No description available.

The dissertation investigates the promises and challenges for application of cardiac signals in biometrics and affective computing, and noninvasive measurement of cardiac signals. We have mainly discussed two major cardiac signals: electrocardiogram (ECG), and photoplethysmogram (PPG). ECG and PPG signals hold strong potential for biometric authentications and identifications. We have shown that by mapping each cardiac beat from time domain to an angular domain using a limit cycle, intra-class variability can be significantly minimized. This is in contrary to conventional time domain analysis. Our experiments with both ECG and PPG signal shows that the proposed method eliminates the effect of instantaneous heart rate on the shape morphology and improves authentication accuracy. For noninvasive measurement of PPG beats, we have developed a systematic algorithm to extract pulse rate from face video in diverse situations using video magnification. We have extracted signals from skin patches and then used frequency domain correlation to filter out non-cardiac signals. We have developed a novel entropy based method to automatically select skin patches from face. We report beat-to-beat accuracy of remote PPG (rPPG) in comparison to conventional average heart rate. The beat-to-beat accuracy is required for applications related to heart rate variability (HRV) and affective computing. The algorithm has been tested on two datasets, one with static illumination condition and the other with unrestricted ambient illumination condition. Automatic skin detection is an intermediate step for rPPG. Existing methods always depend on color information to detect human skin. We have developed a novel standalone skin detection method to show that it is not necessary to have color cues for skin detection. We have used LBP lacunarity based micro-textures features and a region growing algorithm to find skin pixels in an image. Our experiment shows that the proposed method is applicable universally to any image including near infra-red images. This finding helps to extend the domain of many application including rPPG. To the best of our knowledge, this is first such method that is independent of color cues. / Ph. D. / The heart is an integral part of the human body. With every beat, the heart continuously pumps oxygen-enriched blood to providing fuel to our cells and thus enabling life. The heartbeat is initiated by electrical signals generated in the heart muscles. This electrical activity, which are often governed by our autonomic nervous system, can be measured directly by electrocardiogram (ECG) using advanced and often obtrusive instrumentation. Photoplethysmogram (PPG), on the other hand, measures how the blood volume changes and can be readily measured with inexpensive instrumentation at certain locations (e.g. at the fingertip). The ECG and PPG are widely used cardiac signals in medical science for diagnosis and health monitoring. But, these signals hold greater potential than just its medical diagnostic applications. In this work, we have mainly investigated if these signals can be used to identify an individual. Every human heart differs by their size, shape, locations inside body, and internal structure. This motivated us to represent the signals using a mathematical model and use machine learning algorithm to identify individual persons. We have discussed how our method improves the identification accuracy and can be used with current biometric methods like fingerprint in our phone. The measurement procedures of cardiac signals are often cumbersome and need instruments which may not be available outside medical facilities. Therefore, we have investigated alternative method of remote photoplethysmography (rPPG) that are relatively inexpensive and unobtrusive. In this dissertation, we have used face video of an individual to extract the heart rate information. The flow of blood causes small changes in the color of face skin. This is not visible to human eyes without digital magnification, but we have shown how knowledge of distinct behavior of human heart rate and use of advanced computer vision algorithms helped us to extract vital signals like heart rate with a significant accuracy. In addition, to measure rPPG using face video, we integrated a method for automatic detection of skin from images and videos. Existing skin detection methods depended on color information which is not always available within available video sources. We have developed a novel standalone skin detection method to show that it is not necessary to have color cues for skin detection. Our method relies on the context and the texture based appearance of skin. To the best of our knowledge, this is first such method that is independent of color cues. In summary, the dissertation investigates the promises and challenges for application of cardiac signals in biometrics and nonobtrusive measurement of cardiac signals using face video.

