No description available.

Made available in DSpace on 2014-12-17T14:54:49Z (GMT). No. of bitstreams: 1 HelianaBS_da_capa_ate_cap4.pdf: 2361373 bytes, checksum: 3e1e43e8ba1aadc274663b8b8e3de72f (MD5) Previous issue date: 2008-02-22 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / The skin cancer is the most common of all cancers and the increase of its incidence must, in part, caused by the behavior of the people in relation to the exposition to the sun. In Brazil, the non-melanoma skin cancer is the most incident in the majority of the regions. The dermatoscopy and videodermatoscopy are the main types of examinations for the diagnosis of dermatological illnesses of the skin. The field that involves the use of computational tools to help or follow medical diagnosis in dermatological injuries is seen as very recent. Some methods had been proposed for automatic classification of pathology of the skin using images. The present work has the objective to present a new intelligent methodology for analysis and classification of skin cancer images, based on the techniques of digital processing of images for extraction of color characteristics, forms and texture, using Wavelet Packet Transform (WPT) and learning techniques called Support Vector Machine (SVM). The Wavelet Packet Transform is applied for extraction of texture characteristics in the images. The WPT consists of a set of base functions that represents the image in different bands of frequency, each one with distinct resolutions corresponding to each scale. Moreover, the characteristics of color of the injury are also computed that are dependants of a visual context, influenced for the existing colors in its surround, and the attributes of form through the Fourier describers. The Support Vector Machine is used for the classification task, which is based on the minimization principles of the structural risk, coming from the statistical learning theory. The SVM has the objective to construct optimum hyperplanes that represent the separation between classes. The generated hyperplane is determined by a subset of the classes, called support vectors. For the used database in this work, the results had revealed a good performance getting a global rightness of 92,73% for melanoma, and 86% for non-melanoma and benign injuries. The extracted describers and the SVM classifier became a method capable to recognize and to classify the analyzed skin injuries / O c?ncer de pele ? o mais comum de todos os c?nceres e o aumento da sua incid?ncia deve-se, em parte, ao comportamento das pessoas em rela??o ? exposi??o ao sol. No Brasil, o c?ncer de pele n?o melanoma ? o mais incidente na maioria das regi?es. A dermatoscopia e ideodermatoscopia s?o os principais tipos de exames para o diagn?stico de doen?as da pele dermatol?gicas. O campo que envolve o uso de ferramentas computacionais para o aux?lio ou acompanhamento do diagn?stico m?dico em les?es dermatol?gicas ainda ? visto como muito recente. V?rios m?todos foram propostos para classifica??o autom?tica de patologias da pele utilizando imagens. O presente trabalho tem como objetivo apresentar uma nova metodologia inteligente para an?lise e classifica??o de imagens de c?ncer de pele, baseada nas t?cnicas de processamento digital de imagens para extra??o de caracter?sticas de cor, forma e textura, utilizando a Transformada Wavelet Packet (TWP) e a t?cnicas de aprendizado de m?quina denominada M?quina de Vetor de Suporte (SVM Support Vector Machine). A Transformada Wavelet Packet ? aplicada para extra??o de caracter?sticas de textura nas imagens. Esta consiste de um conjunto de fun??es base que representa a imagem em diferentes bandas de freq??ncia, cada uma com resolu??es distintas correspondente a cada escala. Al?m disso, s?o calculadas tamb?m as caracter?sticas de cor da les?o que s?o dependentes de um contexto visual, influenciada pelas cores existentes em sua volta, e os atributos de forma atrav?s dos descritores de Fourier. Para a tarefa de classifica??o ? utilizado a M?quina de Vetor de Suporte, que baseia-se nos princ?pios da minimiza??o do risco estrutural, proveniente da teoria do aprendizado estat?stico. A SVM tem como objetivo construir hiperplanos ?timos que apresentem a maior margem de separa??o entre classes. O hiperplano gerado ? determinado por um subconjunto dos pontos das classes, chamado vetores de suporte. Para o banco de dados utilizado neste trabalho, os resultados apresentaram um bom desempenho obtendo um acerto global de 92,73% para melanoma, e 86% para les?es n?o-melanoma e benigna. O potencial dos descritores extra?dos aliados ao classificador SVM tornou o m?todo capaz de reconhecer e classificar as les?es analisadas

