| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 3.09 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
A existência de diversas arritmias paroxísticas representa um desafio significativo
para a monitorização e o diagnóstico dos pacientes. Muitas vezes, essas arritmias não são
detetadas durante as consultas médicas, por serem intermitentes e, frequentemente,
assintomáticas. É neste enquadramento que se tornam cada vez mais relevantes os
dispositivos das tecnologias móveis (mHealth), que permitem monitorizar de forma contínua
o ritmo cardíaco.
O CardioWheel surge no contexto da indústria automóvel, como um dispositivo que
permite adquirir uma derivação semelhante ao DI de um eletrocardiograma (ECG) clínico,
através de elétrodos secos embutidos num volante. No presente estudo pretende-se realizar
a validação clínica deste dispositivo.
Para tal, foi realizado um estudo clínico que comparou o registo de ECG clínicos de
12 derivações com os registos obtidos simultaneamente com o CardioWheel. A base de
dados obtida consistiu em 143 registos correspondentes a pacientes do Hospital de Santa
Marta, escolhidos de forma aleatória. Destes, 120 cumpriram os critérios de elegibilidade.
A qualidade do sinal do CardioWheel foi avaliada recorrendo a diferentes métodos.
Numa primeira abordagem, foi realizada uma análise clínica que categorizou o sinal do
CardioWheel e da derivação I do ECG clínico de cada paciente, utilizando 5 categorias
predefinidas. Em seguida, cada traçado foi classificado com base nas arritmias que podem
ser identificadas, tendo disponível apenas DI. Foram criadas matrizes de confusão para
comparar os resultados de cada traçado.
Numa segunda abordagem, realizou-se a quantificação da qualidade através de uma
ferramenta de processamento e análise de sinais, tendo em conta 3 aspetos essenciais:
ruído, ritmo e morfologia.
Na generalidade, comprovou-se que o sinal do CardioWheel é sobreponível com o do
ECG clínico em DI, sendo possível utilizá-lo para a deteção de arritmias não emergentes.
Contudo, é essencial salientar as restrições inerentes à aquisição de apenas uma derivação
para análise clínica, que enfatizam a importância de complementar o uso do CardioWheel
com exames hospitalares para uma avaliação mais completa do paciente.
The existence of various paroxysmal arrhythmias represents a significant challenge for monitoring and diagnosing patients. Often, these arrhythmias are not detected during medical consultations because they are intermittent and, regularly, asymptomatic. It is in this context that mHealth devices that enable continuous heart rhythm monitoring are becoming increasingly relevant. The CardioWheel has emerged in the context of the automotive industry as a device that makes it possible to acquire a lead similar to the DI of a clinical electrocardiogram (ECG), using dry electrodes embedded in a steering wheel. The present study aims to carry out the clinical validation of this device. In order to do this, a clinical study was carried out comparing clinical 12-lead ECG recordings with those obtained simultaneously with CardioWheel. The database obtained consisted of 143 records corresponding to patients from the Santa Marta Hospital, chosen at random. Of these, 120 met the eligibility criteria. The quality of the CardioWheel signal was assessed using different methods. In a first approach, a clinical analysis was carried out which categorized the CardioWheel signal and lead I of each patient's clinical ECG using 5 predefined categories. Next, each tracing was classified based on the arrhythmias that could be identified, with only DI available. Confusion matrices were created to compare the results of each tracing. In a second approach, quality was quantified using a signal processing and analysis tool, considering 3 essential aspects: noise, rhythm and morphology. In general, it was found that the CardioWheel signal is comparable to that of the clinical ECG in DI and can be used to detect non-emergent arrhythmias. However, it is essential to be aware of the restrictions inherent in acquiring only one lead for clinical analysis, which emphasize the importance of complementing the use of CardioWheel with hospital exams for a more complete assessment of the patient.
The existence of various paroxysmal arrhythmias represents a significant challenge for monitoring and diagnosing patients. Often, these arrhythmias are not detected during medical consultations because they are intermittent and, regularly, asymptomatic. It is in this context that mHealth devices that enable continuous heart rhythm monitoring are becoming increasingly relevant. The CardioWheel has emerged in the context of the automotive industry as a device that makes it possible to acquire a lead similar to the DI of a clinical electrocardiogram (ECG), using dry electrodes embedded in a steering wheel. The present study aims to carry out the clinical validation of this device. In order to do this, a clinical study was carried out comparing clinical 12-lead ECG recordings with those obtained simultaneously with CardioWheel. The database obtained consisted of 143 records corresponding to patients from the Santa Marta Hospital, chosen at random. Of these, 120 met the eligibility criteria. The quality of the CardioWheel signal was assessed using different methods. In a first approach, a clinical analysis was carried out which categorized the CardioWheel signal and lead I of each patient's clinical ECG using 5 predefined categories. Next, each tracing was classified based on the arrhythmias that could be identified, with only DI available. Confusion matrices were created to compare the results of each tracing. In a second approach, quality was quantified using a signal processing and analysis tool, considering 3 essential aspects: noise, rhythm and morphology. In general, it was found that the CardioWheel signal is comparable to that of the clinical ECG in DI and can be used to detect non-emergent arrhythmias. However, it is essential to be aware of the restrictions inherent in acquiring only one lead for clinical analysis, which emphasize the importance of complementing the use of CardioWheel with hospital exams for a more complete assessment of the patient.
Description
Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia Biomédica
Keywords
ECG CardioWheel Arritmias DI mHealth Arrhythmias
Citation
FERREIRA, Carla Sofia Barão – Validação clínica de um sistema de aquisição de electrocardiografia não intrusivo - estudo preliminar. Lisboa: Instituto Superior de Engenharia de Lisboa, 2024. Dissertação de Mestrado.
Publisher
Instituto Superior de Engenharia de Lisboa