence of patients in clinics which implies high logistical costs. Another limitation of clinical assessment is that the observation in hospital may not accurately represent a patient’s situation at home. For such reasons, the practical frequency of tracking PD symptoms may under-represent the true time scale of PD fluctuations and may result in an overall inaccurate assessment. Current technologies for at-home PD treatment are based on data-driven approaches for which the interpretation and reproduction of results are problematic.  The overall objective of this thesis is to develop and evaluate unobtrusive computer methods for enabling remote monitoring of patients with PD. It investigates first-principle data-driven model based novel signal and image processing techniques for extraction of clinically useful information from audio recordings of speech (in texts read aloud) and video recordings of gait and finger-tapping motor examinations. The aim is to map between PD symptoms severities estimated using novel computer methods and the clinical ratings based on UPDRS part-III (motor examination). A web-based test battery system consisting of self-assessment of symptoms and motor function tests was previously constructed for a touch screen mobile device. A comprehensive speech framework has been developed for this device to analyze text-dependent running speech by: (1) extracting novel signal features that are able to represent PD deficits in each individual component of the speech system, (2) mapping between clinical ratings and feature estimates of speech symptom severity, and (3) classifying between UPDRS part-III severity levels using speech features and statistical machine learning tools. A novel speech processing method called cepstral separation difference showed stronger ability to classify between speech symptom severities as compared to existing features of PD speech. In the case of finger tapping, the recorded videos of rapid finger tapping examination were processed using a novel computer-vision (CV) algorithm that extracts symptom information from video-based tapping signals using motion analysis of the index-finger which incorporates a face detection module for signal calibration. This algorithm was able to discriminate between UPDRS part III severity levels of finger tapping with high classification rates. Further analysis was performed on novel CV based gait features constructed using a standard human model to discriminate between a healthy gait and a Parkinsonian gait. The findings of this study suggest that the symptom severity levels in PD can be discriminated with high accuracies by involving a combination of first-principle (features) and data-driven (classification) approaches. The processing of audio and video recordings on one hand allows remote monitoring of speech, gait and finger-tapping examinations by the clinical staff. On the other hand, the first-principles approach eases the understanding of symptom estimates for clinicians. We have demonstrated that the selected features of speech, gait and finger tapping were able to discriminate between symptom severity levels, as well as, between healthy controls and PD patients with high classification rates. The findings support suitability of these methods to be used as decision support tools in the context of PD assessment.

Algoritmo Q-learning como estratégia de exploração e/ou explotação para metaheurísticas GRASP e algoritmo genético

Techniques of optimization known as metaheuristics have achieved success in the resolution of many problems classified as NP-Hard. These methods use non deterministic approaches that reach very good solutions which, however, don t guarantee the determination of the global optimum. Beyond the inherent difficulties related to the complexity that characterizes the optimization problems, the metaheuristics still face the dilemma of xploration/exploitation, which consists of choosing between a greedy search and a wider exploration of the solution space. A way to guide such algorithms during the searching of better solutions is supplying them with more knowledge of the problem through the use of a intelligent agent, able to recognize promising regions and also identify when they should diversify the direction of the search. This way, this work proposes the use of Reinforcement Learning technique - Q-learning Algorithm - as exploration/exploitation strategy for the metaheuristics GRASP (Greedy Randomized Adaptive Search Procedure) and Genetic Algorithm. The GRASP metaheuristic uses Q-learning instead of the traditional greedy-random algorithm in the construction phase. This replacement has the purpose of improving the quality of the initial solutions that are used in the local search phase of the GRASP, and also provides for the metaheuristic an adaptive memory mechanism that allows the reuse of good previous decisions and also avoids the repetition of bad decisions. In the Genetic Algorithm, the Q-learning algorithm was used to generate an initial population of high fitness, and after a determined number of generations, where the rate of diversity of the population is less than a certain limit L, it also was applied to supply one of the parents to be used in the genetic crossover operator. Another significant change in the hybrid genetic algorithm is the proposal of a mutually interactive cooperation process between the genetic operators and the Q-learning algorithm. In this interactive/cooperative process, the Q-learning algorithm receives an additional update in the matrix of Q-values based on the current best solution of the Genetic Algorithm. The computational experiments presented in this thesis compares the results obtained with the implementation of traditional versions of GRASP metaheuristic and Genetic Algorithm, with those obtained using the proposed hybrid methods. Both algorithms had been applied successfully to the symmetrical Traveling Salesman Problem, which was modeled as a Markov decision process

Aplicação de sistemas multi-classificadores no diagnóstico de falhas em motores de indução trifásicos

