Submitted by FABRÍCIO ELY GOSSLER null (fabricio_ely8@hotmail.com) on 2017-02-09T16:24:59Z No. of bitstreams: 1 Fabrício E. Gossler-Dissertação - Unesp - Feis-PPGEE.pdf: 5023440 bytes, checksum: b5346eb35f509f2283b503acccf22ec3 (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-02-14T16:08:30Z (GMT) No. of bitstreams: 1 gossler_fe_me_ilha.pdf: 5023440 bytes, checksum: b5346eb35f509f2283b503acccf22ec3 (MD5) / Made available in DSpace on 2017-02-14T16:08:30Z (GMT). No. of bitstreams: 1 gossler_fe_me_ilha.pdf: 5023440 bytes, checksum: b5346eb35f509f2283b503acccf22ec3 (MD5) Previous issue date: 2016-12-19 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Existem diferentes tipos de funções wavelets que podem ser utilizadas na Transformada Wavelet. Na maioria das vezes, a função wavelet escolhida para a análise de um determinado sinal vai ser aquela que melhor se ajusta no domínio tempo-frequência do mesmo. Existem vários tipos de funções wavelets que podem ser escolhidas para certas aplicações, sendo que algumas destas pertencem a conjuntos específicos denominados de famílias wavelets, tais como a Haar, Daubechies, Symlets, Morlet, Meyer e Gaussianas. Nesse trabalho é apresentada uma nova família de funções wavelets geradas a partir de polinômios com coeficientes de Fibonacci (FCPs). Essa família recebe o nome de Golden, e cada membro desta é obtido por uma derivada de ordem n do quociente entre dois FCPs distintos. As Golden wavelets foram deduzidas através das observações de que, em alguns casos, a derivada de ordem n, do quociente entre dois FCPs distintos, resulta em uma função que possui as características de uma onda de duração curta. Como aplicação, algumas wavelets apresentadas no decorrer deste trabalho são utilizadas na classificação de arritmias cardíacas em sinais de eletrocardiograma, que foram extraídos da base de dados do MIT-BIH arrhythmia database. / There exist different types of wavelet functions that can be used in the Wavelet Transform. In most cases, the wavelet function chosen for the analysis of a given signal will be the one that best adjusts in the time-frequency domain of the same signal. There are many types of wavelet functions that can be chosen for certain applications, some of which belong to specific sets called wavelet families, such as Haar, Daubechies, Symlets, Morlet, Meyer, and Gaussians. In this work a new wavelet functions family generated from Fibonacci-coefficients polynomials (FCPs) is presented. This family is called Golden, and each member is obtained by the n-th derivative of the quotient between two distinct FCPs. The Golden wavelets were deduced from the observations that in some cases the n-th derivative of the quotient between two distinct FCPs results in a function that has the characteristics of a short-duration wave. As an application, some wavelets presented in the course of this work are used to cardiac arrhythmia classification in electrocardiogram signals, which were extracted from the MITBIH arrhythmia database. / CNPq: 130123/2015-3

Orientador: Francisco Villarreal Alvarado / Resumo: Existem diferentes tipos de funções wavelets que podem ser utilizadas na Transformada Wavelet. Na maioria das vezes, a função wavelet escolhida para a análise de um determinado sinal vai ser aquela que melhor se ajusta no domínio tempo-frequência do mesmo. Existem vários tipos de funções wavelets que podem ser escolhidas para certas aplicações, sendo que algumas destas pertencem a conjuntos específicos denominados de famílias wavelets, tais como a Haar, Daubechies, Symlets, Morlet, Meyer e Gaussianas. Nesse trabalho é apresentada uma nova família de funções wavelets geradas a partir de polinômios com coeficientes de Fibonacci (FCPs). Essa família recebe o nome de Golden, e cada membro desta é obtido por uma derivada de ordem n do quociente entre dois FCPs distintos. As Golden wavelets foram deduzidas através das observações de que, em alguns casos, a derivada de ordem n, do quociente entre dois FCPs distintos, resulta em uma função que possui as características de uma onda de duração curta. Como aplicação, algumas wavelets apresentadas no decorrer deste trabalho são utilizadas na classificação de arritmias cardíacas em sinais de eletrocardiograma, que foram extraídos da base de dados do MIT-BIH arrhythmia database. / Mestre