Tradicionalmente, la detección de faltas en máquinas eléctricas se basa en el uso de la Transformada Rápida de Fourier ya que la mayoría de las faltas pueden ser diagnosticadas con ella con seguridad si las máquinas operan en condiciones de régimen estacionario durante un intervalo de tiempo razonable. Sin embargo, para aplicaciones en las que las máquinas operan en condiciones de carga y velocidad fluctuantes (condiciones no estacionarias) como por ejemplo los aerogeneradores, el uso de la Transformada Rápida de Fourier debe ser reemplazado por otras técnicas. La presente tesis desarrolla una nueva metodología para el diagnóstico de máquinas de inducción de rotor de jaula y rotor bobinado operando en condiciones no estacionarias, basada en el análisis de las componentes de falta de las corrientes en el plano deslizamiento frecuencia. La técnica es aplicada al diagnóstico de asimetrías estatóricas, rotóricas y también para la falta de excentricidad mixta. El diagnóstico de las máquinas eléctricas en el dominio deslizamiento-frecuencia confiere un carácter universal a la metodología ya que puede diagnosticar máquinas eléctricas independientemente de sus características, del modo en el que la velocidad de la máquina varía y de su modo de funcionamiento (motor o generador). El desarrollo de la metodología conlleva las siguientes etapas: (i) Caracterización de las evoluciones de las componentes de falta de asimetría estatórica, rotórica y excentricidad mixta para las máquinas de inducción de rotores de jaula y bobinados en función de la velocidad (deslizamiento) y la frecuencia de alimentación de la red a la que está conectada la máquina. (ii) Debido a la importancia del procesado de la señal, se realiza una introducción a los conceptos básicos del procesado de señal antes de centrarse en las técnicas actuales de procesado de señal para el diagnóstico de máquinas eléctricas. (iii) La extracción de las componentes de falta se lleva a cabo a través de tres técnicas de filtrado diferentes: filtros basados en la Transformada Discreta Wavelet, en la Transformada Wavelet Packet y con una nueva técnica de filtrado propuesta en esta tesis, el Filtrado Espectral. Las dos primeras técnicas de filtrado extraen las componentes de falta en el dominio del tiempo mientras que la nueva técnica de filtrado realiza la extracción en el dominio de la frecuencia. (iv) La extracción de las componentes de falta, en algunos casos, conlleva el desplazamiento de la frecuencia de las componentes de falta. El desplazamiento de la frecuencia se realiza a través de dos técnicas: el Teorema del Desplazamiento de la Frecuencia y la Transformada Hilbert. (v) A diferencia de otras técnicas ya desarrolladas, la metodología propuesta no se basa exclusivamente en el cálculo de la energía de la componente de falta sino que también estudia la evolución de la frecuencia instantánea de ellas, calculándola a través de dos técnicas diferentes (la Transformada Hilbert y el operador Teager-Kaiser), frente al deslizamiento. La representación de la frecuencia instantánea frente al deslizamiento elimina la posibilidad de diagnósticos falsos positivos mejorando la precisión y la calidad del diagnóstico. Además, la representación de la frecuencia instantánea frente al deslizamiento permite realizar diagnósticos cualitativos que son rápidos y requieren bajos requisitos computacionales. (vi) Finalmente, debido a la importancia de la automatización de los procesos industriales y para evitar la posible divergencia presente en el diagnóstico cualitativo, tres parámetros objetivos de diagnóstico son desarrollados: el parámetro de la energía, el coeficiente de similitud y los parámetros de regresión. El parámetro de la energía cuantifica la severidad de la falta según su valor y es calculado en el dominio del tiempo y en el dominio de la frecuencia (consecuencia de la extracción de las componentes de falta en el dominio de la frecuencia). El coeficiente de similitud y los parámetros de regresión son parámetros objetivos que permiten descartar diagnósticos falsos positivos aumentando la robustez de la metodología propuesta. La metodología de diagnóstico propuesta se valida experimentalmente para las faltas de asimetría estatórica y rotórica y para el fallo de excentricidad mixta en máquinas de inducción de rotor de jaula y rotor bobinado alimentadas desde la red eléctrica y desde convertidores de frecuencia en condiciones no estacionarias estocásticas. / Vedreño Santos, FJ. (2013). Diagnosis of electric induction machines in non-stationary regimes working in randomly changing conditions [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/34177 / TESIS

This work deals with the possibilities of filtering the ECG signal, representing the first part, which is the basis for successful delineation and follow diagnosis of the ECG signal. Filtration in this case is mean to suppress interference from electrical grid, noise of electrical grid. The content of the work is description of filters realized trough wavelet transform and linear filtering as a means to successful filtration of interference. There are method of stationary wavelet transform - dyadic wavelet transform, wavelet packet transform and wavelet wiener filtering method. Linear filtering includes two narrow-band FIR filters. The objective of this work is to propose different methods of wavelet and linear filters in Matlab, filtering of ECG signals and compare the success of filtration methods. ECG signals used in this work are from the CSE database.