The electro-optic technique of measuring the cardiovascular pulse wave known as photoplethysmography (PPG) is clinically utilised for noninvasive characterisation of physiological components by dynamic monitoring of tissue optical absorption. There has been a resurgence of interest in this technique in recent years, driven by the demand for a low cost, compact, simple and portable technology for primary care and community-based clinical settings, and the advancement of computer-based pulse wave analysis techniques. PPG signal provides a means of determining cardiovascular properties during the cardiac cycle and changes with ageing and disease. This thesis focuses on the photoplethysmographic signal for cardiovascular assessment. The contour of the PPG pulse wave is influenced by vascular ageing. Contour analysis of the PPG pulse wave provides a rapid means of assessing vascular tone and arterial stiffness. In this thesis, the parameters extracted from the PPG pulse wave are examined in young adults. The results indicate that the contour parameters of the PPG pulse wave could provide a simple and noninvasive means to study the characteristic change relating to arterial stiffness. The pulsatile component of the PPG signal is due to the pumping action of the heart, and thus could reveal the circulation changes of a specific vascular bed. Heart rate variability (HRV) represents one of the most promising quantitative markers of cardiovascular control. Calculation of HRV from the peripheral pulse wave using PPG, called pulse rate variability (PRV), is investigated. The current work has confirmed that the PPG signal could provide basic information about heart rate (HR) and its variability, and highly suggests a good alternative to understanding dynamics pertaining to the autonomic nervous system (ANS) without the use of an electrocardiogram (ECG) device. Hence, PPG measurement has the potential to be readily accepted in ambulatory cardiac monitoring due to its simplicity and comfort. Noncontact PPG (NPPG) is introduced to overcome the current limitations of contact PPG. As a contactless device, NPPG is especially attractive for physiological monitoring in ambulatory units, NICUs, or trauma centres, where attaching electrodes is either inconvenient or unfeasible. In this research, a prototype for noncontact reflection PPG (NRPPG) with a vertical cavity surface emitting laser (VCSEL) as a light source and a high-speed PiN photodiode as a photodetector is developed. The results from physiological experiments suggest that NRPPG is reliable to extract clinically useful information about cardiac condition and function. In summary, recent evidence demonstrates that PPG as a simple noninvasive measurement offers a fruitful avenue for noninvasive cardiovascular monitoring. Key words: Photoplethysmography (PPG), Cardiovascular assessment, Pulse wave contour analysis, Arterial stiffness, Heart rate (HR), Heart rate variability (HRV), Pulse rate variability (PRV), Autonomic nervous system (ANS), Electrocardiogram (ECG).

Obstruções uretrais em felinos geralmente são secundárias à doença do trato urinário inferior felino (DTUIF), afecção comum em gatos. A doença pode ser auto-limitante, mas a obstrução uretral leva à parada da função renal, gerando uma série de distúrbios no organismo, que necessitam tratamento imediato. A maioria dos estudos sobre as alterações decorrentes da obstrução foi feito em animais em estado crítico ou experimentalmente. O objetivo do presente estudo foi descrever as principais alterações observadas em gatos obstruídos atendidos na rotina clínica e correlacioná-las. Foram avaliados 32 gatos machos com obstrução uretral e sem tratamento prévio. A avaliação constou de exame clínico, hemograma, bioquímica sérica (uréia, creatinina, proteína total, ALT, AST, fosfatase alcalina, sódio, potássio, cálcio, fósforo, magnésio, glicemia e lactato), gasometria venosa, exame e cultura de urina e eletrocardiograma (ECG). Os animais foram agrupados conforme o tempo de obstrução (mais e menos de 36 horas). As alterações mais comuns no histórico foram disúria (100% dos animais), disorexia (84,4%), apatia (71,8%), vocalização (68,7%) e oligodipsia (68,7%); no exame físico, desidratação (71,8%), taquipnéia (53,1%) e hipotermia (53,1%). As alterações laboratoriais mais freqüentes foram hipermagnesemia (100%), acidose metabólica (89,6%), hiperglicemia (88,9%), hiperazotemia (84,4%) e hiperpotassemia (80,6%). Vinte por cento dos gatos tinham infecção urinária. Alterações no ECG foram evidenciadas em 39,3% dos casos, sendo a parada atrial com ritmo sinoventricular a mais freqüente. Não houve relação entre as alterações no ECG e os níveis de potássio sérico. A análise dos grupos sugere agravamento da hiperazotemia, hiperpotassemia, hipermagnesemia e do estado geral com a evolução do processo. Nas