Multi-dimensional classification using Bayesian networks for stationary and evolving streaming data

Hoy en día, con la evolución continua y rápida de las tecnologías de la información y los dispositivos de computación, se recogen y almacenan continuamente grandes volúmenes de datos en distintos dominios y a través de diversas aplicaciones del mundo real. La extracción de conocimiento útil de una cantidad tan enorme de datos no se puede realizar habitualmente de forma manual, y requiere el uso de técnicas adecuadas de aprendizaje automático y de minería de datos. La clasificación es una de las técnicas más importantes que ha sido aplicada con éxito a varias áreas. En general, la clasificación se compone de dos pasos principales: en primer lugar, aprender un modelo de clasificación o clasificador a partir de un conjunto de datos de entrenamiento, y en segundo lugar, clasificar las nuevas instancias de datos utilizando el clasificador aprendido. La clasificación es supervisada cuando todas las etiquetas están presentes en los datos de entrenamiento (es decir, datos completamente etiquetados), semi-supervisada cuando sólo algunas etiquetas son conocidas (es decir, datos parcialmente etiquetados), y no supervisada cuando todas las etiquetas están ausentes en los datos de entrenamiento (es decir, datos no etiquetados). Además, aparte de esta taxonomía, el problema de clasificación se puede categorizar en unidimensional o multidimensional en función del número de variables clase, una o más, respectivamente; o también puede ser categorizado en estacionario o cambiante con el tiempo en función de las características de los datos y de la tasa de cambio subyacente. A lo largo de esta tesis, tratamos el problema de clasificación desde tres perspectivas diferentes, a saber, clasificación supervisada multidimensional estacionaria, clasificación semisupervisada unidimensional cambiante con el tiempo, y clasificación supervisada multidimensional cambiante con el tiempo. Para llevar a cabo esta tarea, hemos usado básicamente los clasificadores Bayesianos como modelos. La primera contribución, dirigiéndose al problema de clasificación supervisada multidimensional estacionaria, se compone de dos nuevos métodos de aprendizaje de clasificadores Bayesianos multidimensionales a partir de datos estacionarios. Los métodos se proponen desde dos puntos de vista diferentes. El primer método, denominado CB-MBC, se basa en una estrategia de envoltura de selección de variables que es voraz y hacia delante, mientras que el segundo, denominado MB-MBC, es una estrategia de filtrado de variables con una aproximación basada en restricciones y en el manto de Markov. Ambos métodos han sido aplicados a dos problemas reales importantes, a saber, la predicción de los inhibidores de la transcriptasa inversa y de la proteasa para el problema de infección por el virus de la inmunodeficiencia humana tipo 1 (HIV-1), y la predicción del European Quality of Life-5 Dimensions (EQ-5D) a partir de los cuestionarios de la enfermedad de Parkinson con 39 ítems (PDQ-39). El estudio experimental incluye comparaciones de CB-MBC y MB-MBC con los métodos del estado del arte de la clasificación multidimensional, así como con métodos comúnmente utilizados para resolver el problema de predicción de la enfermedad de Parkinson, a saber, la regresión logística multinomial, mínimos cuadrados ordinarios, y mínimas desviaciones absolutas censuradas. En ambas aplicaciones, los resultados han sido prometedores con respecto a la precisión de la clasificación, así como en relación al análisis de las estructuras gráficas que identifican interacciones conocidas y novedosas entre las variables. La segunda contribución, referida al problema de clasificación semi-supervisada unidimensional cambiante con el tiempo, consiste en un método nuevo (CPL-DS) para clasificar flujos de datos parcialmente etiquetados. Los flujos de datos difieren de los conjuntos de datos estacionarios en su proceso de generación muy rápido y en su aspecto de cambio de concepto. Es decir, los conceptos aprendidos y/o la distribución subyacente están probablemente cambiando y evolucionando en el tiempo, lo que hace que el modelo de clasificación actual sea obsoleto y deba ser actualizado. CPL-DS utiliza la divergencia de Kullback-Leibler y el método de bootstrapping para cuantificar y detectar tres tipos posibles de cambio: en las predictoras, en la a posteriori de la clase o en ambas. Después, si se detecta cualquier cambio, un nuevo modelo de clasificación se aprende usando el algoritmo EM; si no, el modelo de clasificación actual se mantiene sin modificaciones. CPL-DS es general, ya que puede ser aplicado a varios modelos de clasificación. Usando dos modelos diferentes, el clasificador naive Bayes y la regresión logística, CPL-DS se ha probado con flujos de datos sintéticos y también se ha aplicado al problema real de la detección de código malware, en el cual los nuevos ficheros recibidos deben ser continuamente clasificados en malware o goodware. Los resultados experimentales muestran que nuestro método es efectivo para la detección de diferentes tipos de cambio a partir de los flujos de datos parcialmente etiquetados y también tiene una buena precisión de la clasificación. Finalmente, la tercera contribución, sobre el problema de clasificación supervisada multidimensional cambiante con el tiempo, consiste en dos métodos adaptativos, a saber, Locally Adpative-MB-MBC (LA-MB-MBC) y Globally Adpative-MB-MBC (GA-MB-MBC). Ambos métodos monitorizan el cambio de concepto a lo largo del tiempo