Use of phonocardiograms to extract knowledge from heartbeats

[en] In clinical cardiology, the QT interval is an important marker for determining arrhythmia risk and has typically been assessed using electrocardiograms (ECGs). However, due to the expense and complexity of the technology, ECG monitoring may be prohibitive in some situations. This thesis investi...

Descripción completa

Detalles Bibliográficos
Autor: Allam, Majdouline
Tipo de recurso: tesis de maestría
Fecha de publicación:2025
País:España
Institución:Universidad de Jaén (UJA)
Repositorio:CREA. Colección de recursos educativos abiertos
OAI Identifier:oai:crea.ujaen.es:10953.1/26483
Acceso en línea:https://hdl.handle.net/10953.1/26483
Access Level:acceso abierto
Palabra clave:Telecomunicaciones
Procesamiento de Señales
3325
3311
Descripción
Sumario:[en] In clinical cardiology, the QT interval is an important marker for determining arrhythmia risk and has typically been assessed using electrocardiograms (ECGs). However, due to the expense and complexity of the technology, ECG monitoring may be prohibitive in some situations. This thesis investigates the feasibility of measuring the QT interval using phonocardiogram (PCG) data, which provides a more accessible and non-invasive alternative. The project evaluates a database of 409 simultaneous ECG and PCG recordings. PCG features including S1, S2, Systole, and Diastole durations are recovered using hand-corrected annotations as well as the Springer segmentation technique. The project develops two predictive models: a Bayesian-optimized support vector machine (SVM) regression model and an explicit multiple linear regression model