correlações, a temperatura e a freqüência cardíaca apresentaram relação direta com pH sanguíneo, excesso de base e bicarbonato, e relação inversa com uréia, creatinina, potássio e fósforo. A uréia e creatinina se correlacionaram inversamente com sódio, pH sanguíneo, excesso de base e bicarbonato, e diretamente com potássio e fósforo. O estado geral correlacionou-se com a temperatura, uréia, creatinina, potássio, pH sanguíneo, excesso de base e bicarbonato. / Urethral obstruction is a frequent complication in cats with feline lower urinary tract disease (FLUTD), considered a common disease in cats. Most cases are self-limiting, but the urethral obstruction causes renal failure and metabolic alterations which needs immediate treatment. Previous reports selected cats that were critically ill or had a experimental induction of the disease. The goal of this study was to describe the clinical signs, laboratory and electrocardiographyc abnormalities in cats with urethral obstruction and to correlate these results. Thirty-two male cats with natural urethral obstruction and without previous therapy were studied. Complete blood count, serum chemistry profile (urea, creatinine, plasma protein, alanine transferase, aspartate transferase, alkaline phosphatase, sodium, potassium, calcium, phosphorus, magnesium, glucose and lactate), venous blood gas, urinalysis, urine culture and electrocardiogram (ECG) were performed. Two groups of 16 cats were composed (animals under and over 36 hours of obstruction). The most frequent abnormalities described were dysuria (100%), anorexia (84,4%), lethargy (71,8%), vocalization (68,7%), hypodipsia (68,7%), and dehydration (71,8%), tachypnea (53,1%) and hypothermia (53,1%) in physical evaluation. The laboratorial abnormalities most frequently observed were hypermagnesemia (100%), metabolic acidosis (89,6%), hyperglicemia (88,9%), azotemia (84,4%) and hyperkalemia (80,6%). Twenty per cent of cats had urinary infections. Abnormalities in ECG, such as atrial standstill with sinoventricular rhythm, were seen in 39,3% of cases, and there were no correlation with hyperkalemia. The analyses of the groups suggest worsening of azotemia, hyperkalemia, hypermagnesemia and lethargy with evolution of obstruction. Temperature and heart rate were positively correlated with blood pH, base excess and bicarbonate, and inversely correlated with urea, creatinine, potassium and phosphorus. Lethargy was correlated with temperature, blood pH, base excess and bicarbonate.

A cardiomiopatia arritmogênica do ventrículo direito (CAVD) é uma doença miocárdica genética e hereditária, com alta prevalência entre os cães da raça Boxer, causada por alterações nos genes que codificam os receptores de rianodina ou os desmossomos cardíacos. Caracteriza-se, histologicamente, por atrofia de miócitos do ventrículo direito, principalmente, com reposição por tecido adiposo e fibroso. Tais lesões causam instabilidade elétrica do miocárdio, com consequente aparecimento de arritmias ventriculares por re-entrada. Os animais afetados podem apresentar-se assintomáticos ou com frequentes episódios de síncope ou morte súbita; ainda, podem desenvolver uma forma mais grave da doença, com dilatação ventricular e disfunção sistólica, evoluindo para insuficiência cardíaca congestiva. Devido à escassez de estudos determinando fatores preditivos para a ocorrência de morte de origem cardíaca nestes animais, o presente estudo objetivou avaliar o valor prognóstico das variáveis de eletrocardiografia ambulatorial (Holter), eletrocardiografia de alta resolução (ECGAR) e variabilidade de frequência cardíaca (VFC) na CAVD do Boxer. Para tanto, tais parâmetros foram avaliados em 92 cães, distribuídos de acordo com a raça ou o estágio da doença em cinco grupos: cãescontrole da raça Pastor Alemão (grupo PA), cães-controle da raça Boxer (grupo BC), cães Boxer suspeitos para CAVD (grupo BSUS), cães Boxer afetados pela CAVD (grupo BCAVD) e cães Boxer com CAVD e disfunção sistólica (grupo BDS). Foram encontrados potenciais tardios ao ECGAR em poucos cães do grupo CAVD (21,05%); no entanto, dentre os cães do grupo BDS este percentual foi bem mais expressivo (83,33%), demonstrando que a sensibilidade deste exame é maior em estágios mais avançados da doença. Apesar da observação destes potenciais tardios, as variáveis analisadas ao ECGAR não apresentaram correlação com a sobrevida dos cães com CAVD. Também não se constituíram em fatores preditivos de morte os índices de VFC. Apenas a idade do animal no momento do diagnóstico, a ocorrência de taquicardia ventricular não sustentada e de extrassístoles supraventriculares ao Holter apresentaram-se como fatores prognósticos na sobrevida dos cães Boxer com CAVD. As variáveis de ECGAR e VFC não evidenciaram correlação com o número ou grau das arritmias ventriculares registradas ao Holter