utilizando la log-verosimilitud media como métrica y el test de Page-Hinkley. Luego, si se detecta un cambio de concepto, LA-MB-MBC adapta el actual clasificador Bayesiano multidimensional localmente alrededor de cada nodo cambiado, mientras que GA-MB-MBC aprende un nuevo clasificador Bayesiano multidimensional. El estudio experimental realizado usando flujos de datos sintéticos multidimensionales indica los méritos de los métodos adaptativos propuestos. ABSTRACT Nowadays, with the ongoing and rapid evolution of information technology and computing devices, large volumes of data are continuously collected and stored in different domains and through various real-world applications. Extracting useful knowledge from such a huge amount of data usually cannot be performed manually, and requires the use of adequate machine learning and data mining techniques. Classification is one of the most important techniques that has been successfully applied to several areas. Roughly speaking, classification consists of two main steps: first, learn a classification model or classifier from an available training data, and secondly, classify the new incoming unseen data instances using the learned classifier. Classification is supervised when the whole class values are present in the training data (i.e., fully labeled data), semi-supervised when only some class values are known (i.e., partially labeled data), and unsupervised when the whole class values are missing in the training data (i.e., unlabeled data). In addition, besides this taxonomy, the classification problem can be categorized into uni-dimensional or multi-dimensional depending on the number of class variables, one or more, respectively; or can be also categorized into stationary or streaming depending on the characteristics of the data and the rate of change underlying it. Through this thesis, we deal with the classification problem under three different settings, namely, supervised multi-dimensional stationary classification, semi-supervised unidimensional streaming classification, and supervised multi-dimensional streaming classification. To accomplish this task, we basically used Bayesian network classifiers as models. The first contribution, addressing the supervised multi-dimensional stationary classification problem, consists of two new methods for learning multi-dimensional Bayesian network classifiers from stationary data. They are proposed from two different points of view. The first method, named CB-MBC, is based on a wrapper greedy forward selection approach, while the second one, named MB-MBC, is a filter constraint-based approach based on Markov blankets. Both methods are applied to two important real-world problems, namely, the prediction of the human immunodeficiency virus type 1 (HIV-1) reverse transcriptase and protease inhibitors, and the prediction of the European Quality of Life-5 Dimensions (EQ-5D) from 39-item Parkinson’s Disease Questionnaire (PDQ-39). The experimental study includes comparisons of CB-MBC and MB-MBC against state-of-the-art multi-dimensional classification methods, as well as against commonly used methods for solving the Parkinson’s disease prediction problem, namely, multinomial logistic regression, ordinary least squares, and censored least absolute deviations. For both considered case studies, results are promising in terms of classification accuracy as well as regarding the analysis of the learned MBC graphical structures identifying known and novel interactions among variables. The second contribution, addressing the semi-supervised uni-dimensional streaming classification problem, consists of a novel method (CPL-DS) for classifying partially labeled data streams. Data streams differ from the stationary data sets by their highly rapid generation process and their concept-drifting aspect. That is, the learned concepts and/or the underlying distribution are likely changing and evolving over time, which makes the current classification model out-of-date requiring to be updated. CPL-DS uses the Kullback-Leibler divergence and bootstrapping method to quantify and detect three possible kinds of drift: feature, conditional or dual. Then, if any occurs, a new classification model is learned using the expectation-maximization algorithm; otherwise, the current classification model is kept unchanged. CPL-DS is general as it can be applied to several classification models. Using two different models, namely, naive Bayes classifier and logistic regression, CPL-DS is tested with synthetic data streams and applied to the real-world problem of malware detection, where the new received files should be continuously classified into malware or goodware. Experimental results show that our approach is effective for detecting different kinds of drift from partially labeled data streams, as well as having a good classification performance. Finally, the third contribution, addressing the supervised multi-dimensional streaming classification problem, consists of two adaptive methods, namely, Locally Adaptive-MB-MBC (LA-MB-MBC) and Globally Adaptive-MB-MBC (GA-MB-MBC). Both methods monitor the concept drift over time using the average log-likelihood score and the Page-Hinkley test. Then, if a drift is detected, LA-MB-MBC adapts the current multi-dimensional Bayesian network classifier locally around each changed node, whereas GA-MB-MBC learns a new multi-dimensional Bayesian network classifier from scratch. Experimental study carried out using synthetic multi-dimensional data streams shows the merits of both proposed adaptive methods.

