Print Email Facebook Twitter Model-based Feature Engineering of Atrial Fibrillation Title Model-based Feature Engineering of Atrial Fibrillation Author Moghaddasi, Hanie (TU Delft Signal Processing Systems) Contributor van der Veen, A.J. (promotor) de Groot, N.M.S. (promotor) Hunyadi, Borbala (promotor) Degree granting institution Delft University of Technology Date 2024-06-18 Abstract Atrial Fibrillation (AF) is the most common tachyarrhythmia in the heart. Irregular RR intervals and the absence of a P wave before the QRS complex characterize AF. Although many studies have been done to detect atrial fibrillation, many aspects of this intricate disease need further analysis. AF is often diagnosed by the interpretation of multi-lead electrocardiograms (ECGs), which provide a non-invasive comprehensive evaluation of cardiac electrical activity. However, due to the poor spatial resolution of ECG recordings, the characteristics of AF cannot be fully demonstrated by using multi-lead ECG solely. Better spatial resolution is obtained by using high-resolution epicardial electrograms measured directly at the surface of the heart. The combination of multi-lead ECGs and high-resolution electrograms should provide a more detailed view of atrial fibrillation. To analyze such data, we first need to derive relevant features that can reduce the data and capture their essence. An initial aim of this research is to increase the accuracy of AF detection and identification of electropathological regions within the heart using the derived features. The ultimate goal is to (non-invasively) monitor the progression of atrial fibrillation through its subsequent stages.... Subject Atrial FibrillationElectrocardiogramPoincaréVectorcardiogramAtrial ActivityDominant FrequencyElectrogramAction PotentialBody Surface PotentialsRank AnalysisSingular Value Decomposition To reference this document use: https://doi.org/10.4233/uuid:36d20140-0563-4fef-a49b-3f548e604c6c ISBN 978-94-6496-131-7 Part of collection Institutional Repository Document type doctoral thesis Rights © 2024 Hanie Moghaddasi Files PDF Dissertation_Hanie.pdf 13.28 MB Close viewer /islandora/object/uuid:36d20140-0563-4fef-a49b-3f548e604c6c/datastream/OBJ/view