dos cães Boxer com CAVD, demonstrando que a gênese destas arritmias não está relacionada à disfunção do sistema nervoso simpático. / Arhythmogenic right ventricular cardiomyopathy (ARVC) is a genetic and hereditary myocardial disease with a high prevalence among Boxer dogs, caused by changes in genes that encode ryanodine receptors or desmosomes. It is characterized histologically by myocyte atrophy, especially of the right ventricle, with replacement by fat and fibrous tissue. Such injuries cause electrical instability of myocardium, with subsequent appearance of re-entrant ventricular arrhythmias. Affected animals may be asymptomatic, show frequent episodes of syncope or sudden death, or develop a more severe form of disease with ventricular dilatation and systolic dysfunction, progressing to congestive heart failure. Given the paucity of studies determining preditive factors for the occurrence of cardiac death in these animals, this study aimed to evaluate the prognostic value of variables of ambulatory electrocardiography (Holter) signal-averaged electrocardiography (SAECG) and heart rate variability (HRV) in Boxer dogs with ARVC. For this, these parameters were evaluated in 92 dogs, divided according to race or stage of the disease into five groups: German Shepherd control dogs (group PA), Boxer control dogs (group BC), Boxer dogs suspicious for ARVC (group BSUS), Boxer dogs affected by ARVC (group BCAVD) and Boxer dogs with ARVC and systolic dysfunction (group BDS). Late potentials were found in SAECG of few dogs on group CAVD (21.05%), however, among dogs of group BDS, this percentage was much higher (83.33%), demonstrating that the sensitivity of this test is higher in more advanced stages of the disease. Despite the observation of late potentials in SAECG, the variables of this test were not correlated with survival in dogs with ARVC. The indexes of HRV were also not significantly predictive of death in ARVC dogs. Only the age of the animal at diagnosis, the occurrence of nonsustained ventricular tachycardia and supraventricular ectopic beats on Holter presented as prognostic factors in survival of Boxer dogs with ARVC. The SAECG and HRV variables were not correlated with the number or degree of ventricular arrhythmias recorded by Holter of ARVC Boxer dogs, demonstrating that the genesis of these arrhythmias is not related to dysfunction of the sympathetic nervous system.

Métodos de identificação de pessoas sempre foram muito importantes para toda a sociedade. Atualmente, as pesquisas em biometria vêm sendo amplamente incentivadas por diversos setores da indústria mundial com o objetivo de melhorar ou substituir os atuais sistemas de segurança e de identificação de pessoas. O campo da biometria abarca uma grande variedade de tecnologias usadas para identificar e verificar a identidade de uma pessoa por meio da mensuração e análise de diversas características físicas e/ou comportamentais do ser humano. Diversas modalidades biométricas têm sido propostas para reconhecimento de pessoas, como impressão digital, íris, face e fala. Estas modalidades biométricas possuem características distintas em termos de desempenho, mensurabilidade e aceitabilidade. Uma questão a ser considerada com a aplicação biométrica é sua robustez a ataques por circunvenção, repetição e ofuscação. Esses ataques estão se tornando cada vez mais frequentes e questionamentos estão sendo levantados a respeito dos níveis de segurança das formas de reconhecimento. Sinais biomédicos como eletrocardiograma (ECG), eletroencefalograma (EEG) e eletromiograma (EMG) têm sido cada vez mais estudados e aplicados ao reconhecimento biométrico. Em específico, os sinais de ECG têm sido largamente adotados para o reconhecimento biométrico em diversos trabalhos. Por outro lado, análise de séries temporais tem sido usada com sucesso em muitas diferentes aplicações para identificar padrões temporais nos dados. Embora dinâmica simples possa ser observada com ferramentas analíticas tradicionais tais como transformada de fourier, transformada wavelet, a representação simbólica pode melhorar a análise de processos que são complexos e possivelmente caótico. Além disso, representação simbólica pode também reduzir a sensibilidade a ruído e melhorar bastante a eficiência computacional. No entanto, existem aspectos estruturais e paramétricos de projeto que podem conduzir a uma degradação de desempenho. Na ausência de uma metodologia sistemática e de baixo custo para a proposição de técnicas de representação simbólicas otimamente especificadas, os comitês de máquinas, mais especificamente ensemble, se apresentam como alternativas promissoras. Neste estudo, os componentes do ensemble, que correspondem as técnicas de representação simbólicas, e seus respectivos parâmetros foram selecionados via algoritmos evolutivos. O objetivo é explorar