Equipment maintenance is the major cost factor in industrial plants, it is very important the development of fault predict techniques. Three-phase induction motors are key electrical equipments used in industrial applications mainly because presents low cost and large robustness, however, it isn t protected from other fault types such as shorted winding and broken bars. Several acquisition ways, processing and signal analysis are applied to improve its diagnosis. More efficient techniques use current sensors and its signature analysis. In this dissertation, starting of these sensors, it is to make signal analysis through Park s vector that provides a good visualization capability. Faults data acquisition is an arduous task; in this way, it is developed a methodology for data base construction. Park s transformer is applied into stationary reference for machine modeling of the machine s differential equations solution. Faults detection needs a detailed analysis of variables and its influences that becomes the diagnosis more complex. The tasks of pattern recognition allow that systems are automatically generated, based in patterns and data concepts, in the majority cases undetectable for specialists, helping decision tasks. Classifiers algorithms with diverse learning paradigms: k-Neighborhood, Neural Networks, Decision Trees and Naïves Bayes are used to patterns recognition of machines faults. Multi-classifier systems are used to improve classification errors. It inspected the algorithms homogeneous: Bagging and Boosting and heterogeneous: Vote, Stacking and Stacking C. Results present the effectiveness of constructed model to faults modeling, such as the possibility of using multi-classifiers algorithm on faults classification

Estudo avaliativo de um algoritmo genético auto-organizável e multiobjetivo utilizando aprendizado de máquina para aplicações de telecomunicações

This paper presents an evaluative study about the effects of using a machine learning technique on the main features of a self-organizing and multiobjective genetic algorithm (GA). A typical GA can be seen as a search technique which is usually applied in problems involving no polynomial complexity. Originally, these algorithms were designed to create methods that seek acceptable solutions to problems where the global optimum is inaccessible or difficult to obtain. At first, the GAs considered only one evaluation function and a single objective optimization. Today, however, implementations that consider several optimization objectives simultaneously (multiobjective algorithms) are common, besides allowing the change of many components of the algorithm dynamically (self-organizing algorithms). At the same time, they are also common combinations of GAs with machine learning techniques to improve some of its characteristics of performance and use. In this work, a GA with a machine learning technique was analyzed and applied in a antenna design. We used a variant of bicubic interpolation technique, called 2D Spline, as machine learning technique to estimate the behavior of a dynamic fitness function, based on the knowledge obtained from a set of laboratory experiments. This fitness function is also called evaluation function and, it is responsible for determining the fitness degree of a candidate solution (individual), in relation to others in the same population. The algorithm can be applied in many areas, including in the field of telecommunications, as projects of antennas and frequency selective surfaces. In this particular work, the presented algorithm was developed to optimize the design of a microstrip antenna, usually used in wireless communication systems for application in Ultra-Wideband (UWB). The algorithm allowed the optimization of two variables of geometry antenna - the length (Ls) and width (Ws) a slit in the ground plane with respect to three objectives: radiated signal bandwidth, return loss and central frequency deviation. These two dimensions (Ws and Ls) are used as variables in three different interpolation functions, one Spline for each optimization objective, to compose a multiobjective and aggregate fitness function. The final result proposed by the algorithm was compared with the simulation program result and the measured result of a physical prototype of the antenna built in the laboratory. In the present study, the algorithm was analyzed with respect to their success degree in relation to four important characteristics of a self-organizing multiobjective GA: performance, flexibility, scalability and accuracy. At the end of the study, it was observed a time increase in algorithm execution in comparison to a common GA, due to the time required for the machine learning process. On the plus side, we notice a sensitive gain with respect to flexibility and accuracy of results, and a prosperous path that indicates directions to the algorithm to allow the optimization problems with "η" variables

Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

Algoritmos genéticos aplicados a um comitê de LS-SVM em problemas de classificação

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

Experience generalization for multi-agent reinforcement learning

On-line learning methods have been applied successfully in multi-agent systems to achieve coordination among agents. Learning in multi-agent systems implies in a non-stationary scenario perceived by the agents, since the behavior of other agents may change as they simultaneously learn how to improve their actions. Non-stationary scenarios can be modeled as Markov Games, which can be solved using the Minimax-Q algorithm a combination of Q-learning (a Reinforcement Learning (RL) algorithm which directly learns an optimal control policy) and the Minimax algorithm. However, finding optimal control policies using any RL algorithm (Q-learning and Minimax-Q included) can be very time consuming. Trying to improve the learning time of Q-learning, we considered the QS-algorithm. in which a single experience can update more than a single action value by using a spreading function. In this paper, we contribute a Minimax-QS algorithm which combines the Minimax-Q algorithm and the QS-algorithm. We conduct a series of empirical evaluation of the algorithm in a simplified simulator of the soccer domain. We show that even using a very simple domain-dependent spreading function, the performance of the learning algorithm can be improved.

