Retrieving heartbeat information from phonocardiogram
Fecha
2024-04-19
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
En este Trabajo Fin de Máster, se implementaron varios algoritmos. En primer lugar, se adquieren los datos simultáneos de electrocardiografía (ECG) y fonocardiografía (PCG) de bases de datos de acceso abierto. Luego, se implementan técnicas de procesamiento de señales para extraer las características necesarias de las señales de ECG y PCG. El procesamiento se realiza utilizando filtros, técnicas sofisticadas de procesamiento de señales, el algoritmo de Pan Tompkins, brindando una solución estable y confiable para extraer las características. Una vez que se extraen las características necesarias de las señales, se implementan algoritmos de aprendizaje automático para predecir el intervalo QT a partir de las características de PCG. Los resultados se comparan con otros resultados obtenidos previamente.
In this Master Thesis, several algorithms were implemented. In the first place, the electrocardiography (ECG) and phonocardiography (PCG) simultaneous data is acquired from open-source databases. Then signal processing techniques are implemented to extract the necessary features from ECG and PCG signals. The processing is made using filters, sophisticated signal processing techniques, Pan Tompkins algorithm, bringing a stable and reliable solution to extract the features. Once the necessary features are extracted from the signals, machine learning algorithms are implemented in order to predict the QT interval from PCG features. The results are compared to other previously obtained results.
In this Master Thesis, several algorithms were implemented. In the first place, the electrocardiography (ECG) and phonocardiography (PCG) simultaneous data is acquired from open-source databases. Then signal processing techniques are implemented to extract the necessary features from ECG and PCG signals. The processing is made using filters, sophisticated signal processing techniques, Pan Tompkins algorithm, bringing a stable and reliable solution to extract the features. Once the necessary features are extracted from the signals, machine learning algorithms are implemented in order to predict the QT interval from PCG features. The results are compared to other previously obtained results.