conjuntamente potencialidades advindas das técnicas de representação simbólicas e comitê de máquinas para reconhecimento biométrico baseado em sinais de ECG. Resultados experimentais conduzidos sobre dois conjuntos de dados disponíveis publicamente indicam que a abordagem proposta pode melhorar o desempenho do reconhecimento quando comparada com as técnicas tradicionais / Identification people methods have been very important for the whole society. Currently, research on biometrics have been widely encouraged by various sectors of the industry worldwide in order to improve or replace existing security systems and people identification. The field of biometrics includes a variety of technologies used to identify or verify the identity of a person by measuring and analyzing various physical and/or behavioral aspects of the human being. Several biometric methods have been proposed for recognition of people, such as fingerprint, iris, face and speech. These biometric modalities have different characteristics in terms of performance, measurability and acceptability. One issue to be considered with the biometric application in the real world is its robustness to attacks by circumvention, repetition and obfuscation. These attacks are becoming more frequent and more questions are being raised about the levels of security that this technology can offer. Biomedical signals such as electrocardiogram (ECG), electroencephalogram (EEG) and electromyogram (EMG) have been increasingly studied and applied to biometric recognition. Specifically, ECG signals have been widely adopted for biometric recognition in various works. On the other hand, time series analysis has been used successfully in many different applications to identify temporal patterns in the data. Although simple dynamics can be observed with traditional analytical tools such as fourier transform, wavelet transform, the symbolic representation can improve the analysis of processes that are complex and possibly chaotic. In addition, symbolic representation can also reduce noise sensitivity and greatly improve computational efficiency. However, there are structural and parametric design aspects that can lead to performance degradation. In the absence of a systematic and inexpensive methodology for proposing optimally specified symbolic representation techniques, machine committees, more specifically ensemble, present themselves as promising alternatives. In this study, the components of the committee, which correspond to techniques of symbolic representation, and their respective parameters were selected via evolutionary algorithms. The objective is to jointly explore the potentialities of both symbolic representation techniques and machine committee for biometric recognition based on ECG signals. Experimental results conducted on two publicly available datasets indicate that the proposed approach may improve recognition performance when compared to traditional techniques

A análise da variabilidade da frequência cardíaca é um método útil para avaliar o funcionamento anormal do sistema nervoso autônomo e para prever eventos cardíacos em pacientes com insuficiência cardíaca. As medidas da variabilidade da frequência cardíaca por meio de cardiofrequencímetro têm sido validadas em comparação com eletrocardiograma em indivíduos saudáveis, mas não em pacientes com insuficiência cardíaca. Nós exploramos a reprodutibilidade dos índices da variabilidade da frequência cardíaca obtidos por meio de um cardiofrequencímetro (PolarS810i) e um eletrocardiograma portátil (Holter) nas fases de repouso e de recuperação em dois testes da caminhada de seis minutos consecutivos, com 60 minutos de intervalo entre os dois testes em 50 pacientes com insuficiência cardíaca (~59 anos, New York Heart Association classe funcional II, fração de ejeção do ventrículo esquerdo ~35%). A reprodutibilidade das medidas para cada dispositivo foi analisada por meio do t-test pareado ou Wilcoxon signed-rank test. Adicionalmente, nós avaliamos a concordância entre os dois dispositivos na análise dos índices da variabilidade da frequência cardíaca em repouso, durante o teste da caminhada de seis minutos e durante a fase de recuperação por meio do Coeficiente de Correlação e Concordância (CCC) com 95% de intervalo de confiança e gráficos Bland-Altman. O teste-reteste para análise da variabilidade da frequência cardíaca foi reprodutível com o uso de ambos, o Holter e o PolarS810i, no repouso mas não na fase de recuperação. No segundo teste da caminhada de seis minutos, os pacientes apresentaram aumentos significativo do rMSSD e da distância percorrida. A confiabilidade das medidas do PolarS810i foram consideravelmente altas [0,86 < CCC < 0,99) com base nas medidas do Holter nas três fases: em repouso, durante o teste da caminhada de seis minutos e durante a recuperação. A menor