Global rise in life expectancy has resulted in an increased demand for affordable healthcare and monitoring services. The advent of miniature and low–power sensor technologies coupled with the emergence of the Internet–of–Things has paved the way towards affordable health monitoring tools in wearable platforms. However, ensuring power–efficient operation, data accuracy and user comfort are critical for such wearable systems. This thesis focuses on the development of accurate and computationally efficient algorithms and low–cost, unobtrusive devices with potential predictive capability for monitoring mobility and cardiac health in a wearable platform. A three–stage complementary filter–based approach is developed to realize a computationally efficient method to estimate sensor orientation in real–time. A gradient descent–based approach is used to estimate the gyroscope integration drift, which is subsequently subtracted from the integrated gyroscope data to get the sensor orientation. This predominantly gyroscope–based orientation estimation approach is least affected by external acceleration and magnetic disturbances. A two–stage complementary filter–based efficient sensor fusion algorithm is developed for real–time monitoring of lower–limb joints that estimates the IMU inclinations in the first stage and uses a gradient descent–based approach in the second stage to estimate the joint angles. The proposed method estimates joint angles primarily from the gyroscope measurements without incorporating the magnetic field measurement, rendering the estimated angles least affected by any external acceleration and insensitive to magnetic disturbances. An IMU–based simple, low–cost and computationally efficient gait–analyzer is developed to track the course of an individual's gait health in a continuous fashion. Continuous monitoring of gait patterns can potentially enable detecting musculoskeletal or neurodegenerative diseases at the early onset. The proposed gait analyzer identifies an anomalous gait with moderate to high accuracy by evaluating the gait features with respect to the baseline clusters corresponding to an individual’s healthy peer group. The adoption of a computationally efficient signal analysis technique renders the analyzer suitable for systems with limited processing capabilities. A flexible dry capacitive electrode and a wireless ECG monitoring system with automatic anomaly detection capability are developed. The flexible capacitive electrode reduces motion artifacts and enables sensing bio–potential over a dielectric material such as cotton cloth. The virtual ground of the electrode allows for obtaining single–lead ECG using two electrodes only. ECG measurements obtained over different types of textile materials and in presence of body movements show comparable performance to other reported ECG monitoring systems. An algorithm is developed separately as a potential extension of the software to realize automatic identification of Atrial Fibrillation from short single–lead ECGs. The association between human gait and cardiac activities is studied. The gait is measured using wearable IMUs and the cardiac activity is measured with a single–lead handheld ECG monitor. Some key cardiac parameters, such as heart rate and heart rate variability and physical parameters, such as age and BMI show good association with gait asymmetry and gait variation. These associations between gait and heart can be useful in realizing low–cost in–home personal monitoring tool for early detecting CVD–related changes in gait features before the CVD symptoms are manifested. / Thesis / Doctor of Philosophy (PhD) / Wearable health monitoring systems can be a viable solution to meet the increased demand for affordable healthcare and monitoring services. However, such systems need to be energy–efficient, accurate and ergonomic to enable long–term monitoring of health reliably while preserving user comfort. In this thesis, we develop efficient algorithms to obtain real–time estimates of on–body sensors' orientation, gait parameters such as stride length, and gait velocity and lower–limb joint angles. Furthermore, we develop a simple, low–cost and computationally efficient gait–analyzer using miniature and low–power inertial motion units to track the health of human gait in a continuous fashion. In addition, we design flexible, dry capacitive electrodes and use them to develop a portable single–lead electrocardiogram (ECG) device. The flexible design ensures better conformity of the electrode to the skin, resulting in better signal quality. The capacitive nature allows for obtaining ECG signals over insulating materials such as cloth, thereby potentially enabling a comfortable means of long–term cardiac health monitoring at home. Besides, we implement an automatic anomaly detection algorithm that detects Atrial Fibrillation with good accuracy from short single–lead ECGs. Finally, we investigate the association between gait and cardiac activities. We observe that some important cardiac signs, such as heart rate and heart rate variability and physical parameters, such as age and BMI show good association with gait asymmetry and gait variation.