Hardware-Based Particle Filter with Evolutionary Resampling Stage

Autonomous systems require, in most of the cases, reasoning and decision-making capabilities. Moreover, the decision process has to occur in real time. Real-time computing means that every situation or event has to have an answer before a temporal deadline. In complex applications, these deadlines are usually in the order of milliseconds or even microseconds if the application is very demanding. In order to comply with these timing requirements, computing tasks have to be performed as fast as possible. The problem arises when computations are no longer simple, but very time-consuming operations. A good example can be found in autonomous navigation systems with visual-tracking submodules where Kalman filtering is the most extended solution. However, in recent years, some interesting new approaches have been developed. Particle filtering, given its more general problem-solving features, has reached an important position in the field. The aim of this thesis is to design, implement and validate a hardware platform that constitutes itself an embedded intelligent system. The proposed system would combine particle filtering and evolutionary computation algorithms to generate intelligent behavior. Traditional approaches to particle filtering or evolutionary computation have been developed in software platforms, including parallel capabilities to some extent. In this work, an additional goal is fully exploiting hardware implementation advantages. By using the computational resources available in a FPGA device, better performance results in terms of computation time are expected. These hardware resources will be in charge of extensive repetitive computations. With this hardware-based implementation, real-time features are also expected.

3D Frangi-based lung vessel enhancement filter penalizing airways

This paper describes a fully automatic simultaneous lung vessel and airway enhancement filter. The approach consists of a Frangi-based multiscale vessel enhancement filtering specifically designed for lung vessel and airway detection, where arteries and veins have high contrast with respect to the lung parenchyma, and airway walls are hollow tubular structures with a non negative response using the classical Frangi's filter. The features extracted from the Hessian matrix are used to detect centerlines and approximate walls of airways, decreasing the filter response in those areas by applying a penalty function to the vesselness measure. We validate the segmentation method in 20 CT scans with different pathological states within the VESSEL12 challenge framework. Results indicate that our approach obtains good results, decreasing the number of false positives in airway walls.

Adaptive spatio-temporal filter for low-cost camera depth maps

In this paper we present an adaptive spatio-temporal filter that aims to improve low-cost depth camera accuracy and stability over time. The proposed system is composed by three blocks that are used to build a reliable depth map of static scenes. An adaptive joint-bilateral filter is used to obtain consistent depth maps by jointly considering depth and video information and by adapting its parameters to different levels of estimated noise. Kalman filters are used to reduce the temporal random fluctuations of the measurements. Finally an interpolation algorithm is used to obtain consistent depth maps in the regions where the depth information is not available. Results show that this approach allows to considerably improve the depth maps quality by considering spatio-temporal information and by adapting its parameters to different levels of noise.

Design of a Two-Phase buck converter with fourth-order output filter for envelope amplifiers of limited bandwidth

The use of techniques such as envelope tracking (ET) and envelope elimination and restoration (EER) can improve the efficiency of radio frequency power amplifiers (RFPA). In both cases, high-bandwidth DC/DC converters called envelope amplifiers (EA) are used to modulate the supply voltage of the RFPA. This paper addresses the analysis and design of a modified two-phase Buck converter optimized to operate as EA. The effects of multiphase operation on the tracking capabilities are analyzed. The use of a fourth-order output filter is proposed to increase the attenuation of the harmonics generated by the PWM operation, thus allowing a reduction of the ratio between the switching frequency and the converter bandwidth. The design of the output filter is addressed considering envelope tracking accuracy and distortion caused by the side bands arising from the nonlinear modulation process. Finally, the proposed analysis and design methods are supported by simulation results, as well as demonstrated by experiments obtained using two 100-W, 10-MHz, two-phase Buck EAs capable of accurately tracking a 1.5-MHz bandwidth OFDM signal.