concordância [CCC=0,86] entre os dois dispositivos foi observada no pNN50 durante o teste da caminhada de seis minutos e na fase de recuperação. Em conclusão, nosso estudo mostrou boa reprodutibilidade dos índices da variabilidade da frequência cardíaca em repouso em dois consecutivos testes da caminhada de seis minutos utilizando Holter e PolarS810i. Adicionalmente, o PolarS810i produziu índices confiáveis da variabilidade da frequência cardíaca a partir de registros de curta duração e com base nas gravações simultâneas com o Holter nas três fases: em repouso, durante o teste da caminhada de seis minutos e durante a recuperação em pacientes com insuficiência cardíaca / Heart rate variability analysis is a useful method to assess abnormal functioning in the autonomic nervous system and to predict cardiac events in patients with heart failure. Heart rate variability measurements with heart rate monitors have been validated with an electrocardiograph in healthy subjects but not in patients with heart failure. We explored the reproducibility of heart rate variability indexes using a heart rate monitor (Polar S810i) and a portable electrocardiograph (Holter) at rest and at recovery of two consecutive six-min walk tests, 60 minutes apart in 50 heart failure patients (~59 years, New York Heart Association functional class II, left ventricular ejection fraction ~35%). The reproducibility for each device was analysed using a paired t-test or the Wilcoxon signed-rank test. Additionally, we assessed the agreement between the two devices based on the heart rate variability indexes at rest, during the six-min walk test and during recovery using Concordance Correlation Coefficients (CCC), 95% confidence intervals and Bland-Altman plots. The test-retest for the heart rate variability analyses was reproducible using Holter and PolarS810i at rest but not during recovery. In the second six-min walk test, patients showed significant increases in rMSSD and walking distance. The reliability of PolarS810i measurements was remarkably high [0.86 < CCC < 0.99] based on Holter in the three phases: at rest, during six-min walk test and during recovery. The lowest agreement [CCC=0.86] between the two devices was observed in pNN50 during the six-min walk test and recovery. In conclusion, our study showed good reproducibility of heart rate variability indexes at rest in two consecutive six-min walk test using Holter and Polar S810i. Additionally, PolarS810i produced reliable short-term heart rate variability indexes based on Holter simultaneous recordings at rest, during the six-min walk test and during recovery in heart failure patients

The goal for this master thesis is to develop a prototype that uses ultra-wideband (UWB) communications to wirelessly transfer electrocardiogram (ECG) data from an ECG measurement unit to an Android device (smartphone or similar) which is used to process and display the ECG signals. The prototype should consist of two hardware nodes; (1) Node one having a ECG measurement unit (an AD8232 single lead heart rate monitor), an UWB communication module (a Decawave DWM1000 module) and a microcontroller (an Arduino DUE); and (2) Node two having an Android device (an Android smartphone), an UWB communication module (a Decawave DWM1000 module) and a microcontroller (an Arduino DUE). On Node one the AD8232 monitor for ECG measurements is connected to an analog input (with an analog to digital converter (ADC)) on the Arduino and the DWM1000 module is connected to the Arduino via serial peripheral interface (SPI). On Node two the DWM1000 is connected to the Arduino via SPI to receive ECG data from Node one, and the Arduino is connected to the smartphone through a serial USB cable with an USB on-the-go adapter to send the ECG data to the smartphone, where it is filtered and displayed with an Android application. The application has the potential to add, for example, ECG analysis for diagnosing heart activities with artificial intelligence (AI) and further transmit the ECG data for remote medical care. The Arduino is programmed in Arduino IDE (integrated development environment) to handle ECG measurements and UWB communications (transmitting and receiving ECG data), which is limited to a single UWB channel because of limitations of the DWM1000 module. The Android application is created using Android studio, and it can process (with a notch filter) and display 1-12 channel ECG. The prototype has been built and tested. The results show that a single lead ECG measurement can be sent via UWB communication to a smartphone to display in real time. Multiple data channels (1-12 analog inputs on the Arduino) can be multiplexed, transmitted and displayed in real time. This thesis concludes that UWB has huge development potential, and will likely be used for various wireless devices in the future.