An Optimum-Path Forest framework for intrusion detection in computer networks

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Efficient supervised optimum-path forest classification for large datasets

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Otimização em comitês de classificadores: uma abordagem baseada em filtro para seleção de subconjuntos de atributos

Traditional applications of feature selection in areas such as data mining, machine learning and pattern recognition aim to improve the accuracy and to reduce the computational cost of the model. It is done through the removal of redundant, irrelevant or noisy data, finding a representative subset of data that reduces its dimensionality without loss of performance. With the development of research in ensemble of classifiers and the verification that this type of model has better performance than the individual models, if the base classifiers are diverse, comes a new field of application to the research of feature selection. In this new field, it is desired to find diverse subsets of features for the construction of base classifiers for the ensemble systems. This work proposes an approach that maximizes the diversity of the ensembles by selecting subsets of features using a model independent of the learning algorithm and with low computational cost. This is done using bio-inspired metaheuristics with evaluation filter-based criteria

Abordagem de construção de arquitetura homogênea para comitês via meta-aprendizagem

In the world we are constantly performing everyday actions. Two of these actions are frequent and of great importance: classify (sort by classes) and take decision. When we encounter problems with a relatively high degree of complexity, we tend to seek other opinions, usually from people who have some knowledge or even to the extent possible, are experts in the problem domain in question in order to help us in the decision-making process. Both the classification process as the process of decision making, we are guided by consideration of the characteristics involved in the specific problem. The characterization of a set of objects is part of the decision making process in general. In Machine Learning this classification happens through a learning algorithm and the characterization is applied to databases. The classification algorithms can be employed individually or by machine committees. The choice of the best methods to be used in the construction of a committee is a very arduous task. In this work, it will be investigated meta-learning techniques in selecting the best configuration parameters of homogeneous committees for applications in various classification problems. These parameters are: the base classifier, the architecture and the size of this architecture. We investigated nine types of inductors candidates for based classifier, two methods of generation of architecture and nine medium-sized groups for architecture. Dimensionality reduction techniques have been applied to metabases looking for improvement. Five classifiers methods are investigated as meta-learners in the process of choosing the best parameters of a homogeneous committee.

Towards the prediction of essential genes by integration of network topology, cellular localization and biological process information

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Particle Competition and Cooperation in Networks for Semi-Supervised Learning

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Aprendizado não-supervisionado em redes neurais pulsadas de base radial: um estudo da capacidade de agrupamento para a classificação de pixels

Redes neurais pulsadas - redes que utilizam uma codificação temporal da informação - têm despontado como uma promissora abordagem dentro do paradigma conexionista, emergente da ciência cognitiva. Um desses novos modelos é a rede neural pulsada com função de base radial, que é capaz de armazenar informação nos tempos de atraso axonais dos neurônios. Um algoritmo de aprendizado foi aplicado com sucesso nesta rede pulsada, que se mostrou capaz de mapear uma seqüência de pulsos de entrada em uma seqüência de pulsos de saída. Mais recentemente, um método baseado no uso de campos receptivos gaussianos foi proposto para codificar dados constantes em uma seqüência de pulsos temporais. Este método tornou possível a essa rede lidar com dados computacionais. O processo de aprendizado desta nova rede não se encontra plenamente compreendido e investigações mais profundas são necessárias para situar este modelo dentro do contexto do aprendizado de máquinas e também para estabelecer as habilidades e limitações desta rede. Este trabalho apresenta uma investigação desse novo classificador e um estudo de sua capacidade de agrupar dados em três dimensões, particularmente procurando estabelecer seus domínios de aplicação e horizontes no campo da visão computacional.

Robust and fast vowel recognition using optimum-path forest

The applications of Automatic Vowel Recognition (AVR), which is a sub-part of fundamental importance in most of the speech processing systems, vary from automatic interpretation of spoken language to biometrics. State-of-the-art systems for AVR are based on traditional machine learning models such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs), however, such classifiers can not deal with efficiency and effectiveness at the same time, existing a gap to be explored when real-time processing is required. In this work, we present an algorithm for AVR based on the Optimum-Path Forest (OPF), which is an emergent pattern recognition technique recently introduced in literature. Adopting a supervised training procedure and using speech tags from two public datasets, we observed that OPF has outperformed ANNs, SVMs, plus other classifiers, in terms of training time and accuracy. ©2010 IEEE.