<p dir="ltr">Cardiac disease is one of the prominent reasons of deaths worldwide. The timely de-<br>tection of arrhythmias, one of the highly prevalent cardiac abnormalities, is very important<br>and promising for treatment. Electrocardiography (ECG) is well applied to probe the car-<br>diac dynamics, nevertheless, it is still challenging to robustly detect the arrhythmia with<br>automatic algorithms, especially when the noise may contaminate the signal to some extent.<br>In this research study, we have not only built and assessed different neural network models<br>to understand their capability in terms of ECE-based arrhythmia detection, but also com-<br>prehensively investigated the detection under different kinds of signal-to-noise ratio (SNR).<br>Both Long Short-Term Memory (LSTM) model and Multi-Layer Perception (MLP) model<br>have been developed in the study. Further, we have studied the necessity of fine-tuning<br>of the neural network models, which are pre-trained on other data and demonstrated that<br>it is very important to boost the performance when ECG is contaminated by noise. In<br>the experiments, the LSTM model achieves an accuracy of 99.0%, F1 score of 97.9%, and<br>high precision and recall, with the clean ECE signal. Further, in the high SNR scenario,<br>the LSTM maintains an attractive performance. With the low SNR scenario, though there<br>is some performance drop, the fine-tuning approach helps performance improvement criti-<br>cally. Overall, this study has built the neural network models, and investigated different<br>kinds of signal fidelity including clean, high-SNR, and low-SNR, towards robust arrhythmia<br>detection.</p>

Over the last century, physiological signals have been broadly analyzed and processed not only to assess the function of the human physiology, but also to better diagnose illnesses or injuries and provide treatment options for patients. In particular, Electrocardiogram (ECG), blood pressure (BP) and impedance are among the most important biomedical signals processed and analyzed. The majority of studies that utilize these signals attempt to diagnose important irregularities such as arrhythmia or blood loss by processing one of these signals. However, the relationship between them is not yet fully studied using computational methods. Therefore, a system that extract and combine features from all physiological signals representative of states such as arrhythmia and loss of blood volume to predict the presence and the severity of such complications is of paramount importance for care givers. This will not only enhance diagnostic methods, but also enable physicians to make more accurate decisions; thereby the overall quality of care provided to patients will improve significantly. In the first part of the dissertation, analysis and processing of ECG signal to detect the most important waves i.e. P, QRS, and T, are described. A wavelet-based method is implemented to facilitate and enhance the detection process. The method not only provides high detection accuracy, but also efficient in regards to memory and execution time. In addition, the method is robust against noise and baseline drift, as supported by the results. The second part outlines a method that extract features from ECG signal in order to classify and predict the severity of arrhythmia. Arrhythmia can be life-threatening or benign. Several methods exist to detect abnormal heartbeats. However, a clear criterion to identify whether the detected arrhythmia is malignant or benign still an open problem. The method discussed in this dissertation will address a novel solution to this important issue. In the third part, a classification model that predicts the severity of loss of blood volume by incorporating multiple physiological signals is elaborated. The features are extracted in time and frequency domains after transforming the signals with Wavelet Transformation (WT). The results support the desirable reliability and accuracy of the system.

Atrial fibrillation is one of the most common cardiac rhythm disorders characterized by ever-increasing prevalence and incidence in the Czech Republic and abroad. The incidence of atrial fibrillation is reported at 2-4 % of the population, but due to the often asymptomatic course, the real prevalence is even higher. The aim of this work is to design an algorithm for automatic detection of atrial fibrillation in the ECG record. In the practical part of this work, an algorithm for the detection of atrial fibrillation is proposed. For the detection itself, the k-nearest neighbor method, the support vector method and the multilayer neural network were used to classify ECG signals using features indicating the variability of RR intervals and the presence of the P wave in the ECG recordings. The best detection was achieved by a model using a multilayer neural network classification with two hidden layers. Results of success indicators: Sensitivity 91.23 %, Specificity 99.20 %, PPV 91.23 %, F-measure 91.23 % and Accuracy 98.53 %.