Métodos de identificação de pessoas sempre foram muito importantes para toda a sociedade. Atualmente, as pesquisas em biometria vêm sendo amplamente incentivadas por diversos setores da indústria mundial com o objetivo de melhorar ou substituir os atuais sistemas de segurança e de identificação de pessoas. O campo da biometria abarca uma grande variedade de tecnologias usadas para identificar e verificar a identidade de uma pessoa por meio da mensuração e análise de diversas características físicas e/ou comportamentais do ser humano. Diversas modalidades biométricas têm sido propostas para reconhecimento de pessoas, como impressão digital, íris, face e fala. Estas modalidades biométricas possuem características distintas em termos de desempenho, mensurabilidade e aceitabilidade. Uma questão a ser considerada com a aplicação biométrica é sua robustez a ataques por circunvenção, repetição e ofuscação. Esses ataques estão se tornando cada vez mais frequentes e questionamentos estão sendo levantados a respeito dos níveis de segurança das formas de reconhecimento. Sinais biomédicos como eletrocardiograma (ECG), eletroencefalograma (EEG) e eletromiograma (EMG) têm sido cada vez mais estudados e aplicados ao reconhecimento biométrico. Em específico, os sinais de ECG têm sido largamente adotados para o reconhecimento biométrico em diversos trabalhos. Por outro lado, análise de séries temporais tem sido usada com sucesso em muitas diferentes aplicações para identificar padrões temporais nos dados. Embora dinâmica simples possa ser observada com ferramentas analíticas tradicionais tais como transformada de fourier, transformada wavelet, a representação simbólica pode melhorar a análise de processos que são complexos e possivelmente caótico. Além disso, representação simbólica pode também reduzir a sensibilidade a ruído e melhorar bastante a eficiência computacional. No entanto, existem aspectos estruturais e paramétricos de projeto que podem conduzir a uma degradação de desempenho. Na ausência de uma metodologia sistemática e de baixo custo para a proposição de técnicas de representação simbólicas otimamente especificadas, os comitês de máquinas, mais especificamente ensemble, se apresentam como alternativas promissoras. Neste estudo, os componentes do ensemble, que correspondem as técnicas de representação simbólicas, e seus respectivos parâmetros foram selecionados via algoritmos evolutivos. O objetivo é explorar conjuntamente potencialidades advindas das técnicas de representação simbólicas e comitê de máquinas para reconhecimento biométrico baseado em sinais de ECG. Resultados experimentais conduzidos sobre dois conjuntos de dados disponíveis publicamente indicam que a abordagem proposta pode melhorar o desempenho do reconhecimento quando comparada com as técnicas tradicionais / Identification people methods have been very important for the whole society. Currently, research on biometrics have been widely encouraged by various sectors of the industry worldwide in order to improve or replace existing security systems and people identification. The field of biometrics includes a variety of technologies used to identify or verify the identity of a person by measuring and analyzing various physical and/or behavioral aspects of the human being. Several biometric methods have been proposed for recognition of people, such as fingerprint, iris, face and speech. These biometric modalities have different characteristics in terms of performance, measurability and acceptability. One issue to be considered with the biometric application in the real world is its robustness to attacks by circumvention, repetition and obfuscation. These attacks are becoming more frequent and more questions are being raised about the levels of security that this technology can offer. Biomedical signals such as electrocardiogram (ECG), electroencephalogram (EEG) and electromyogram (EMG) have been increasingly studied and applied to biometric recognition. Specifically, ECG signals have been widely adopted for biometric recognition in various works. On the other hand, time series analysis has been used successfully in many different applications to identify temporal patterns in the data. Although simple dynamics can be observed with traditional analytical tools such as fourier transform, wavelet transform, the symbolic representation can improve the analysis of processes that are complex and possibly chaotic. In addition, symbolic representation can also reduce noise sensitivity and greatly improve computational efficiency. However, there are structural and parametric design aspects that can lead to performance degradation. In the absence of a systematic and inexpensive methodology for proposing optimally specified symbolic representation techniques, machine committees, more specifically ensemble, present themselves as promising alternatives. In this study, the components of the committee, which correspond to techniques of symbolic representation, and their respective parameters were selected via evolutionary algorithms. The objective is to jointly explore the potentialities of both symbolic representation techniques and machine committee for biometric recognition based on ECG signals. Experimental results conducted on two publicly available datasets indicate that the proposed approach may improve recognition performance when compared to traditional techniques