Extended phone log-likelihood ratio features and acoustic-based I-vectors for language recognition

This paper presents new techniques with relevant improvements added to the primary system presented by our group to the Albayzin 2012 LRE competition, where the use of any additional corpora for training or optimizing the models was forbidden. In this work, we present the incorporation of an additional phonotactic subsystem based on the use of phone log-likelihood ratio features (PLLR) extracted from different phonotactic recognizers that contributes to improve the accuracy of the system in a 21.4% in terms of Cavg (we also present results for the official metric during the evaluation, Fact). We will present how using these features at the phone state level provides significant improvements, when used together with dimensionality reduction techniques, especially PCA. We have also experimented with applying alternative SDC-like configurations on these PLLR features with additional improvements. Also, we will describe some modifications to the MFCC-based acoustic i-vector system which have also contributed to additional improvements. The final fused system outperformed the baseline in 27.4% in Cavg.

University employees work schedule

The aim of this project is to create a website which is useful both employees and students of a university, so employees can add information, if they log in with username and password access, and students can view this information . Employees may modify and display information such as title, room, or their faculty (from a list defined by the administrator), and most importantly, their schedule, whether class, tutoring, free time, or any of tasks that the administrator define. There will be a manager, responsible for managing employees, the availables faculties and the types of tasks that employees can use on their schedule. Students may see the employees schedules and rooms on the homepage. They differentiate between differents tasks of employees, because these are in different colors. They can also filter information for faculty, employee or day. To achieve our goal, we decided to program in Java using Servlets, which we will use to generate response to requests coming from users from the website. We will also use JSP, allowing us to create different websites files. We use JSP files and not HTML, because we need that the pages are dynamic, since not only want to show specific information, we like that information can change depending on user requests. The JSP file allows us to generate HTML, but also using JAVA language, which is necessary for our purpose. As the information we store is not fixed. We want this information can be modified at any time by employees and admin, so we need a database, which can be accessed from anywhere. We decided SQLite databases because are integrated quite well in our application, and offer a quick response. To access the database from our program, we simply connect it to the database, and with very few lines of code, add, delete or modify entries in different tables that owns the database. To facilitate the initial creation of the database, and the first tables, we use a Mozilla Firefox browser plugin, called SQLite Manager, which allows us to do so from a more friendly interface. Finally, we need a server that supports and implements specifications Servlets and JSP. We decided on the TomCat server, which is a container Servlets, because is free, easy to use, and compatible with our program. We realized all the project with Eclipse environment, also free program that allows integrating database, server and program the JSP and Servlets. Once submitted all the tools we used, we must first organize the structure of the web, relating each Servlets with JSP files. Next, create the database and the different Servlets, and adjust the database accesses to make sure we do it right. From here simply is to build up the page step by step, showing in each place we need, and redirect to different pages. In this way, we can build a complex website, free, and without being an expert in the field. RESUMEN. El objetivo de este proyecto, es crear una página web que sirva tanto a empleados como a alumnos de una universidad, de tal manera que los empleados podrán añadir información, mediante el acceso con usuario y contraseña, y los alumnos podrán visualizar está información. Los empleados podrán modificar y mostrar información como su título, despacho, facultad a la que pertenecen (de entre una lista definida por el administrador), y lo más importante, sus horarios, ya sean de clase, tutorías, tiempo libre, o cualquiera de las tareas que el administrador defina. Habrá un administrador, encargado de gestionar los empleados existentes, las facultades disponibles y los tipos de tareas que podrán usar los empleados en su horario. Los alumnos, podrán visualizar los horarios y despacho de los empleados en la página principal. Diferenciarán entre las distintas tareas de los profesores, porque estas se encuentran en colores diferentes. Además, podrán filtrar la información, por facultad, empleado o día de la semana. Para conseguir nuestro objetivo, hemos decidido programar en Java, mediante el uso de Servlets, los cuales usaremos para generar respuesta antes las peticiones que llegan de los usuarios desde la página web. También usaremos archivos JSP, que nos permitirán crear las diferentes páginas webs. Usamos archivos JSP y no HTML, porque necesitamos que las diferentes páginas sean dinámicas, ya que no solo queremos mostrar una información concreta, si no que esta información puede variar en función de las peticiones de usuario. El archivo JSP nos permite generar HTML, pero a la vez usar lenguaje JAVA, algo necesario para nuestro cometido. Como la información que queremos almacenar no es fija, si no que en todo momento debe poder ser modificada por empleados y administrador, necesitamos una base de datos, a la que podamos acceder desde la web. Nos hemos decidido por bases SQLite, ya que se integran bastante bien en nuestra aplicación, y además ofrecen una rápida respuesta. Para acceder a la base de datos desde nuestro programa, simplemente debemos conectar el mismo a la base de datos, y con muy pocas líneas de código, añadir, eliminar o modificar entradas de las diferentes tablas que posee la base de datos. Para facilitar la creación inicial de la base de datos, y de las primeras tablas, usamos un complemento del navegador Mozilla Firefox, llamado SQLite Manager, que nos permite hacerlo desde una interfaz más amigable. Por último, necesitamos de un servidor que soporte e implemente las especificaciones de los Servlets y JSP. Nos decidimos por el servidor TomCat, que es un contenedor de Servlets gratuito, de fácil manejo, y compatible con nuestro programa. Todo el desarrollo del proyecto, lo realizamos desde el entorno Eclipse, programa también gratuito, que permite integrar la base de datos, el servidor y programar los JSP y Servlets. Una vez presentadas todas las herramientas que hemos utilizado, primero debemos organizar la estructura de la web, relacionando cada archivo JSP con los Servlets a los que debe acceder. A continuación creamos la base de datos y los diferentes Servlets, y ajustamos bien los accesos a la base de datos para comprobar que lo hacemos correctamente. A partir de aquí, simplemente es ir construyendo la página paso a paso, mostrando en cada lugar lo que necesitemos, y redirigiendo a las diferentes páginas. De esta manera, podremos construir una página web compleja, de manera gratuita, y sin ser un experto en la materia.

