Mathematical techniques for assessing cardiac wavefront dynamics
File(s)
Author(s)
Roney, Caroline Helen
Type
Thesis
Abstract
Electrical signals measured during cardiac arrhythmias may be difficult to analyse and there are multiple challenges associated with their interpretation, including noise, variability between and within datasets, data resolution, and complicated wavefront behaviour. This thesis aimed to develop and test techniques for analysing cardiac electrical wavefront dynamics applied to in vitro, in silico in vivo data.
One such technique was developed to determine conduction velocity and to estimate the location of the origin of a focal source, assuming either a planar or circular wavefront, measured from any arbitrary arrangement of recording points. This algorithmic technique performed well on simulated and clinical data, and was extended to incorporate curvature of the atrium.
The sensitivity to filtering and post-processing parameters of a technique for tracking rotational sources in optical mapping experiments was investigated, and an optimal parameter set was determined. This was used alongside a technique developed to calculate the distribution of new wavefront initiation sites, in order to gain insight into the mechanisms sustaining atrial fibrillation. The technique for tracking rotational sources was extended to work for unipolar and bipolar electrogram data. This extension was tested in simulation against action potential data, where the modalities (unipolar electrogram, bipolar electrogram and action potential phase) were found to perform similarly; and then applied to experimental and clinical electrograms. Rotational content and activation maps were found to be similar for clinical unipolar and bipolar phase.
The developed techniques were used to investigate the effects of spatial resolution on the identification of rotors and focal sources, as a function of the distance between successive wavefronts. A technique was presented to estimate this distance from a small number of measuring points, providing experimental and clinical utility. These requirements were then tested for simulated high-density mapping catheters and basket catheters, where all catheters provided sufficient resolution.
One such technique was developed to determine conduction velocity and to estimate the location of the origin of a focal source, assuming either a planar or circular wavefront, measured from any arbitrary arrangement of recording points. This algorithmic technique performed well on simulated and clinical data, and was extended to incorporate curvature of the atrium.
The sensitivity to filtering and post-processing parameters of a technique for tracking rotational sources in optical mapping experiments was investigated, and an optimal parameter set was determined. This was used alongside a technique developed to calculate the distribution of new wavefront initiation sites, in order to gain insight into the mechanisms sustaining atrial fibrillation. The technique for tracking rotational sources was extended to work for unipolar and bipolar electrogram data. This extension was tested in simulation against action potential data, where the modalities (unipolar electrogram, bipolar electrogram and action potential phase) were found to perform similarly; and then applied to experimental and clinical electrograms. Rotational content and activation maps were found to be similar for clinical unipolar and bipolar phase.
The developed techniques were used to investigate the effects of spatial resolution on the identification of rotors and focal sources, as a function of the distance between successive wavefronts. A technique was presented to estimate this distance from a small number of measuring points, providing experimental and clinical utility. These requirements were then tested for simulated high-density mapping catheters and basket catheters, where all catheters provided sufficient resolution.
Version
Open Access
Date Issued
2015-06
Date Awarded
2015-09
Copyright Statement
Attribution NoDerivatives 4.0 International Licence (CC BY-ND)
Advisor
Siggers, Jennifer
Peters, Nicholas
Ng, Fu Siong
Sponsor
British Heart Foundation
Imperial College London
Grant Number
BHF Programme Grant RG/10/11/28457
Academy of medical sciences starter grant AMS-SGCL8-Ng
Publisher Department
Bioengineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)