A cardiomiopatia arritmogênica do ventrículo direito (CAVD) é uma doença miocárdica genética e hereditária, com alta prevalência entre os cães da raça Boxer, causada por alterações nos genes que codificam os receptores de rianodina ou os desmossomos cardíacos. Caracteriza-se, histologicamente, por atrofia de miócitos do ventrículo direito, principalmente, com reposição por tecido adiposo e fibroso. Tais lesões causam instabilidade elétrica do miocárdio, com consequente aparecimento de arritmias ventriculares por re-entrada. Os animais afetados podem apresentar-se assintomáticos ou com frequentes episódios de síncope ou morte súbita; ainda, podem desenvolver uma forma mais grave da doença, com dilatação ventricular e disfunção sistólica, evoluindo para insuficiência cardíaca congestiva. Devido à escassez de estudos determinando fatores preditivos para a ocorrência de morte de origem cardíaca nestes animais, o presente estudo objetivou avaliar o valor prognóstico das variáveis de eletrocardiografia ambulatorial (Holter), eletrocardiografia de alta resolução (ECGAR) e variabilidade de frequência cardíaca (VFC) na CAVD do Boxer. Para tanto, tais parâmetros foram avaliados em 92 cães, distribuídos de acordo com a raça ou o estágio da doença em cinco grupos: cãescontrole da raça Pastor Alemão (grupo PA), cães-controle da raça Boxer (grupo BC), cães Boxer suspeitos para CAVD (grupo BSUS), cães Boxer afetados pela CAVD (grupo BCAVD) e cães Boxer com CAVD e disfunção sistólica (grupo BDS). Foram encontrados potenciais tardios ao ECGAR em poucos cães do grupo CAVD (21,05%); no entanto, dentre os cães do grupo BDS este percentual foi bem mais expressivo (83,33%), demonstrando que a sensibilidade deste exame é maior em estágios mais avançados da doença. Apesar da observação destes potenciais tardios, as variáveis analisadas ao ECGAR não apresentaram correlação com a sobrevida dos cães com CAVD. Também não se constituíram em fatores preditivos de morte os índices de VFC. Apenas a idade do animal no momento do diagnóstico, a ocorrência de taquicardia ventricular não sustentada e de extrassístoles supraventriculares ao Holter apresentaram-se como fatores prognósticos na sobrevida dos cães Boxer com CAVD. As variáveis de ECGAR e VFC não evidenciaram correlação com o número ou grau das arritmias ventriculares registradas ao Holter dos cães Boxer com CAVD, demonstrando que a gênese destas arritmias não está relacionada à disfunção do sistema nervoso simpático. / Arhythmogenic right ventricular cardiomyopathy (ARVC) is a genetic and hereditary myocardial disease with a high prevalence among Boxer dogs, caused by changes in genes that encode ryanodine receptors or desmosomes. It is characterized histologically by myocyte atrophy, especially of the right ventricle, with replacement by fat and fibrous tissue. Such injuries cause electrical instability of myocardium, with subsequent appearance of re-entrant ventricular arrhythmias. Affected animals may be asymptomatic, show frequent episodes of syncope or sudden death, or develop a more severe form of disease with ventricular dilatation and systolic dysfunction, progressing to congestive heart failure. Given the paucity of studies determining preditive factors for the occurrence of cardiac death in these animals, this study aimed to evaluate the prognostic value of variables of ambulatory electrocardiography (Holter) signal-averaged electrocardiography (SAECG) and heart rate variability (HRV) in Boxer dogs with ARVC. For this, these parameters were evaluated in 92 dogs, divided according to race or stage of the disease into five groups: German Shepherd control dogs (group PA), Boxer control dogs (group BC), Boxer dogs suspicious for ARVC (group BSUS), Boxer dogs affected by ARVC (group BCAVD) and Boxer dogs with ARVC and systolic dysfunction (group BDS). Late potentials were found in SAECG of few dogs on group CAVD (21.05%), however, among dogs of group BDS, this percentage was much higher (83.33%), demonstrating that the sensitivity of this test is higher in more advanced stages of the disease. Despite the observation of late potentials in SAECG, the variables of this test were not correlated with survival in dogs with ARVC. The indexes of HRV were also not significantly predictive of death in ARVC dogs. Only the age of the animal at diagnosis, the occurrence of nonsustained ventricular tachycardia and supraventricular ectopic beats on Holter presented as prognostic factors in survival of Boxer dogs with ARVC. The SAECG and HRV variables were not correlated with the number or degree of ventricular arrhythmias recorded by Holter of ARVC Boxer dogs, demonstrating that the genesis of these arrhythmias is not related to dysfunction of the sympathetic nervous system.