Sodium channel selectivity filter regulates antiarrhythmic drug binding

Local anesthetic antiarrhythmic drugs block Na+ channels and have important clinical uses. However, the molecular mechanism by which these drugs block the channel has not been established. The family of drugs is characterized by having an ionizable amino group and a hydrophobic tail. We hypothesized that the charged amino group of the drug may interact with charged residues in the channel’s selectivity filter. Mutation of the putative domain III selectivity filter residue of the adult rat skeletal muscle Na+ channel (μ1) K1237E increased resting lidocaine block, but no change was observed in block by neutral analogs of lidocaine. An intermediate effect on the lidocaine block resulted from K1237S and there was no effect from K1237R, implying an electrostatic effect of Lys. Mutation of the other selectivity residues, D400A (domain I), E755A (domain II), and A1529D (domain IV) allowed block by externally applied quaternary membrane-impermeant derivatives of lidocaine (QX314 and QX222) and accelerated recovery from block by internal QX314. Neo-saxitoxin and tetrodotoxin, which occlude the channel pore, reduced the amount of QX314 bound in D400A and A1529D, respectively. Block by outside QX314 in E755A was inhibited by mutation of residues in transmembrane segment S6 of domain IV that are thought to be part of an internal binding site. The results demonstrate that the Na+ channel selectivity filter is involved in interactions with the hydrophilic part of the drugs, and it normally limits extracellular access to and escape from their binding site just within the selectivity filter. Participation of the selectivity ring in antiarrhythmic drug binding and access locates this structure adjacent to the S6 segment.

Isolation of anti-angiogenesis antibodies from a large combinatorial repertoire by colony filter screening

We describe here a method, based on iterative colony filter screening, for the rapid isolation of binding specificities from a large synthetic repertoire of human antibody fragments in single-chain Fv configuration. Escherichia coli cells, expressing the library of antibody fragments, are grown on a porous master filter, in contact with a second filter coated with the antigen, onto which antibodies secreted by the bacteria are able to diffuse. Detection of antigen binding on the second filter allows the recovery of a number of E.coli cells, including those expressing the binding specificity of interest, which can be submitted to a second round of screening for the isolation of specific monoclonal antibodies. We tested the methodology using as antigen the ED-B domain of fibronectin, a marker of angiogenesis. From an antibody library of 7 × 108 clones, we recovered a number of specifically-binding antibodies of different aminoacid sequence. The antibody clone showing the strongest enzyme-linked immunosorbent assay signal (ME4C) was further characterised. Its epitope on the ED-B domain was mapped using the SPOT synthesis method, which uses a set of decapeptides spanning the antigen sequence synthesised and anchored on cellulose. ME4C binds to the ED-B domain with a dissociation constant Kd = 1 × 10–7 M and specifically stains tumour blood vessels, as shown by immunohistochemical analysis on tumour sections of human and murine origin.

An O(N log N) algorithm for shape modeling.

We present a shape-recovery technique in two dimensions and three dimensions with specific applications in modeling anatomical shapes from medical images. This algorithm models extremely corrugated structures like the brain, is topologically adaptable, and runs in O(N log N) time, where N is the total number of points in the domain. Our technique is based on a level set shape-recovery scheme recently introduced by the authors and the fast marching method for computing solutions to static Hamilton-Jacobi equations.

Quantitative analysis of senile plaques in Alzheimer disease: observation of log-normal size distribution and molecular epidemiology of differences associated with apolipoprotein E genotype and trisomy 21 (Down syndrome).

The discovery that the epsilon 4 allele of the apolipoprotein E (apoE) gene is a putative risk factor for Alzheimer disease (AD) in the general population has highlighted the role of genetic influences in this extremely common and disabling illness. It has long been recognized that another genetic abnormality, trisomy 21 (Down syndrome), is associated with early and severe development of AD neuropathological lesions. It remains a challenge, however, to understand how these facts relate to the pathological changes in the brains of AD patients. We used computerized image analysis to examine the size distribution of one of the characteristic neuropathological lesions in AD, deposits of A beta peptide in senile plaques (SPs). Surprisingly, we find that a log-normal distribution fits the SP size distribution quite well, motivating a porous model of SP morphogenesis. We then analyzed SP size distribution curves in genotypically defined subgroups of AD patients. The data demonstrate that both apoE epsilon 4/AD and trisomy 21/AD lead to increased amyloid deposition, but by apparently different mechanisms. The size distribution curve is shifted toward larger plaques in trisomy 21/AD, probably reflecting increased A beta production. In apoE epsilon 4/AD, the size distribution is unchanged but the number of SP is increased compared to apoE epsilon 3, suggesting increased probability of SP initiation. These results demonstrate that subgroups of AD patients defined on the basis of molecular characteristics have quantitatively different neuropathological phenotypes.

Joint interpretation of magnetotelluric, seismic, and well-log data in Hontomín (Spain)

Acknowledgements. This work is dedicated to the memory of Andrés Pérez-Estaún, brilliant scientist, colleague, and friend. The authors sincerely thank Ian Ferguson and an anonymous reviewer for their useful comments on the manuscript. Xènia Ogaya is currently supported in the Dublin Institute for Advanced Studies by a Science Foundation Ireland grant IRECCSEM (SFI grant 12/IP/1313). Juan Alcalde is funded by NERC grant NE/M007251/1, on interpretational uncertainty. Juanjo Ledo, Pilar Queralt and Alex Marcuello thank Ministerio de Economía y Competitividad and EU Feder Funds through grant CGL2014- 54118-C2-1-R. Funding for this Project has been partially provided by the Spanish Ministry of Industry, Tourism and Trade, through the CIUDEN-CSIC-Inst. Jaume Almera agreement (ALM-09-027: Characterization, Development and Validation of Seismic Techniques applied to CO2 Geological Storage Sites), the CIUDEN-Fundació Bosch i Gimpera agreement (ALM-09-009 Development and Adaptation of Electromagnetic techniques: Characterisation of Storage Sites) and the project PIERCO2 (Progress In Electromagnetic Research for CO2 geological reservoirs CGL2009-07604). The CIUDEN project is co-financed by the European Union through the Technological Development Plant of Compostilla OXYCFB300 Project (European Energy Programme for Recovery).

A nova família de distribuições odd log-logística: teoria e aplicações

Neste trabalho, foi proposta uma nova família de distribuições, a qual permite modelar dados de sobrevivência quando a função de risco tem formas unimodal e U (banheira). Ainda, foram consideradas as modificações das distribuições Weibull, Fréchet, half-normal generalizada, log-logística e lognormal. Tomando dados não-censurados e censurados, considerou-se os estimadores de máxima verossimilhança para o modelo proposto, a fim de verificar a flexibilidade da nova família. Além disso, um modelo de regressão locação-escala foi utilizado para verificar a influência de covariáveis nos tempos de sobrevida. Adicionalmente, conduziu-se uma análise de resíduos baseada nos resíduos deviance modificada. Estudos de simulação, utilizando-se de diferentes atribuições dos parâmetros, porcentagens de censura e tamanhos amostrais, foram conduzidos com o objetivo de verificar a distribuição empírica dos resíduos tipo martingale e deviance modificada. Para detectar observações influentes, foram utilizadas medidas de influência local, que são medidas de diagnóstico baseadas em pequenas perturbações nos dados ou no modelo proposto. Podem ocorrer situações em que a suposição de independência entre os tempos de falha e censura não seja válida. Assim, outro objetivo desse trabalho é considerar o mecanismo de censura informativa, baseado na verossimilhança marginal, considerando a distribuição log-odd log-logística Weibull na modelagem. Por fim, as metodologias descritas são aplicadas a conjuntos de dados reais.

Filtro respiratório reduz efeitos cardiovasculares associados à poluição: estudo randomizado, duplo-cego, controlado e cruzado em pacientes com insuficiência cardíaca (FILTER-HF trial)

Introdução A poluição do ar é um fator de risco associado com descompensação e mortalidade em pacientes com insuficiência cardíaca (IC). Objetivo Avaliar o impacto de um filtro de polipropileno sobre desfechos cardiovasculares em pacientes com IC e voluntários saudáveis durante exposição controlada à poluição. Métodos Ensaio clínico duplocego, controlado e cruzado, incluindo 26 pacientes com IC e 15 voluntários saudáveis, expostos a três protocolos diferentes de inalação randomizados por ordem: Ar Limpo; Exposição à Partículas de Exaustão do Diesel (ED); e ED filtrada. Os desfechos estudados foram função endotelial por índice de hiperemia reativa (RHi) e índice de aumento (Aix), biomarcadores séricos, variáveis de teste cardiopulmonar submáximo (caminhada de seis-minutos [tc6m]; consumo de oxigênio [VO2]; equivalente ventilatório de gás carbônico [VE/VCO2 slope]; consumo de O2 por batida [PulsoO2]) e variabilidade da frequência cardíaca (VFC). Resultados No grupo IC, a ED piorou o RHi [de 2,17 (IQR: 1,8-2,5) para 1,72 (IQR: 1,5-2,2); p=0,002], reduziu o VO2 [de 11.0 ± 3.9 para 8.4±2.8ml/Kg/min; p < 0.001], o tc6m [de 243,3±13 para 220,8 ± 14m; p=0,030] e o PulsoO2 [de 8.9 ± 1.0 para 7.8±0.7ml/bpm; p < 0.001]; e aumentou o BNP [de 47,0pg/ml (IQR: 17,3-118,0) para 66,5pg/ml (IQR: 26,5-155,5); p=0,004]. O filtro foi capaz de reduzir a concentração de poluição de 325±31 para 25±6?g/m3 (p < 0,001 vs. ED). No grupo IC, o filtro foi associado com melhora no RHi [2,06 (IQR: 1,5-2,6); p=0,019 vs. ED); aumento no VO2 (10.4 ± 3.8ml/Kg/min; p < 0.001 vs. ED) e PulsoO2 (9.7±1.1ml/bpm; p < 0.001 vs. ED); e redução no BNP [44,0pg/ml (IQR: 20,0-110,0); p=0,015 vs. ED]. Em ambos os grupos, a ED reduziu o Aix, sem efeito do filtro. O uso do filtro foi associado com maior ventilação e reinalação de CO2. Outras variáveis pesquisadas como VE/VCO2 slope e VFC não sofreram influências entre os protocolos. Conclusão A poluição do ar afetou adversamente o desempenho cardiovascular de pacientes com IC. Este é o primeiro ensaio clínico demonstrando que um simples filtrorespiratório pode prevenir a disfunção endotelial, a intolerância ao exercício e o aumento do BNP associados à poluição em pacientes com IC. O uso de máscaras com filtro tem o potencial de reduzir a morbidade associada à IC. Identificador ClinicalTrials.gov: NCT01960920